diff options
| author | Bruce Hill <bruce@bruce-hill.com> | 2025-09-01 16:44:58 -0400 |
|---|---|---|
| committer | Bruce Hill <bruce@bruce-hill.com> | 2025-09-01 16:44:58 -0400 |
| commit | ed50c5fefb8892ad2ba5262491669f268ddbd436 (patch) | |
| tree | e93a6a8b7e963c37a001691751d6845d10e0cbf8 | |
| parent | 02a99d24a310c04622a875dcf4b0c6fd2de71332 (diff) | |
Overhaul code to stop keeping examples and libraries in the same repo,
but instead spin each out into its own repo.
87 files changed, 85 insertions, 6400 deletions
@@ -184,10 +184,12 @@ man/man1/tomo.1: docs/tomo.1.md pandoc --lua-filter=docs/.pandoc/bold-code.lua -s $< -t man -o $@ examples: - ./local-tomo -qIL examples/log examples/ini examples/vectors examples/http examples/wrap examples/colorful - ./local-tomo -e examples/game/game.tm examples/http-server/http-server.tm + ./local-tomo -L modules/examples.ini ./local-tomo examples/learnxiny.tm +core-libs: + ./local-tomo -L modules/core.ini + deps: bash ./install_dependencies.sh @@ -217,14 +219,7 @@ install-files: build/bin/$(EXE_FILE) build/lib/$(LIB_FILE) build/lib/$(AR_FILE) cp man/man3/* "$(PREFIX)/man/man3/"; \ sh link_versions.sh -install-libs: build/bin/$(EXE_FILE) check-utilities - if ! [ -w "$(PREFIX)" ]; then \ - $(SUDO) -u $(OWNER) $(MAKE) install-libs; \ - exit 0; \ - fi; \ - ./local-tomo -qIL lib/patterns lib/json lib/time lib/commands lib/shell lib/random lib/base64 lib/pthreads lib/uuid lib/core - -install: install-files install-libs +install: install-files uninstall: if ! [ -w "$(PREFIX)" ]; then \ @@ -238,4 +233,4 @@ uninstall: endif .SUFFIXES: -.PHONY: all clean install install-files install-libs uninstall test tags examples deps check-utilities check-c-compiler check-libs version +.PHONY: all clean install install-files uninstall test tags core-libs examples deps check-utilities check-c-compiler check-libs version diff --git a/examples/README.md b/examples/README.md index 9e9291e5..d527d692 100644 --- a/examples/README.md +++ b/examples/README.md @@ -4,19 +4,20 @@ This folder contains some example programs and libraries. ## Example Programs +- [hello.tm](hello.tm): A Hello World program. - [learnxiny.tm](learnxiny.tm): A quick overview of language features in the style of [learnxinyminutes.com](https://learnxinyminutes.com/). -- [game](game/): An example game using raylib. -- [http-server](http-server/): A multithreaded HTTP server. -- [wrap](wrap/): A command-line program to wrap text. +- [game](https://github.com/bruce-hill/tomo-game): An example game using raylib. +- [http-server](https://github.com/bruce-hill/http-server): A multithreaded HTTP server. +- [wrap](https://github.com/bruce-hill/wrap): A command-line program to wrap text. ## Example Libraries Libraries can be installed with `tomo -IL ./library-folder` -- [colorful](colorful/): A DSL useful for rendering terminal colors. -- [coroutines](coroutines/): A library for stackful coroutines similar to Lua's. (Note: only works on x86_64) -- [http](http/): An HTTP library to make basic synchronous HTTP requests. -- [ini](ini/): An INI configuration file reader tool. -- [log](log/): A logging utility. -- [vectors](vectors/): A math vector library. +- [colorful](https://github.com/bruce-hill/tomo-colorful): A DSL useful for rendering terminal colors. +- [coroutines](https://github.com/bruce-hill/tomo-coroutines): A library for stackful coroutines similar to Lua's. (Note: only works on x86_64) +- [http](https://github.com/bruce-hill/tomo-http): An HTTP library to make basic synchronous HTTP requests. +- [ini](https://github.com/bruce-hill/tomo-ini): An INI configuration file reader tool. +- [log](https://github.com/bruce-hill/tomo-log): A logging utility. +- [vectors](https://github.com/bruce-hill/tomo-vectors): A math vector library. diff --git a/examples/colorful/CHANGES.md b/examples/colorful/CHANGES.md deleted file mode 100644 index 093cc077..00000000 --- a/examples/colorful/CHANGES.md +++ /dev/null @@ -1,13 +0,0 @@ -# Version History - -## v1.2 - -- Version bump for patterns. - -## v1.1 - -- Added syntax for `@(strikethrough:...)` or `@(s:...)` - -## v1.0 - -Initial version diff --git a/examples/colorful/README.md b/examples/colorful/README.md deleted file mode 100644 index faded9b1..00000000 --- a/examples/colorful/README.md +++ /dev/null @@ -1,70 +0,0 @@ -# Colorful Lang - -Colorful is a `lang` that lets you write colorful text for the terminal without -having to stress about managing state for color highlighting. - -## Grammar - -The grammar looks like this: - -``` -colorful <- ("@(at)" / "@(lparen)" / "@(rparen)" # Escapes - / "@(" attributes ":" colorful ")" # Colorful text - / .)* # Plain text - -attributes <- (attribute ("," attribute)*)? - -attribute <- color # Color defaults to foreground - / "fg=" color # Foreground color - / "bg=" color # Background color - / "ul=" color # Underline color - / "b" / "bold" - / "d" / "dim" - / "u" / "underline" - / "i" / "italic" - / "B" / "blink" - / "r" / "reverse" - # These are rarely supported by terminals: - / "fraktur" - / "frame" - / "encircle" - / "overline" - / "super" / "superscript" - / "sub" / "subscript" - -color <- "black" / "red" / "green" / "yellow" / "blue" / "magenta" / "cyan" / "white" - # All caps colors are "bright" colors (not always supported): - / "BLACK" / "RED" / "GREEN" / "YELLOW" / "BLUE" / "MAGENTA" / "CYAN" / "WHITE" - / "default" - / "#" 6 hex # Values 0x000000-0xFFFFFF - / "#" 3 hex # Values 0x000-0xFFF - / 1-3 digit # Values 0-255 -``` - -## Command Line Usage - -You can run `colorful` as a standalone executable to render colorful text with -ANSI escape sequences so it looks nice on a terminal. - -``` -colorful [--help] [texts...] [--by-line] [--files ...] -``` - -## Library Usage - -`colorful` can also be used as a Tomo library: - -```ini -# modules.ini -[colorful] -version=v1.0 -``` - -```tomo -use colorful - -$Colorful" - @(blue:Welcome to the @(bold:party)!) - We have @(green,bold:colors)! -".print() -``` diff --git a/examples/colorful/colorful.tm b/examples/colorful/colorful.tm deleted file mode 100644 index 5b01cfd5..00000000 --- a/examples/colorful/colorful.tm +++ /dev/null @@ -1,220 +0,0 @@ -# Colorful language - -HELP := " - colorful: A domain-specific language for writing colored text to the terminal - Usage: colorful [args...] [--by-line] [--files files...] -" - -CSI := "\033[" - -use patterns - -lang Colorful - convert(text:Text -> Colorful) - text = text.translate({"@"="@(at)", "("="@(lparen)", ")"="@(rparen)"}) - return Colorful.from_text(text) - - convert(i:Int -> Colorful) return Colorful.from_text("$i") - convert(n:Num -> Colorful) return Colorful.from_text("$n") - - func for_terminal(c:Colorful -> Text) - return CSI ++ "m" ++ _for_terminal(c, _TermState()) - - func print(c:Colorful, newline=yes) - say(c.for_terminal(), newline=newline) - - -func main(texts:[Text], files:[Path]=[], by_line=no) - for i,text in texts - colorful := Colorful.from_text(text) - colorful.print(newline=no) - if i == texts.length then say("") - else say(" ", newline=no) - - if texts.length == 0 and files.length == 0 - files = [(/dev/stdin)] - - for file in files - if by_line - for line in file.by_line() or exit("Could not read file: $file") - colorful := Colorful.from_text(line) - colorful.print() - else - colorful := Colorful.from_text(file.read() or exit("Could not read file: $file")) - colorful.print(newline=no) - - -func _for_terminal(c:Colorful, state:_TermState -> Text) - return c.text.map_pattern(recursive=no, $Pat/@(?)/, func(m:PatternMatch) _add_ansi_sequences(m.captures[1], state)) - -enum _Color(Default, Bright(color:Int16), Color8Bit(color:Int16), Color24Bit(color:Int32)) - func from_text(text:Text -> _Color?) - if text.matches_pattern($Pat/#{3-6 hex}/) - hex := text.from(2) - return none unless hex.length == 3 or hex.length == 6 - if hex.length == 3 - hex = hex[1]++hex[1]++hex[2]++hex[2]++hex[3]++hex[3] - n := Int32.parse("0x" ++ hex) or return none - return Color24Bit(n) - else if text.matches_pattern($Pat/{1-3 digit}/) - n := Int16.parse(text) or return none - if n >= 0 and n <= 255 return Color8Bit(n) - else if text == "black" return _Color.Color8Bit(0) - else if text == "red" return _Color.Color8Bit(1) - else if text == "green" return _Color.Color8Bit(2) - else if text == "yellow" return _Color.Color8Bit(3) - else if text == "blue" return _Color.Color8Bit(4) - else if text == "magenta" return _Color.Color8Bit(5) - else if text == "cyan" return _Color.Color8Bit(6) - else if text == "white" return _Color.Color8Bit(7) - else if text == "default" return _Color.Default - else if text == "BLACK" return _Color.Bright(0) - else if text == "RED" return _Color.Bright(1) - else if text == "GREEN" return _Color.Bright(2) - else if text == "YELLOW" return _Color.Bright(3) - else if text == "BLUE" return _Color.Bright(4) - else if text == "MAGENTA" return _Color.Bright(5) - else if text == "CYAN" return _Color.Bright(6) - else if text == "WHITE" return _Color.Bright(7) - return none - - func fg(c:_Color -> Text) - when c is Color8Bit(color) - if color >= 0 and color <= 7 return "$(30+color)" - else if color >= 0 and color <= 255 return "38;5;$color" - is Color24Bit(hex) - if hex >= 0 and hex <= 0xFFFFFF - return "38;2;$((hex >> 16) and 0xFF);$((hex >> 8) and 0xFF);$((hex >> 0) and 0xFF)" - is Bright(color) - if color <= 7 return "$(90+color)" - is Default - return "39" - fail("Invalid foreground color: '$c'") - - func bg(c:_Color -> Text) - when c is Color8Bit(color) - if color >= 0 and color <= 7 return "$(40+color)" - else if color >= 0 and color <= 255 return "48;5;$color" - is Color24Bit(hex) - if hex >= 0 and hex <= 0xFFFFFF - return "48;2;$((hex >> 16) and 0xFF);$((hex >> 8) and 0xFF);$((hex >> 0) and 0xFF)" - is Bright(color) - if color <= 7 return "$(90+color)" - is Default - return "49" - fail("Invalid background color: '$c'") - - func underline(c:_Color -> Text) - when c is Color8Bit(color) - if color >= 0 and color <= 255 return "58;5;$color" - is Color24Bit(hex) - if hex >= 0 and hex <= 0xFFFFFF - return "58;2;$((hex >> 16) and 0xFF);$((hex >> 8) and 0xFF);$((hex >> 0) and 0xFF)" - is Default - return "59" - is Bright(color) - pass - fail("Invalid underline color: '$c'") - -func _toggle(sequences:&[Text], cur,new:Bool, apply,unapply:Text; inline) - if new and not cur - sequences.insert(apply) - else if cur and not new - sequences.insert(unapply) - -func _toggle2(sequences:&[Text], cur1,cur2,new1,new2:Bool, apply1,apply2,unapply:Text; inline) - return if new1 == cur1 and new2 == cur2 - if (cur1 and not new1) or (cur2 and not new2) # Gotta wipe at least one - sequences.insert(unapply) - cur1, cur2 = no, no # Wiped out - - if new1 and not cur1 - sequences.insert(apply1) - if new2 and not cur2 - sequences.insert(apply2) - -struct _TermState( - bold=no, dim=no, italic=no, underline=no, blink=no, - reverse=no, conceal=no, strikethrough=no, fraktur=no, frame=no, - encircle=no, overline=no, superscript=no, subscript=no, - bg=_Color.Default, fg=_Color.Default, underline_color=_Color.Default, -) - - func apply(old,new:_TermState -> Text) - sequences : &[Text] - _toggle2(sequences, old.bold, old.dim, new.bold, new.dim, "1", "2", "22") - _toggle2(sequences, old.italic, old.fraktur, new.italic, new.fraktur, "3", "20", "23") - _toggle(sequences, old.underline, new.underline, "4", "24") - _toggle(sequences, old.blink, new.blink, "5", "25") - _toggle(sequences, old.reverse, new.reverse, "7", "27") - _toggle(sequences, old.conceal, new.conceal, "8", "28") - _toggle(sequences, old.strikethrough, new.strikethrough, "9", "29") - _toggle2(sequences, old.frame, old.encircle, new.frame, new.frame, "51", "52", "54") - _toggle(sequences, old.overline, new.overline, "53", "55") - _toggle2(sequences, old.subscript, old.subscript, new.superscript, new.superscript, "73", "74", "75") - - if new.bg != old.bg - sequences.insert(new.bg.bg()) - - if new.fg != old.fg - sequences.insert(new.fg.fg()) - - if new.underline_color != old.underline_color - sequences.insert(new.underline_color.underline()) - - if sequences.length == 0 - return "" - return CSI ++ ";".join(sequences) ++ "m" - -func _add_ansi_sequences(text:Text, prev_state:_TermState -> Text) - if text == "lparen" return "(" - else if text == "rparen" return ")" - else if text == "@" or text == "at" return "@" - parts := ( - text.pattern_captures($Pat/{0+..}:{0+..}/) or - return "@("++_for_terminal(Colorful.from_text(text), prev_state)++")" - ) - attributes := parts[1].split_pattern($Pat/{0+space},{0+space}/) - new_state := prev_state - for attr in attributes - if attr.starts_with("fg=") - new_state.fg = _Color.from_text(attr.from(4))! - else if attr.starts_with("bg=") - new_state.bg = _Color.from_text(attr.from(4))! - else if attr.starts_with("ul=") - new_state.underline_color = _Color.from_text(attr.from(4))! - else if color := _Color.from_text(attr) - new_state.fg = color - else if attr == "b" or attr == "bold" - new_state.bold = yes - else if attr == "d" or attr == "dim" - new_state.dim = yes - else if attr == "i" or attr == "italic" - new_state.italic = yes - else if attr == "u" or attr == "underline" - new_state.underline = yes - else if attr == "s" or attr == "strikethrough" - new_state.strikethrough = yes - else if attr == "B" or attr == "blink" - new_state.blink = yes - else if attr == "r" or attr == "reverse" - new_state.reverse = yes - else if attr == "fraktur" - new_state.fraktur = yes - else if attr == "frame" - new_state.frame = yes - else if attr == "encircle" - new_state.encircle = yes - else if attr == "overline" - new_state.overline = yes - else if attr == "super" or attr == "superscript" - new_state.superscript = yes - else if attr == "sub" or attr == "subscript" - new_state.subscript = yes - else - fail("Invalid attribute: '$attr'") - - result := prev_state.apply(new_state) - result ++= parts[2].map_pattern(recursive=no, $Pat/@(?)/, func(m:PatternMatch) _add_ansi_sequences(m.captures[1], new_state)) - result ++= new_state.apply(prev_state) - return result diff --git a/examples/colorful/modules.ini b/examples/colorful/modules.ini deleted file mode 100644 index 5e4b5b0a..00000000 --- a/examples/colorful/modules.ini +++ /dev/null @@ -1,2 +0,0 @@ -[patterns] -version=v1.1 diff --git a/examples/colorful/test.colors b/examples/colorful/test.colors deleted file mode 100644 index 314b8b05..00000000 --- a/examples/colorful/test.colors +++ /dev/null @@ -1,9 +0,0 @@ -This is some text that has @(bold:@(red:c)@(yellow:o)@(green:l)@(cyan:o)@(blue:r)@(magenta:s))! - -@(fg=#aaf,b:You can have @(red:nested) color directives and stuff (even in -parens) will be handled @(i:right)) - -@(dim:The output @(bold:ANSI) sequences will be optimal, even if you have -nested stuff like bold and dim) - -(which is distinct from @(bold:BOLD) by itself) diff --git a/examples/coroutines/ACO_LICENSE b/examples/coroutines/ACO_LICENSE deleted file mode 100644 index ef4f82f0..00000000 --- a/examples/coroutines/ACO_LICENSE +++ /dev/null @@ -1,202 +0,0 @@ - Apache License - Version 2.0, January 2004 - http://www.apache.org/licenses/ - - TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION - - 1. Definitions. - - "License" shall mean the terms and conditions for use, reproduction, - and distribution as defined by Sections 1 through 9 of this document. - - "Licensor" shall mean the copyright owner or entity authorized by - the copyright owner that is granting the License. - - "Legal Entity" shall mean the union of the acting entity and all - other entities that control, are controlled by, or are under common - control with that entity. For the purposes of this definition, - "control" means (i) the power, direct or indirect, to cause the - direction or management of such entity, whether by contract or - otherwise, or (ii) ownership of fifty percent (50%) or more of the - outstanding shares, or (iii) beneficial ownership of such entity. - - "You" (or "Your") shall mean an individual or Legal Entity - exercising permissions granted by this License. - - "Source" form shall mean the preferred form for making modifications, - including but not limited to software source code, documentation - source, and configuration files. - - "Object" form shall mean any form resulting from mechanical - transformation or translation of a Source form, including but - not limited to compiled object code, generated documentation, - and conversions to other media types. - - "Work" shall mean the work of authorship, whether in Source or - Object form, made available under the License, as indicated by a - copyright notice that is included in or attached to the work - (an example is provided in the Appendix below). - - "Derivative Works" shall mean any work, whether in Source or Object - form, that is based on (or derived from) the Work and for which the - editorial revisions, annotations, elaborations, or other modifications - represent, as a whole, an original work of authorship. For the purposes - of this License, Derivative Works shall not include works that remain - separable from, or merely link (or bind by name) to the interfaces of, - the Work and Derivative Works thereof. - - "Contribution" shall mean any work of authorship, including - the original version of the Work and any modifications or additions - to that Work or Derivative Works thereof, that is intentionally - submitted to Licensor for inclusion in the Work by the copyright owner - or by an individual or Legal Entity authorized to submit on behalf of - the copyright owner. For the purposes of this definition, "submitted" - means any form of electronic, verbal, or written communication sent - to the Licensor or its representatives, including but not limited to - communication on electronic mailing lists, source code control systems, - and issue tracking systems that are managed by, or on behalf of, the - Licensor for the purpose of discussing and improving the Work, but - excluding communication that is conspicuously marked or otherwise - designated in writing by the copyright owner as "Not a Contribution." - - "Contributor" shall mean Licensor and any individual or Legal Entity - on behalf of whom a Contribution has been received by Licensor and - subsequently incorporated within the Work. - - 2. Grant of Copyright License. Subject to the terms and conditions of - this License, each Contributor hereby grants to You a perpetual, - worldwide, non-exclusive, no-charge, royalty-free, irrevocable - copyright license to reproduce, prepare Derivative Works of, - publicly display, publicly perform, sublicense, and distribute the - Work and such Derivative Works in Source or Object form. - - 3. Grant of Patent License. Subject to the terms and conditions of - this License, each Contributor hereby grants to You a perpetual, - worldwide, non-exclusive, no-charge, royalty-free, irrevocable - (except as stated in this section) patent license to make, have made, - use, offer to sell, sell, import, and otherwise transfer the Work, - where such license applies only to those patent claims licensable - by such Contributor that are necessarily infringed by their - Contribution(s) alone or by combination of their Contribution(s) - with the Work to which such Contribution(s) was submitted. If You - institute patent litigation against any entity (including a - cross-claim or counterclaim in a lawsuit) alleging that the Work - or a Contribution incorporated within the Work constitutes direct - or contributory patent infringement, then any patent licenses - granted to You under this License for that Work shall terminate - as of the date such litigation is filed. - - 4. Redistribution. You may reproduce and distribute copies of the - Work or Derivative Works thereof in any medium, with or without - modifications, and in Source or Object form, provided that You - meet the following conditions: - - (a) You must give any other recipients of the Work or - Derivative Works a copy of this License; and - - (b) You must cause any modified files to carry prominent notices - stating that You changed the files; and - - (c) You must retain, in the Source form of any Derivative Works - that You distribute, all copyright, patent, trademark, and - attribution notices from the Source form of the Work, - excluding those notices that do not pertain to any part of - the Derivative Works; and - - (d) If the Work includes a "NOTICE" text file as part of its - distribution, then any Derivative Works that You distribute must - include a readable copy of the attribution notices contained - within such NOTICE file, excluding those notices that do not - pertain to any part of the Derivative Works, in at least one - of the following places: within a NOTICE text file distributed - as part of the Derivative Works; within the Source form or - documentation, if provided along with the Derivative Works; or, - within a display generated by the Derivative Works, if and - wherever such third-party notices normally appear. The contents - of the NOTICE file are for informational purposes only and - do not modify the License. You may add Your own attribution - notices within Derivative Works that You distribute, alongside - or as an addendum to the NOTICE text from the Work, provided - that such additional attribution notices cannot be construed - as modifying the License. - - You may add Your own copyright statement to Your modifications and - may provide additional or different license terms and conditions - for use, reproduction, or distribution of Your modifications, or - for any such Derivative Works as a whole, provided Your use, - reproduction, and distribution of the Work otherwise complies with - the conditions stated in this License. - - 5. Submission of Contributions. Unless You explicitly state otherwise, - any Contribution intentionally submitted for inclusion in the Work - by You to the Licensor shall be under the terms and conditions of - this License, without any additional terms or conditions. - Notwithstanding the above, nothing herein shall supersede or modify - the terms of any separate license agreement you may have executed - with Licensor regarding such Contributions. - - 6. Trademarks. This License does not grant permission to use the trade - names, trademarks, service marks, or product names of the Licensor, - except as required for reasonable and customary use in describing the - origin of the Work and reproducing the content of the NOTICE file. - - 7. Disclaimer of Warranty. Unless required by applicable law or - agreed to in writing, Licensor provides the Work (and each - Contributor provides its Contributions) on an "AS IS" BASIS, - WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or - implied, including, without limitation, any warranties or conditions - of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A - PARTICULAR PURPOSE. You are solely responsible for determining the - appropriateness of using or redistributing the Work and assume any - risks associated with Your exercise of permissions under this License. - - 8. Limitation of Liability. In no event and under no legal theory, - whether in tort (including negligence), contract, or otherwise, - unless required by applicable law (such as deliberate and grossly - negligent acts) or agreed to in writing, shall any Contributor be - liable to You for damages, including any direct, indirect, special, - incidental, or consequential damages of any character arising as a - result of this License or out of the use or inability to use the - Work (including but not limited to damages for loss of goodwill, - work stoppage, computer failure or malfunction, or any and all - other commercial damages or losses), even if such Contributor - has been advised of the possibility of such damages. - - 9. Accepting Warranty or Additional Liability. While redistributing - the Work or Derivative Works thereof, You may choose to offer, - and charge a fee for, acceptance of support, warranty, indemnity, - or other liability obligations and/or rights consistent with this - License. However, in accepting such obligations, You may act only - on Your own behalf and on Your sole responsibility, not on behalf - of any other Contributor, and only if You agree to indemnify, - defend, and hold each Contributor harmless for any liability - incurred by, or claims asserted against, such Contributor by reason - of your accepting any such warranty or additional liability. - - END OF TERMS AND CONDITIONS - - APPENDIX: How to apply the Apache License to your work. - - To apply the Apache License to your work, attach the following - boilerplate notice, with the fields enclosed by brackets "[]" - replaced with your own identifying information. (Don't include - the brackets!) The text should be enclosed in the appropriate - comment syntax for the file format. We also recommend that a - file or class name and description of purpose be included on the - same "printed page" as the copyright notice for easier - identification within third-party archives. - - Copyright [2018] [Sen Han <00hnes@gmail.com>] - Copyright [2024] [Bruce Hill <bruce@bruce-hill.com>] - - Licensed under the Apache License, Version 2.0 (the "License"); - you may not use this file except in compliance with the License. - You may obtain a copy of the License at - - http://www.apache.org/licenses/LICENSE-2.0 - - Unless required by applicable law or agreed to in writing, software - distributed under the License is distributed on an "AS IS" BASIS, - WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - See the License for the specific language governing permissions and - limitations under the License. diff --git a/examples/coroutines/CHANGES.md b/examples/coroutines/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/coroutines/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/coroutines/README.md b/examples/coroutines/README.md deleted file mode 100644 index 644c0e07..00000000 --- a/examples/coroutines/README.md +++ /dev/null @@ -1,30 +0,0 @@ -# Tomo Coroutine Library - -This is a coroutine library built on top of a modified version of -[libaco](https://libaco.org). - -## Example Usage - -```ini -# modules.ini -[coroutines] -version=v1.0 -``` - -```tomo -use coroutines - -func main() - co := Coroutine(func() - say("I'm in the coroutine!") - yield() - say("I'm back in the coroutine!") - ) - >> co - say("I'm in the main func") - >> co.resume() - say("I'm back in the main func") - >> co.resume() - say("I'm back in the main func again") - >> co.resume() -``` diff --git a/examples/coroutines/aco.c b/examples/coroutines/aco.c deleted file mode 100644 index 258efe28..00000000 --- a/examples/coroutines/aco.c +++ /dev/null @@ -1,458 +0,0 @@ -// Copyright 2018 Sen Han <00hnes@gmail.com> -// Modifications copyright 2025 Bruce Hill <bruce@bruce-hill.com> -// -// Licensed under the Apache License, Version 2.0 (the "License"); -// you may not use this file except in compliance with the License. -// You may obtain a copy of the License at -// -// http://www.apache.org/licenses/LICENSE-2.0 -// -// Unless required by applicable law or agreed to in writing, software -// distributed under the License is distributed on an "AS IS" BASIS, -// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -// See the License for the specific language governing permissions and -// limitations under the License. - -#define _GNU_SOURCE - -#include "aco.h" -#include <stdint.h> -#include <stdio.h> - -#ifndef public -#define public __attribute__((visibility("default"))) -#endif - -#define aco_size_t_safe_add_assert(a, b) aco_assert((a) + (b) >= (a)) - -static void aco_default_protector_last_word(void *); - -void *(*aco_alloc_fn)(size_t) = malloc; -void (*aco_dealloc_fn)(void *) = free; - -#define aco_alloc(size) \ - ({ \ - void *_ptr = aco_alloc_fn(size); \ - if (aco_unlikely((_ptr) == NULL)) { \ - fprintf(stderr, "Aborting: failed to allocate memory: %s:%d:%s\n", __FILE__, __LINE__, \ - __PRETTY_FUNCTION__); \ - abort(); \ - } \ - _ptr; \ - }) - -// aco's Global Thread Local Storage variable `co` -public -__thread aco_t *aco_gtls_co; -static __thread aco_cofuncp_t aco_gtls_last_word_fp = aco_default_protector_last_word; - -#ifdef __i386__ -static __thread void *aco_gtls_fpucw_mxcsr[2]; -#elif __x86_64__ -static __thread void *aco_gtls_fpucw_mxcsr[1]; -#else -#error "platform not supporteded yet" -#endif - -public -void aco_runtime_test(void) { -#ifdef __i386__ - _Static_assert(sizeof(void *) == 4, "require 'sizeof(void*) == 4'"); -#elif __x86_64__ - _Static_assert(sizeof(void *) == 8, "require 'sizeof(void*) == 8'"); - _Static_assert(sizeof(__uint128_t) == 16, "require 'sizeof(__uint128_t) == 16'"); -#else -#error "platform not supporteded yet" -#endif - _Static_assert(sizeof(int) >= 4, "require 'sizeof(int) >= 4'"); - aco_assert(sizeof(int) >= 4); - _Static_assert(sizeof(int) <= sizeof(size_t), "require 'sizeof(int) <= sizeof(size_t)'"); - aco_assert(sizeof(int) <= sizeof(size_t)); -} - -#ifdef __x86_64__ -static inline void aco_fast_memcpy(void *dst, const void *src, size_t sz) { - if (((uintptr_t)src & 0x0f) != 0 || ((uintptr_t)dst & 0x0f) != 0 || (sz & 0x0f) != 0x08 || (sz >> 4) > 8) { - memcpy(dst, src, sz); - return; - } - - __uint128_t xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7; - switch (sz >> 4) { - case 0: break; - case 1: - xmm0 = *((__uint128_t *)src + 0); - *((__uint128_t *)dst + 0) = xmm0; - break; - case 2: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - break; - case 3: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - xmm2 = *((__uint128_t *)src + 2); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - *((__uint128_t *)dst + 2) = xmm2; - break; - case 4: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - xmm2 = *((__uint128_t *)src + 2); - xmm3 = *((__uint128_t *)src + 3); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - *((__uint128_t *)dst + 2) = xmm2; - *((__uint128_t *)dst + 3) = xmm3; - break; - case 5: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - xmm2 = *((__uint128_t *)src + 2); - xmm3 = *((__uint128_t *)src + 3); - xmm4 = *((__uint128_t *)src + 4); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - *((__uint128_t *)dst + 2) = xmm2; - *((__uint128_t *)dst + 3) = xmm3; - *((__uint128_t *)dst + 4) = xmm4; - break; - case 6: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - xmm2 = *((__uint128_t *)src + 2); - xmm3 = *((__uint128_t *)src + 3); - xmm4 = *((__uint128_t *)src + 4); - xmm5 = *((__uint128_t *)src + 5); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - *((__uint128_t *)dst + 2) = xmm2; - *((__uint128_t *)dst + 3) = xmm3; - *((__uint128_t *)dst + 4) = xmm4; - *((__uint128_t *)dst + 5) = xmm5; - break; - case 7: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - xmm2 = *((__uint128_t *)src + 2); - xmm3 = *((__uint128_t *)src + 3); - xmm4 = *((__uint128_t *)src + 4); - xmm5 = *((__uint128_t *)src + 5); - xmm6 = *((__uint128_t *)src + 6); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - *((__uint128_t *)dst + 2) = xmm2; - *((__uint128_t *)dst + 3) = xmm3; - *((__uint128_t *)dst + 4) = xmm4; - *((__uint128_t *)dst + 5) = xmm5; - *((__uint128_t *)dst + 6) = xmm6; - break; - case 8: - xmm0 = *((__uint128_t *)src + 0); - xmm1 = *((__uint128_t *)src + 1); - xmm2 = *((__uint128_t *)src + 2); - xmm3 = *((__uint128_t *)src + 3); - xmm4 = *((__uint128_t *)src + 4); - xmm5 = *((__uint128_t *)src + 5); - xmm6 = *((__uint128_t *)src + 6); - xmm7 = *((__uint128_t *)src + 7); - *((__uint128_t *)dst + 0) = xmm0; - *((__uint128_t *)dst + 1) = xmm1; - *((__uint128_t *)dst + 2) = xmm2; - *((__uint128_t *)dst + 3) = xmm3; - *((__uint128_t *)dst + 4) = xmm4; - *((__uint128_t *)dst + 5) = xmm5; - *((__uint128_t *)dst + 6) = xmm6; - *((__uint128_t *)dst + 7) = xmm7; - break; - } - *((uint64_t *)((uintptr_t)dst + sz - 8)) = *((uint64_t *)((uintptr_t)src + sz - 8)); -} -#endif - -void aco_default_protector_last_word(void *_) { - aco_t *co = aco_get_co(); - // do some log about the offending `co` - fprintf(stderr, "error: aco_default_protector_last_word triggered\n"); - fprintf(stderr, - "error: co:%p should call `aco_exit()` instead of direct " - "`return` in co_fp:%p to finish its execution\n", - co, (void *)co->fp); - aco_assert(0); -} - -public -void aco_set_allocator(void *(*alloc)(size_t), void (*dealloc)(void *)) { - aco_alloc_fn = alloc; - aco_dealloc_fn = dealloc; -} - -public -void aco_thread_init(aco_cofuncp_t last_word_co_fp) { - aco_save_fpucw_mxcsr(aco_gtls_fpucw_mxcsr); - - if ((void *)last_word_co_fp != NULL) aco_gtls_last_word_fp = last_word_co_fp; -} - -// This function `aco_funcp_protector` should never be -// called. If it's been called, that means the offending -// `co` didn't call aco_exit(co) instead of `return` to -// finish its execution. -public -void aco_funcp_protector(void) { - if ((void *)(aco_gtls_last_word_fp) != NULL) { - aco_gtls_last_word_fp(NULL); - } else { - aco_default_protector_last_word(NULL); - } - aco_assert(0); -} - -public -aco_shared_stack_t *aco_shared_stack_new(size_t sz) { return aco_shared_stack_new2(sz, 1); } - -public -aco_shared_stack_t *aco_shared_stack_new2(size_t sz, bool guard_page_enabled) { - if (sz == 0) { - sz = 1024 * 1024 * 2; - } - if (sz < 4096) { - sz = 4096; - } - aco_assert(sz > 0); - - size_t u_pgsz = 0; - if (guard_page_enabled) { - // although gcc's Built-in Functions to Perform Arithmetic with - // Overflow Checking is better, but it would require gcc >= 5.0 - long pgsz = sysconf(_SC_PAGESIZE); - // pgsz must be > 0 && a power of two - aco_assert(pgsz > 0 && (((pgsz - 1) & pgsz) == 0)); - u_pgsz = (size_t)((unsigned long)pgsz); - // it should be always true in real life - aco_assert(u_pgsz == (unsigned long)pgsz && ((u_pgsz << 1) >> 1) == u_pgsz); - if (sz <= u_pgsz) { - sz = u_pgsz << 1; - } else { - size_t new_sz; - if ((sz & (u_pgsz - 1)) != 0) { - new_sz = (sz & (~(u_pgsz - 1))); - aco_assert(new_sz >= u_pgsz); - aco_size_t_safe_add_assert(new_sz, (u_pgsz << 1)); - new_sz = new_sz + (u_pgsz << 1); - aco_assert(sz / u_pgsz + 2 == new_sz / u_pgsz); - } else { - aco_size_t_safe_add_assert(sz, u_pgsz); - new_sz = sz + u_pgsz; - aco_assert(sz / u_pgsz + 1 == new_sz / u_pgsz); - } - sz = new_sz; - aco_assert((sz / u_pgsz > 1) && ((sz & (u_pgsz - 1)) == 0)); - } - } - - aco_shared_stack_t *p = aco_alloc(sizeof(aco_shared_stack_t)); - memset(p, 0, sizeof(aco_shared_stack_t)); - - if (guard_page_enabled) { - p->real_ptr = mmap(NULL, sz, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); - if (aco_unlikely(p->real_ptr == MAP_FAILED)) { - fprintf(stderr, "Aborting: failed to allocate memory: %s:%d:%s\n", __FILE__, __LINE__, __PRETTY_FUNCTION__); - abort(); - } - p->guard_page_enabled = true; - aco_assert(0 == mprotect(p->real_ptr, u_pgsz, PROT_READ)); - - p->ptr = (void *)(((uintptr_t)p->real_ptr) + u_pgsz); - p->real_sz = sz; - aco_assert(sz >= (u_pgsz << 1)); - p->sz = sz - u_pgsz; - } else { - // p->guard_page_enabled = 0; - p->sz = sz; - p->ptr = aco_alloc(sz); - } - - p->owner = NULL; -#ifdef ACO_USE_VALGRIND - p->valgrind_stk_id = VALGRIND_STACK_REGISTER(p->ptr, (void *)((uintptr_t)p->ptr + p->sz)); -#endif -#if defined(__i386__) || defined(__x86_64__) - uintptr_t u_p = (uintptr_t)(p->sz - (sizeof(void *) << 1) + (uintptr_t)p->ptr); - u_p = (u_p >> 4) << 4; - p->align_highptr = (void *)u_p; - p->align_retptr = (void *)(u_p - sizeof(void *)); - *((void **)(p->align_retptr)) = (void *)(aco_funcp_protector_asm); - aco_assert(p->sz > (16 + (sizeof(void *) << 1) + sizeof(void *))); - p->align_limit = p->sz - 16 - (sizeof(void *) << 1); -#else -#error "platform not supporteded yet" -#endif - return p; -} - -public -void aco_shared_stack_destroy(aco_shared_stack_t *sstk) { - aco_assert(sstk != NULL && sstk->ptr != NULL); -#ifdef ACO_USE_VALGRIND - VALGRIND_STACK_DEREGISTER(sstk->valgrind_stk_id); -#endif - if (sstk->guard_page_enabled) { - aco_assert(0 == munmap(sstk->real_ptr, sstk->real_sz)); - sstk->real_ptr = NULL; - sstk->ptr = NULL; - } else { - if (aco_dealloc_fn != NULL) aco_dealloc_fn(sstk->ptr); - sstk->ptr = NULL; - } - if (aco_dealloc_fn != NULL) aco_dealloc_fn(sstk); -} - -public -aco_t *aco_create(aco_t *main_co, aco_shared_stack_t *shared_stack, size_t saved_stack_sz, aco_cofuncp_t fp, - void *arg) { - aco_t *p = aco_alloc(sizeof(aco_t)); - memset(p, 0, sizeof(aco_t)); - - if (main_co != NULL) { // non-main co - aco_assertptr(shared_stack); - p->shared_stack = shared_stack; -#ifdef __i386__ - // POSIX.1-2008 (IEEE Std 1003.1-2008) - General Information - Data Types - Pointer Types - // http://pubs.opengroup.org/onlinepubs/9699919799.2008edition/functions/V2_chap02.html#tag_15_12_03 - p->reg[ACO_REG_IDX_RETADDR] = (void *)fp; - // push retaddr - p->reg[ACO_REG_IDX_SP] = p->shared_stack->align_retptr; -#ifndef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - p->reg[ACO_REG_IDX_FPU] = aco_gtls_fpucw_mxcsr[0]; - p->reg[ACO_REG_IDX_FPU + 1] = aco_gtls_fpucw_mxcsr[1]; -#endif -#elif __x86_64__ - p->reg[ACO_REG_IDX_RETADDR] = (void *)fp; - p->reg[ACO_REG_IDX_SP] = p->shared_stack->align_retptr; -#ifndef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - p->reg[ACO_REG_IDX_FPU] = aco_gtls_fpucw_mxcsr[0]; -#endif -#else -#error "platform not supporteded yet" -#endif - p->main_co = main_co; - p->arg = arg; - p->fp = fp; - if (saved_stack_sz == 0) { - saved_stack_sz = 64; - } - p->saved_stack.ptr = aco_alloc(saved_stack_sz); - p->saved_stack.sz = saved_stack_sz; -#if defined(__i386__) || defined(__x86_64__) - p->saved_stack.valid_sz = 0; -#else -#error "platform not supporteded yet" -#endif - return p; - } else { // main co - p->main_co = NULL; - p->arg = arg; - p->fp = fp; - p->shared_stack = NULL; - p->saved_stack.ptr = NULL; - return p; - } - aco_assert(0); -} - -public -aco_attr_no_asan void aco_resume(aco_t *resume_co) { - aco_assert(resume_co != NULL && resume_co->main_co != NULL && !resume_co->is_finished); - if (resume_co->shared_stack->owner != resume_co) { - if (resume_co->shared_stack->owner != NULL) { - aco_t *owner_co = resume_co->shared_stack->owner; - aco_assert(owner_co->shared_stack == resume_co->shared_stack); -#if defined(__i386__) || defined(__x86_64__) - aco_assert(((uintptr_t)(owner_co->shared_stack->align_retptr) >= (uintptr_t)(owner_co->reg[ACO_REG_IDX_SP])) - && ((uintptr_t)(owner_co->shared_stack->align_highptr) - - (uintptr_t)(owner_co->shared_stack->align_limit) - <= (uintptr_t)(owner_co->reg[ACO_REG_IDX_SP]))); - owner_co->saved_stack.valid_sz = - (uintptr_t)(owner_co->shared_stack->align_retptr) - (uintptr_t)(owner_co->reg[ACO_REG_IDX_SP]); - if (owner_co->saved_stack.sz < owner_co->saved_stack.valid_sz) { - if (aco_dealloc_fn != NULL) aco_dealloc_fn(owner_co->saved_stack.ptr); - owner_co->saved_stack.ptr = NULL; - while (1) { - owner_co->saved_stack.sz = owner_co->saved_stack.sz << 1; - aco_assert(owner_co->saved_stack.sz > 0); - if (owner_co->saved_stack.sz >= owner_co->saved_stack.valid_sz) { - break; - } - } - owner_co->saved_stack.ptr = aco_alloc(owner_co->saved_stack.sz); - } - // TODO: optimize the performance penalty of memcpy function call - // for very short memory span - if (owner_co->saved_stack.valid_sz > 0) { -#ifdef __x86_64__ - aco_fast_memcpy(owner_co->saved_stack.ptr, owner_co->reg[ACO_REG_IDX_SP], - owner_co->saved_stack.valid_sz); -#else - memcpy(owner_co->saved_stack.ptr, owner_co->reg[ACO_REG_IDX_SP], owner_co->saved_stack.valid_sz); -#endif - owner_co->saved_stack.ct_save++; - } - if (owner_co->saved_stack.valid_sz > owner_co->saved_stack.max_cpsz) { - owner_co->saved_stack.max_cpsz = owner_co->saved_stack.valid_sz; - } - owner_co->shared_stack->owner = NULL; - owner_co->shared_stack->align_validsz = 0; -#else -#error "platform not supporteded yet" -#endif - } - aco_assert(resume_co->shared_stack->owner == NULL); -#if defined(__i386__) || defined(__x86_64__) - aco_assert(resume_co->saved_stack.valid_sz <= resume_co->shared_stack->align_limit - sizeof(void *)); - // TODO: optimize the performance penalty of memcpy function call - // for very short memory span - if (resume_co->saved_stack.valid_sz > 0) { - void *dst = (void *)((uintptr_t)(resume_co->shared_stack->align_retptr) - resume_co->saved_stack.valid_sz); -#ifdef __x86_64__ - aco_fast_memcpy(dst, resume_co->saved_stack.ptr, resume_co->saved_stack.valid_sz); -#else - memcpy(dst, resume_co->saved_stack.ptr, resume_co->saved_stack.valid_sz); -#endif - resume_co->saved_stack.ct_restore++; - } - if (resume_co->saved_stack.valid_sz > resume_co->saved_stack.max_cpsz) { - resume_co->saved_stack.max_cpsz = resume_co->saved_stack.valid_sz; - } - resume_co->shared_stack->align_validsz = resume_co->saved_stack.valid_sz + sizeof(void *); - resume_co->shared_stack->owner = resume_co; -#else -#error "platform not supporteded yet" -#endif - } - aco_gtls_co = resume_co; - aco_yield_asm(resume_co->main_co, resume_co); - aco_gtls_co = resume_co->main_co; -} - -public -void aco_destroy(aco_t *co) { - aco_assertptr(co); - if (aco_is_main_co(co)) { - if (aco_dealloc_fn != NULL) aco_dealloc_fn(co); - } else { - if (co->shared_stack->owner == co) { - co->shared_stack->owner = NULL; - co->shared_stack->align_validsz = 0; - } - if (aco_dealloc_fn != NULL) aco_dealloc_fn(co->saved_stack.ptr); - co->saved_stack.ptr = NULL; - if (aco_dealloc_fn != NULL) aco_dealloc_fn(co); - } -} - -public -void aco_exit_fn(void *_) { aco_exit(); } diff --git a/examples/coroutines/aco.h b/examples/coroutines/aco.h deleted file mode 100644 index 05ba0bdb..00000000 --- a/examples/coroutines/aco.h +++ /dev/null @@ -1,213 +0,0 @@ -// A coroutine library -// Copyright 2018 Sen Han <00hnes@gmail.com> -// Modifications copyright 2025 Bruce Hill <bruce@bruce-hill.com> -// -// Licensed under the Apache License, Version 2.0 (the "License"); -// you may not use this file except in compliance with the License. -// You may obtain a copy of the License at -// -// http://www.apache.org/licenses/LICENSE-2.0 -// -// Unless required by applicable law or agreed to in writing, software -// distributed under the License is distributed on an "AS IS" BASIS, -// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -// See the License for the specific language governing permissions and -// limitations under the License. - -#pragma once - -#include <limits.h> -#include <stdbool.h> -#include <stdint.h> -#include <stdio.h> -#include <stdlib.h> -#include <string.h> -#include <sys/mman.h> -#include <time.h> -#include <unistd.h> - -#ifdef ACO_USE_VALGRIND -#include <valgrind/valgrind.h> -#endif - -#ifdef __cplusplus -extern "C" { -#endif - -#define ACO_VERSION_MAJOR 2 -#define ACO_VERSION_MINOR 0 -#define ACO_VERSION_PATCH 0 - -#ifdef __i386__ -#define ACO_REG_IDX_RETADDR 0 -#define ACO_REG_IDX_SP 1 -#define ACO_REG_IDX_BP 2 -#define ACO_REG_IDX_ARG1 0 -#define ACO_REG_IDX_FPU 6 -#elif __x86_64__ -#define ACO_REG_IDX_RETADDR 4 -#define ACO_REG_IDX_SP 5 -#define ACO_REG_IDX_BP 7 -#define ACO_REG_IDX_EDI 8 -#define ACO_REG_IDX_FPU 8 -#else -#error "platform not supported yet" -#endif - -typedef struct { - void *ptr; - size_t sz; - size_t valid_sz; - // max copy size in bytes - size_t max_cpsz; - // copy from shared stack to this saved stack - size_t ct_save; - // copy from this saved stack to shared stack - size_t ct_restore; -} aco_saved_stack_t; - -struct aco_s; -typedef struct aco_s aco_t; - -typedef struct { - void *ptr; - size_t sz; - void *align_highptr; - void *align_retptr; - size_t align_validsz; - size_t align_limit; - aco_t *owner; - - bool guard_page_enabled; - void *real_ptr; - size_t real_sz; - -#ifdef ACO_USE_VALGRIND - unsigned long valgrind_stk_id; -#endif -} aco_shared_stack_t; - -typedef void (*aco_cofuncp_t)(void *); - -struct aco_s { - // cpu registers' state -#ifdef __i386__ -#ifdef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - void *reg[6]; -#else - void *reg[8]; -#endif -#elif __x86_64__ -#ifdef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - void *reg[8]; -#else - void *reg[9]; -#endif -#else -#error "platform not supported yet" -#endif - aco_t *main_co; - void *arg; - bool is_finished; - - aco_cofuncp_t fp; - - aco_saved_stack_t saved_stack; - aco_shared_stack_t *shared_stack; -}; - -#define aco_likely(x) (__builtin_expect(!!(x), 1)) - -#define aco_unlikely(x) (__builtin_expect(!!(x), 0)) - -#define aco_assert(EX) ((aco_likely(EX)) ? ((void)0) : (abort())) - -#define aco_assertptr(ptr) ((aco_likely((ptr) != NULL)) ? ((void)0) : (abort())) - -#if defined(aco_attr_no_asan) -#error "aco_attr_no_asan already defined" -#endif -#if defined(ACO_USE_ASAN) -#if defined(__has_feature) -#if __has_feature(__address_sanitizer__) -#define aco_attr_no_asan __attribute__((__no_sanitize_address__)) -#endif -#endif -#if defined(__SANITIZE_ADDRESS__) && !defined(aco_attr_no_asan) -#define aco_attr_no_asan __attribute__((__no_sanitize_address__)) -#endif -#endif -#ifndef aco_attr_no_asan -#define aco_attr_no_asan -#endif - -void aco_runtime_test(void); - -void aco_set_allocator(void *(*alloc)(size_t), void (*dealloc)(void *)); - -void aco_thread_init(aco_cofuncp_t last_word_co_fp); - -void aco_yield_asm(aco_t *from_co, aco_t *to_co) __asm__("aco_yield_asm"); // asm - -void aco_save_fpucw_mxcsr(void *p) __asm__("aco_save_fpucw_mxcsr"); // asm - -void aco_funcp_protector_asm(void) __asm__("aco_funcp_protector_asm"); // asm - -void aco_funcp_protector(void); - -aco_shared_stack_t *aco_shared_stack_new(size_t sz); - -aco_shared_stack_t *aco_shared_stack_new2(size_t sz, bool guard_page_enabled); - -void aco_shared_stack_destroy(aco_shared_stack_t *sstk); - -aco_t *aco_create(aco_t *main_co, aco_shared_stack_t *shared_stack, size_t saved_stack_sz, aco_cofuncp_t fp, void *arg); - -// aco's Global Thread Local Storage variable `co` -#ifdef __TINYC__ -#error "TinyCC doesn't support thread-local storage!" -#else -extern __thread aco_t *aco_gtls_co; -#endif - -aco_attr_no_asan void aco_resume(aco_t *resume_co); - -// void aco_yield1(aco_t* yield_co); -#define aco_yield1(yield_co) \ - do { \ - aco_assertptr((yield_co)); \ - aco_assertptr((yield_co)->main_co); \ - aco_yield_asm((yield_co), (yield_co)->main_co); \ - } while (0) - -#define aco_yield() aco_yield1(aco_gtls_co) - -#define aco_get_arg() (aco_gtls_co->arg) - -#define aco_get_co() \ - ({ \ - (void)0; \ - aco_gtls_co; \ - }) - -void aco_destroy(aco_t *co); - -#define aco_is_main_co(co) ({ ((co)->main_co) == NULL; }) - -#define aco_exit1(co) \ - do { \ - (co)->is_finished = true; \ - aco_assert((co)->shared_stack->owner == (co)); \ - (co)->shared_stack->owner = NULL; \ - (co)->shared_stack->align_validsz = 0; \ - aco_yield1((co)); \ - aco_assert(0); \ - } while (0) - -#define aco_exit() aco_exit1(aco_gtls_co) - -void aco_exit_fn(void *); - -#ifdef __cplusplus -} -#endif diff --git a/examples/coroutines/acoyield.S b/examples/coroutines/acoyield.S deleted file mode 100644 index 7bc87ff1..00000000 --- a/examples/coroutines/acoyield.S +++ /dev/null @@ -1,208 +0,0 @@ -.text -.globl aco_yield_asm -#if defined(__APPLE__) -#else -.type aco_yield_asm, @function -#endif -.intel_syntax noprefix -aco_yield_asm: -/* - extern void aco_yield_asm(aco_t* from_co, aco_t* to_co); - - struct aco_t { - void* reg[X]; - // ... - } - - reference: - https://github.com/hjl-tools/x86-psABI/wiki/X86-psABI - - pitfall: - http://man7.org/linux/man-pages/man7/signal.7.html - http://man7.org/linux/man-pages/man2/sigaltstack.2.html - - > $ man 7 signal - > ... - > By default, the signal handler is invoked on the normal process - > stack. It is possible to arrange that the signal handler - > uses an alternate stack; see sigaltstack(2) for a discussion of - > how to do this and when it might be useful. - > ... - - This is a BUG example: - https://github.com/Tencent/libco/blob/v1.0/coctx_swap.S#L27 - - proof of correctness: - https://github.com/hnes/libaco - - mxcsr & fpu: - fnstcw * m2byte - Store FPU control word to m2byte without checking for - pending unmasked floating-point exceptions. - - fldcw m2byte - Load FPU control word from m2byte. - - stmxcsr m32 - Store contents of MXCSR register to m32 - - ldmxcsr m32 - Load MXCSR register from m32. -*/ -/* - 0x00 --> 0xff - eip esp ebp edi esi ebx fpucw16 mxcsr32 - 0 4 8 c 10 14 18 1c -*/ -#ifdef __i386__ - mov eax,DWORD PTR [esp+0x4] // from_co - mov edx,DWORD PTR [esp] // retaddr - lea ecx,[esp+0x4] // esp - mov DWORD PTR [eax+0x8],ebp //<ebp - mov DWORD PTR [eax+0x4],ecx //<esp - mov DWORD PTR [eax+0x0],edx //<retaddr - mov DWORD PTR [eax+0xc],edi //<edi - mov ecx,DWORD PTR [esp+0x8] // to_co - mov DWORD PTR [eax+0x10],esi //<esi - mov DWORD PTR [eax+0x14],ebx //<ebx -#ifndef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - fnstcw WORD PTR [eax+0x18] //<fpucw - stmxcsr DWORD PTR [eax+0x1c] //<mxcsr -#endif - mov edx,DWORD PTR [ecx+0x4] //>esp - mov ebp,DWORD PTR [ecx+0x8] //>ebp - mov eax,DWORD PTR [ecx+0x0] //>retaddr - mov edi,DWORD PTR [ecx+0xc] //>edi - mov esi,DWORD PTR [ecx+0x10] //>esi - mov ebx,DWORD PTR [ecx+0x14] //>ebx -#ifndef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - fldcw WORD PTR [ecx+0x18] //>fpucw - ldmxcsr DWORD PTR [ecx+0x1c] //>mxcsr -#endif - xor ecx,ecx - mov esp,edx - mov edx,eax - - // Pass the user-provided argument as first argument: -#ifdef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - mov eax,DWORD PTR [ecx+0x24] -#else - mov eax,DWORD PTR [ecx+0x28] -#endif - - jmp edx -#elif __x86_64__ -/* - 0x00 --> 0xff - r12 r13 r14 r15 rip rsp rbx rbp fpucw16 mxcsr32 - 0 8 10 18 20 28 30 38 40 44 -*/ - // rdi - from_co | rsi - to_co - mov rdx,QWORD PTR [rsp] // retaddr - lea rcx,[rsp+0x8] // rsp - mov QWORD PTR [rdi+0x0], r12 - mov QWORD PTR [rdi+0x8], r13 - mov QWORD PTR [rdi+0x10],r14 - mov QWORD PTR [rdi+0x18],r15 - mov QWORD PTR [rdi+0x20],rdx // retaddr - mov QWORD PTR [rdi+0x28],rcx // rsp - mov QWORD PTR [rdi+0x30],rbx - mov QWORD PTR [rdi+0x38],rbp -#ifndef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - fnstcw WORD PTR [rdi+0x40] - stmxcsr DWORD PTR [rdi+0x44] -#endif - mov r12,QWORD PTR [rsi+0x0] - mov r13,QWORD PTR [rsi+0x8] - mov r14,QWORD PTR [rsi+0x10] - mov r15,QWORD PTR [rsi+0x18] - mov rax,QWORD PTR [rsi+0x20] // retaddr - mov rcx,QWORD PTR [rsi+0x28] // rsp - mov rbx,QWORD PTR [rsi+0x30] - mov rbp,QWORD PTR [rsi+0x38] -#ifndef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - fldcw WORD PTR [rsi+0x40] - ldmxcsr DWORD PTR [rsi+0x44] -#endif - - // Pass the user-provided argument as first argument: -#ifdef ACO_CONFIG_SHARE_FPU_MXCSR_ENV - mov rdi,QWORD PTR [rsi+0x48] -#else - mov rdi,QWORD PTR [rsi+0x50] -#endif - - mov rsp,rcx - jmp rax -#else - #error "platform not supported" -#endif - -.globl aco_save_fpucw_mxcsr -#if defined(__APPLE__) -#else -.type aco_save_fpucw_mxcsr, @function -#endif -.intel_syntax noprefix -aco_save_fpucw_mxcsr: -#ifdef __i386__ - mov eax,DWORD PTR [esp+0x4] // ptr - fnstcw WORD PTR [eax] - stmxcsr DWORD PTR [eax+0x4] - ret -#elif __x86_64__ - fnstcw WORD PTR [rdi] - stmxcsr DWORD PTR [rdi+0x4] - ret -#else - #error "platform not supported" -#endif - -#if defined(__APPLE__) -.globl _abort -.globl _aco_funcp_protector -#else -.globl abort -.globl aco_funcp_protector -#endif - -.globl aco_funcp_protector_asm -#if defined(__APPLE__) -#else -.type aco_funcp_protector_asm, @function -#endif -.intel_syntax noprefix -aco_funcp_protector_asm: -#ifdef __i386__ - and esp,0xfffffff0 - #if defined(__APPLE__) - call _aco_funcp_protector - call _abort - #else - #if defined(__pic__) || defined(__PIC__) - call aco_funcp_protector@PLT - call abort@PLT - #else - call aco_funcp_protector - call abort - #endif - #endif - ret -#elif __x86_64__ - and rsp,0xfffffffffffffff0 - #if defined(__APPLE__) - call _aco_funcp_protector - call _abort - #else - #if defined(__pic__) || defined(__PIC__) - call aco_funcp_protector@PLT - call abort@PLT - #else - call aco_funcp_protector - call abort - #endif - #endif - ret -#else - #error "platform not supported" -#endif diff --git a/examples/coroutines/coroutines.tm b/examples/coroutines/coroutines.tm deleted file mode 100644 index b530a685..00000000 --- a/examples/coroutines/coroutines.tm +++ /dev/null @@ -1,67 +0,0 @@ -# This is a coroutine library that uses libaco (https://libaco.org) -# -# Lua programmers will recognize this as similar to Lua's stackful coroutines. -# -# Async/Await programmers will weep at its beauty and gnash their teeth and -# rend their garments in despair at what they could have had. - -use ./aco.h -use ./aco.c -use ./acoyield.S - -func main() - say("Example usage") - co := Coroutine(func() - say("I'm in the coroutine!") - yield() - say("I'm back in the coroutine!") - ) - >> co - say("I'm in the main func") - >> co.resume() - say("I'm back in the main func") - >> co.resume() - say("I'm back in the main func again") - >> co.resume() - -struct aco_t(; extern, opaque) -struct aco_shared_stack_t(; extern, opaque) - -_main_co : @aco_t? = none -_shared_stack : @aco_shared_stack_t? = none - -struct Coroutine(co:@aco_t) - convert(fn:func() -> Coroutine) - if not _main_co - _init() - - main_co := _main_co - shared_stack := _shared_stack - aco_ptr := C_code:@aco_t( - aco_create(@main_co, @shared_stack, 0, (void*)@fn.fn, @fn.userdata) - ) - return Coroutine(aco_ptr) - - func is_finished(co:Coroutine->Bool; inline) - return C_code:Bool(((aco_t*)@co.co)->is_finished) - - func resume(co:Coroutine->Bool) - if co.is_finished() - return no - C_code { aco_resume(@co.co); } - return yes - -func _init() - C_code { - aco_set_allocator(GC_malloc, NULL); - aco_thread_init(aco_exit_fn); - } - _main_co = C_code:@aco_t(aco_create(NULL, NULL, 0, NULL, NULL)) - - _shared_stack = C_code:@aco_shared_stack_t(aco_shared_stack_new(0)) - -func yield(; inline) - C_code { - aco_yield(); - } - diff --git a/examples/game/CHANGES.md b/examples/game/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/game/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/game/Makefile b/examples/game/Makefile deleted file mode 100644 index 7cf46ce6..00000000 --- a/examples/game/Makefile +++ /dev/null @@ -1,17 +0,0 @@ - - -game: game.tm box.tm color.tm player.tm world.tm - tomo -e game.tm - -# Disable built-in makefile rules: -%: %.c -%.o: %.c -%: %.o - -clean: - rm -vf game *.tm.* - -play: game - ./game - -.PHONY: play, clean diff --git a/examples/game/README.md b/examples/game/README.md deleted file mode 100644 index 475a8299..00000000 --- a/examples/game/README.md +++ /dev/null @@ -1,13 +0,0 @@ -# Example Game - -This is a simple example game that uses [raylib](https://www.raylib.com/) to -demonstrate a project that spans multiple files and showcases some game -programming concepts used in idiomatic Tomo code. It also showcases how to -interact with an external C library. To run the game: - -```bash -tomo game.tm -``` - -An example [Makefile](Makefile) is also provided if you want to use `make` to -build the game and `make clean` to clean up the built files. diff --git a/examples/game/box.tm b/examples/game/box.tm deleted file mode 100644 index 41ae10e5..00000000 --- a/examples/game/box.tm +++ /dev/null @@ -1,7 +0,0 @@ -# Defines a struct representing boxes on the terrain -use ./world.tm -use ./raylib.tm - -struct Box(pos:Vector2, size=Vector2(50, 50), color=Color(0x80,0x80,0x80)) - func draw(b:Box) - DrawRectangleV(b.pos, b.size, b.color) diff --git a/examples/game/game.tm b/examples/game/game.tm deleted file mode 100644 index f82e4f40..00000000 --- a/examples/game/game.tm +++ /dev/null @@ -1,29 +0,0 @@ -# This game demo uses Raylib to present a simple maze-type game -use ./raylib.tm -use ./world.tm - -func main(map=(./map.txt)) - InitWindow(1600, 900, CString("raylib [core] example - 2d camera")) - - map_contents := map.read() or exit("Could not find the game map: $map") - - world := @World( - player=@Player(Vector2(0,0), Vector2(0,0)), - goal=@Box(Vector2(0,0), Vector2(50,50), color=Color(0x10,0xa0,0x10)), - boxes=@[], - ) - world.load_map(map_contents) - - SetTargetFPS(60) - - while not WindowShouldClose() - dt := GetFrameTime() - world.update(dt) - - BeginDrawing() - ClearBackground(Color(0xCC, 0xCC, 0xCC, 0xFF)) - world.draw() - EndDrawing() - - CloseWindow() - diff --git a/examples/game/map.txt b/examples/game/map.txt deleted file mode 100644 index 46fccd09..00000000 --- a/examples/game/map.txt +++ /dev/null @@ -1,18 +0,0 @@ -################################ -#@ # # # -# #### #### # # #### ##### -# # # # # # -# ####### # ####### # # ### -# # # # # # # -# #### ########## # #### # -# # # # # # # -#### # # # # # # # # ## -# # # # # # # # -####### ########## # ######## -# # # # # -# ########## #### # # # # -# # # # # # # # # -# ####### # # ########## # -# # # # # # # ? # -# # # # # # # -################################ diff --git a/examples/game/player.tm b/examples/game/player.tm deleted file mode 100644 index 2e5e54f6..00000000 --- a/examples/game/player.tm +++ /dev/null @@ -1,28 +0,0 @@ -# Defines a struct representing the player, which is controlled by WASD keys -use ./world.tm -use ./raylib.tm - -struct Player(pos,prev_pos:Vector2) - WALK_SPEED := Num32(500.) - ACCEL := Num32(0.3) - FRICTION := Num32(0.99) - SIZE := Vector2(30, 30) - COLOR := Color(0x60, 0x60, 0xbF) - - func update(p:@Player) - target_x := C_code:Num32( - (Num32_t)((IsKeyDown(KEY_A) ? -1 : 0) + (IsKeyDown(KEY_D) ? 1 : 0)) - ) - target_y := C_code:Num32( - (Num32_t)((IsKeyDown(KEY_W) ? -1 : 0) + (IsKeyDown(KEY_S) ? 1 : 0)) - ) - target_vel := Vector2(target_x, target_y).norm() * Player.WALK_SPEED - - vel := (p.pos - p.prev_pos)/World.DT - vel *= Player.FRICTION - vel = vel.mix(target_vel, Player.ACCEL) - - p.prev_pos, p.pos = p.pos, p.pos + World.DT*vel - - func draw(p:Player) - DrawRectangleV(p.pos, Player.SIZE, Player.COLOR) diff --git a/examples/game/raylib.tm b/examples/game/raylib.tm deleted file mode 100644 index b2ba53d7..00000000 --- a/examples/game/raylib.tm +++ /dev/null @@ -1,63 +0,0 @@ -# Raylib wrapper for some functions and structs -use -lraylib -use <raylib.h> -use <raymath.h> - -struct Color(r,g,b:Byte,a=Byte(255); extern) -struct Rectangle(x,y,width,height:Num32; extern) - func draw(r:Rectangle, color:Color) - DrawRectangleRec(r, color) - -struct Vector2(x,y:Num32; extern) - ZERO := Vector2(0, 0) - func plus(a,b:Vector2->Vector2; inline) - return Vector2(a.x+b.x, a.y+b.y) - func minus(a,b:Vector2->Vector2; inline) - return Vector2(a.x-b.x, a.y-b.y) - func times(a,b:Vector2->Vector2; inline) - return Vector2(a.x*b.x, a.y*b.y) - func negative(v:Vector2->Vector2; inline) - return Vector2(-v.x, -v.y) - func dot(a,b:Vector2->Num32; inline) - return ((a.x-b.x)*(a.x-b.x) + (a.y-b.y)*(a.y-b.y)) - func cross(a,b:Vector2->Num32; inline) - return a.x*b.y - a.y*b.x - func scaled_by(v:Vector2, k:Num32->Vector2; inline) - return Vector2(v.x*k, v.y*k) - func divided_by(v:Vector2, divisor:Num32->Vector2; inline) - return Vector2(v.x/divisor, v.y/divisor) - func length(v:Vector2->Num32; inline) - return (v.x*v.x + v.y*v.y).sqrt() - func dist(a,b:Vector2->Num32; inline) - return a.minus(b).length() - func angle(v:Vector2->Num32; inline) - return Num32.atan2(v.y, v.x) - func norm(v:Vector2->Vector2; inline) - if v.x == 0 and v.y == 0 - return v - len := v.length() - return Vector2(v.x/len, v.y/len) - func rotated(v:Vector2, radians:Num32 -> Vector2) - cos := radians.cos() or return v - sin := radians.sin() or return v - return Vector2(cos*v.x - sin*v.y, sin*v.x + cos*v.y) - func mix(a,b:Vector2, amount:Num32 -> Vector2) - return Vector2( - amount.mix(a.x, b.x), - amount.mix(a.y, b.y), - ) - -extern InitWindow : func(width:Int32, height:Int32, title:CString) -extern SetTargetFPS : func(fps:Int32) -extern WindowShouldClose : func(->Bool) -extern GetFrameTime : func(->Num32) -extern BeginDrawing : func() -extern EndDrawing : func() -extern CloseWindow : func() -extern ClearBackground : func(color:Color) -extern DrawRectangle : func(x,y,width,height:Int32, color:Color) -extern DrawRectangleRec : func(rec:Rectangle, color:Color) -extern DrawRectangleV : func(pos:Vector2, size:Vector2, color:Color) -extern DrawText : func(text:CString, x,y:Int32, text_height:Int32, color:Color) -extern GetScreenWidth : func(->Int32) -extern GetScreenHeight : func(->Int32) diff --git a/examples/game/world.tm b/examples/game/world.tm deleted file mode 100644 index 0de8ea4b..00000000 --- a/examples/game/world.tm +++ /dev/null @@ -1,88 +0,0 @@ -# This file defines a World object for keeping track of everything, as well -# as the collision logic. -use ./player.tm -use ./raylib.tm -use ./box.tm - -# Return a displacement relative to `a` that will push it out of `b` -func solve_overlap(a_pos:Vector2, a_size:Vector2, b_pos:Vector2, b_size:Vector2 -> Vector2) - a_left := a_pos.x - a_right := a_pos.x + a_size.x - a_top := a_pos.y - a_bottom := a_pos.y + a_size.y - - b_left := b_pos.x - b_right := b_pos.x + b_size.x - b_top := b_pos.y - b_bottom := b_pos.y + b_size.y - - # Calculate the overlap in each dimension - overlap_x := (a_right _min_ b_right) - (a_left _max_ b_left) - overlap_y := (a_bottom _min_ b_bottom) - (a_top _max_ b_top) - - # If either axis is not overlapping, then there is no collision: - if overlap_x <= 0 or overlap_y <= 0 - return Vector2(0, 0) - - if overlap_x < overlap_y - if a_right > b_left and a_right < b_right - return Vector2(-(overlap_x), 0) - else if a_left < b_right and a_left > b_left - return Vector2(overlap_x, 0) - else - if a_top < b_bottom and a_top > b_top - return Vector2(0, overlap_y) - else if a_bottom > b_top and a_bottom < b_bottom - return Vector2(0, -overlap_y) - - return Vector2(0, 0) - -struct World(player:@Player, goal:@Box, boxes:@[@Box], dt_accum=Num32(0.0), won=no) - DT := (Num32(1.)/Num32(60.)) - STIFFNESS := Num32(0.3) - - func update(w:@World, dt:Num32) - w.dt_accum += dt - while w.dt_accum > 0 - w.update_once() - w.dt_accum -= World.DT - - func update_once(w:@World) - w.player.update() - - if solve_overlap(w.player.pos, Player.SIZE, w.goal.pos, w.goal.size) != Vector2(0,0) - w.won = yes - - # Resolve player overlapping with any boxes: - for i in 3 - for b in w.boxes - w.player.pos += World.STIFFNESS * solve_overlap(w.player.pos, Player.SIZE, b.pos, b.size) - - func draw(w:@World) - for b in w.boxes - b.draw() - w.goal.draw() - w.player.draw() - - if w.won - DrawText(CString("WINNER"), GetScreenWidth()/Int32(2)-Int32(48*3), GetScreenHeight()/Int32(2)-Int32(24), 48, Color(0,0,0)) - - func load_map(w:@World, map:Text) - if map.has("[]") - map = map.translate({"[]"="#", "@ "="@", " "=" "}) - w.boxes = @[] - box_size := Vector2(50., 50.) - for y,line in map.lines() - for x,cell in line.split() - if cell == "#" - pos := Vector2((Num32(x)-1) * box_size.x, (Num32(y)-1) * box_size.y) - box := @Box(pos, size=box_size, color=Color(0x80,0x80,0x80)) - w.boxes.insert(box) - else if cell == "@" - pos := Vector2((Num32(x)-1) * box_size.x, (Num32(y)-1) * box_size.y) - pos += box_size/Num32(2) - Player.SIZE/Num32(2) - w.player = @Player(pos,pos) - else if cell == "?" - pos := Vector2((Num32(x)-1) * box_size.x, (Num32(y)-1) * box_size.y) - w.goal.pos = pos - diff --git a/examples/hello.tm b/examples/hello.tm new file mode 100644 index 00000000..09ee3ab4 --- /dev/null +++ b/examples/hello.tm @@ -0,0 +1,3 @@ +# A simple hello world program: +func main() + say("Hello world!") diff --git a/examples/http-server/CHANGES.md b/examples/http-server/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/http-server/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/http-server/README.md b/examples/http-server/README.md deleted file mode 100644 index 78c8d793..00000000 --- a/examples/http-server/README.md +++ /dev/null @@ -1,8 +0,0 @@ -# HTTP Server - -This is a simple multithreaded Tomo HTTP server that can be run like this: - -``` -tomo -e http-server.tm -./http-server ./sample-site -``` diff --git a/examples/http-server/connection-queue.tm b/examples/http-server/connection-queue.tm deleted file mode 100644 index c56069e1..00000000 --- a/examples/http-server/connection-queue.tm +++ /dev/null @@ -1,25 +0,0 @@ -use pthreads - -func _assert_success(name:Text, val:Int32; inline) - fail("$name() failed!") if val < 0 - -struct ConnectionQueue(_connections:@[Int32]=@[], _mutex=pthread_mutex_t.new(), _cond=pthread_cond_t.new()) - func enqueue(queue:ConnectionQueue, connection:Int32) - queue._mutex.lock() - queue._connections.insert(connection) - queue._mutex.unlock() - queue._cond.signal() - - - func dequeue(queue:ConnectionQueue -> Int32) - conn : Int32? - - queue._mutex.lock() - - while queue._connections.length == 0 - queue._cond.wait(queue._mutex) - - conn = queue._connections.pop(1) - queue._mutex.unlock() - queue._cond.signal() - return conn! diff --git a/examples/http-server/http-server.tm b/examples/http-server/http-server.tm deleted file mode 100644 index 8e8aff7e..00000000 --- a/examples/http-server/http-server.tm +++ /dev/null @@ -1,149 +0,0 @@ -#!/bin/env tomo - -# This file provides an HTTP server module and standalone executable - -use <stdio.h> -use <stdlib.h> -use <string.h> -use <unistd.h> -use <arpa/inet.h> -use <err.h> - -use commands -use pthreads -use patterns - -use ./connection-queue.tm - -func serve(port:Int32, handler:func(request:HTTPRequest -> HTTPResponse), num_threads=16) - connections := ConnectionQueue() - workers : &[@pthread_t] - for i in num_threads - workers.insert(pthread_t.new(func() - repeat - connection := connections.dequeue() - request_text := C_code:Text( - Text_t request = EMPTY_TEXT; - char buf[1024] = {}; - for (ssize_t n; (n = read(@connection, buf, sizeof(buf) - 1)) > 0; ) { - buf[n] = 0; - request = Text$concat(request, Text$from_strn(buf, n)); - if (request.length > 1000000 || strstr(buf, "\r\n\r\n")) - break; - } - request - ) - - request := HTTPRequest.from_text(request_text) or skip - response := handler(request).bytes() - C_code { - if (@response.stride != 1) - List$compact(&@response, 1); - write(@connection, @response.data, @response.length); - close(@connection); - } - )) - - - sock := C_code:Int32( - int s = socket(AF_INET, SOCK_STREAM, 0); - if (s < 0) err(1, "Couldn't connect to socket!"); - - int opt = 1; - if (setsockopt(s, SOL_SOCKET, SO_REUSEADDR, &opt, sizeof(opt)) < 0) - err(1, "Couldn't set socket option"); - - struct sockaddr_in addr = {AF_INET, htons(@port), INADDR_ANY}; - if (bind(s, (struct sockaddr*)&addr, sizeof(addr)) < 0) - err(1, "Couldn't bind to socket"); - if (listen(s, 8) < 0) - err(1, "Couldn't listen on socket"); - - s - ) - - repeat - conn := C_code:Int32(accept(@sock, NULL, NULL)) - stop if conn < 0 - connections.enqueue(conn) - - say("Shutting down...") - for w in workers - w.cancel() - -struct HTTPRequest(method:Text, path:Text, version:Text, headers:[Text], body:Text) - func from_text(text:Text -> HTTPRequest?) - m := text.pattern_captures($Pat'{word} {..} HTTP/{..}{crlf}{..}') or return none - method := m[1] - path := m[2].replace_pattern($Pat'{2+ /}', '/') - version := m[3] - rest := m[-1].pattern_captures($Pat/{..}{2 crlf}{0+ ..}/) or return none - headers := rest[1].split_pattern($Pat/{crlf}/) - body := rest[-1] - return HTTPRequest(method, path, version, headers, body) - -struct HTTPResponse(body:Text, status=200, content_type="text/plain", headers:{Text=Text}={}) - func bytes(r:HTTPResponse -> [Byte]) - body_bytes := r.body.bytes() - extra_headers := (++: "$k: $v\r\n" for k,v in r.headers) or "" - return " - HTTP/1.1 $(r.status) OK\r - Content-Length: $(body_bytes.length + 2)\r - Content-Type: $(r.content_type)\r - Connection: close\r - $extra_headers - \r\n - ".bytes() ++ body_bytes - -func _content_type(file:Path -> Text) - when file.extension() is "html" return "text/html" - is "tm" return "text/html" - is "js" return "text/javascript" - is "css" return "text/css" - else return "text/plain" - -enum RouteEntry(ServeFile(file:Path), Redirect(destination:Text)) - func respond(entry:RouteEntry, request:HTTPRequest -> HTTPResponse) - when entry is ServeFile(file) - body := if file.can_execute() - Command(Text(file)).get_output()! - else - file.read()! - return HTTPResponse(body, content_type=_content_type(file)) - is Redirect(destination) - return HTTPResponse("Found", 302, headers={"Location"=destination}) - -func load_routes(directory:Path -> {Text=RouteEntry}) - routes : &{Text=RouteEntry} - for file in (directory ++ (./*)).glob() - skip unless file.is_file() - contents := file.read() or skip - server_path := "/" ++ "/".join(file.relative_to(directory).components) - if file.base_name() == "index.html" - canonical := server_path.without_suffix("index.html") - routes[server_path] = Redirect(canonical) - routes[canonical] = ServeFile(file) - else if file.extension() == "html" - canonical := server_path.without_suffix(".html") - routes[server_path] = Redirect(canonical) - routes[canonical] = ServeFile(file) - else if file.extension() == "tm" - canonical := server_path.without_suffix(".tm") - routes[server_path] = Redirect(canonical) - routes[canonical] = ServeFile(file) - else - routes[server_path] = ServeFile(file) - return routes[] - -func main(directory:Path, port=Int32(8080)) - say("Serving on port $port") - routes := load_routes(directory) - say("Hosting: $routes") - - serve(port, func(request:HTTPRequest) - if handler := routes[request.path] - return handler.respond(request) - else - return HTTPResponse("Not found!", 404) - ) - diff --git a/examples/http-server/modules.ini b/examples/http-server/modules.ini deleted file mode 100644 index 171e11d2..00000000 --- a/examples/http-server/modules.ini +++ /dev/null @@ -1,8 +0,0 @@ -[pthreads] -version=v1.0 - -[patterns] -version=v1.0 - -[commands] -version=v1.0 diff --git a/examples/http-server/sample-site/foo.html b/examples/http-server/sample-site/foo.html deleted file mode 100644 index 162a7146..00000000 --- a/examples/http-server/sample-site/foo.html +++ /dev/null @@ -1,6 +0,0 @@ -<!DOCTYPE HTML> -<html> - <body> - This is the <b>foo</b> page. - </body> -</html> diff --git a/examples/http-server/sample-site/hello.txt b/examples/http-server/sample-site/hello.txt deleted file mode 100644 index e965047a..00000000 --- a/examples/http-server/sample-site/hello.txt +++ /dev/null @@ -1 +0,0 @@ -Hello diff --git a/examples/http-server/sample-site/index.html b/examples/http-server/sample-site/index.html deleted file mode 100644 index 8e1573bb..00000000 --- a/examples/http-server/sample-site/index.html +++ /dev/null @@ -1,16 +0,0 @@ -<!DOCTYPE HTML> -<html> - <head> - <title>HTTP Example</title> - <link rel="stylesheet" href="styles.css"> - </head> - <body> - <p> - Hello <b>world!</b> - </p> - - <p> - Try going to <a href="/random">/random</a> or <a href="/foo">/foo</a> or <a href="/hello.txt">/hello.txt</a> - </p> - </body> -</html> diff --git a/examples/http-server/sample-site/random.tm b/examples/http-server/sample-site/random.tm deleted file mode 100755 index 153ac2af..00000000 --- a/examples/http-server/sample-site/random.tm +++ /dev/null @@ -1,17 +0,0 @@ -#!/bin/env tomo -use random - -func main() - say(" - <!DOCTYPE HTML> - <html> - <head> - <title>Random Number</title> - <link rel="stylesheet" href="styles.css"> - </head> - <body> - <h1>Random Number</h1> - Your random number is: $(random.int(1,100)) - </body> - </html> - ") diff --git a/examples/http-server/sample-site/styles.css b/examples/http-server/sample-site/styles.css deleted file mode 100644 index f15d25de..00000000 --- a/examples/http-server/sample-site/styles.css +++ /dev/null @@ -1,11 +0,0 @@ -body{ - margin:40px auto; - max-width:650px; - line-height:1.6; - font-size:18px; - color:#444; - padding:0 10px; -} -h1,h2,h3{ - line-height:1.2 -} diff --git a/examples/http/CHANGES.md b/examples/http/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/http/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/http/http.tm b/examples/http/http.tm deleted file mode 100644 index 3fe41ae2..00000000 --- a/examples/http/http.tm +++ /dev/null @@ -1,111 +0,0 @@ -# A simple HTTP library built using CURL - -use -lcurl -use <curl/curl.h> - -struct HTTPResponse(code:Int, body:Text) - -enum _Method(GET, POST, PUT, PATCH, DELETE) - -func _send(method:_Method, url:Text, data:Text?, headers:[Text]=[] -> HTTPResponse) - chunks : @[Text] - save_chunk := func(chunk:CString, size:Int64, n:Int64) - chunks.insert(C_code:Text(Text$from_strn(@chunk, @size*@n))) - return n*size - - C_code { - CURL *curl = curl_easy_init(); - struct curl_slist *chunk = NULL; - curl_easy_setopt(curl, CURLOPT_URL, @(CString(url))); - curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, @save_chunk.fn); - curl_easy_setopt(curl, CURLOPT_WRITEDATA, @save_chunk.userdata); - } - - defer - C_code { - if (chunk) - curl_slist_free_all(chunk); - curl_easy_cleanup(curl); - } - - when method is POST - C_code { - curl_easy_setopt(curl, CURLOPT_POST, 1L); - } - if posting := data - C_code { - curl_easy_setopt(curl, CURLOPT_POSTFIELDS, @(CString(posting))); - } - is PUT - C_code { - curl_easy_setopt(curl, CURLOPT_CUSTOMREQUEST, "PUT"); - } - if putting := data - C_code { - curl_easy_setopt(curl, CURLOPT_POSTFIELDS, @(CString(putting))); - } - is PATCH - C_code { - curl_easy_setopt(curl, CURLOPT_CUSTOMREQUEST, "PATCH"); - } - if patching := data - C_code { - curl_easy_setopt(curl, CURLOPT_POSTFIELDS, @(CString(patching))); - } - is DELETE - C_code { - curl_easy_setopt(curl, CURLOPT_CUSTOMREQUEST, "DELETE"); - } - else - pass - - for header in headers - C_code { - chunk = curl_slist_append(chunk, @(CString(header))); - } - - C_code { - if (chunk) - curl_easy_setopt(curl, CURLOPT_HTTPHEADER, chunk); - } - - code := Int64(0) - C_code { - CURLcode res = curl_easy_perform(curl); - if (res != CURLE_OK) - fprintf(stderr, "curl_easy_perform() failed: %s\n", curl_easy_strerror(res)); - - curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &@code); - } - return HTTPResponse(Int(code), "".join(chunks)) - -func get(url:Text, headers:[Text]=[] -> HTTPResponse) - return _send(GET, url, none, headers) - -func post(url:Text, data="", headers=["Content-Type: application/json", "Accept: application/json"] -> HTTPResponse) - return _send(POST, url, data, headers) - -func put(url:Text, data="", headers=["Content-Type: application/json", "Accept: application/json"] -> HTTPResponse) - return _send(PUT, url, data, headers) - -func patch(url:Text, data="", headers=["Content-Type: application/json", "Accept: application/json"] -> HTTPResponse) - return _send(PATCH, url, data, headers) - -func delete(url:Text, data:Text?=none, headers=["Content-Type: application/json", "Accept: application/json"] -> HTTPResponse) - return _send(DELETE, url, data, headers) - -func main() - say("GET:") - say(get("https://httpbin.org/get").body) - say("Waiting 1sec...") - sleep(1) - say("POST:") - say(post("https://httpbin.org/post", `{"key": "value"}`).body) - say("Waiting 1sec...") - sleep(1) - say("PUT:") - say(put("https://httpbin.org/put", `{"key": "value"}`).body) - say("Waiting 1sec...") - sleep(1) - say("DELETE:") - say(delete("https://httpbin.org/delete").body) diff --git a/examples/ini/CHANGES.md b/examples/ini/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/ini/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/ini/ini.tm b/examples/ini/ini.tm deleted file mode 100644 index 4dc27725..00000000 --- a/examples/ini/ini.tm +++ /dev/null @@ -1,61 +0,0 @@ - -use patterns - -_USAGE := " - Usage: ini <filename> "[section[/key]]" -" -_HELP := " - ini: A .ini config file reader tool. - $_USAGE -" - -func parse_ini(path:Path -> {Text={Text=Text}}) - text := path.read() or exit("Could not read INI file: \[31;1]$(path)\[]") - sections : @{Text=@{Text=Text}} - current_section : @{Text=Text} - - # Line wraps: - text = text.replace_pattern($Pat/\\{1 nl}{0+space}/, " ") - - for line in text.lines() - line = line.trim() - skip if line.starts_with(";") or line.starts_with("#") - if line.matches_pattern($Pat/[?]/) - section_name := line.replace($Pat/[?]/, "@1").trim().lower() - current_section = @{} - sections[section_name] = current_section - else if line.matches_pattern($Pat/{..}={..}/) - key := line.replace_pattern($Pat/{..}={..}/, "@1").trim().lower() - value := line.replace_pattern($Pat/{..}={..}/, "@2").trim() - current_section[key] = value - - return {k=v[] for k,v in sections[]} - -func main(path:Path, key:Text?) - keys := (key or "").split($|/|) - if keys.length > 2 - exit(" - Too many arguments! - $_USAGE - ") - - data := parse_ini(path) - if keys.length < 1 or keys[1] == '*' - say("$data") - return - - section := keys[1].lower() - section_data := data[section] or exit(" - Invalid section name: \[31;1]$section\[] - Valid names: \[1]$(", ".join([k.quoted() for k in data.keys]))\[] - ") - if keys.length < 2 or keys[2] == '*' - say("$section_data") - return - - section_key := keys[2].lower() - value := section_data[section_key] or exit(" - Invalid key: \[31;1]$section_key\[] - Valid keys: \[1]$(", ".join([s.quoted() for s in section_data.keys]))\[] - ") - say(value) diff --git a/examples/ini/modules.ini b/examples/ini/modules.ini deleted file mode 100644 index fb52a859..00000000 --- a/examples/ini/modules.ini +++ /dev/null @@ -1,2 +0,0 @@ -[patterns] -version=v1.0 diff --git a/examples/ini/test.ini b/examples/ini/test.ini deleted file mode 100644 index 782dc76f..00000000 --- a/examples/ini/test.ini +++ /dev/null @@ -1,8 +0,0 @@ -[ Book ] -title = Dirk Gently's Holistic Detective Agency -author = Douglas Adams -published = 1987 - -[ Protagonist ] -name = Dirk Gently -age = 42 diff --git a/examples/log/CHANGES.md b/examples/log/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/log/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/log/log.tm b/examples/log/log.tm deleted file mode 100644 index 4b7893fd..00000000 --- a/examples/log/log.tm +++ /dev/null @@ -1,50 +0,0 @@ -use <time.h> -use <stdio.h> - -timestamp_format := CString("%F %T") - -logfiles : @|Path| - -func _timestamp(->Text) - c_str := C_code:CString( - char *str = GC_MALLOC_ATOMIC(20); - time_t t; time(&t); - struct tm *tm_info = localtime(&t); - strftime(str, 20, "%F %T", tm_info); - str - ) - return c_str.as_text() - -func info(text:Text, newline=yes) - say("\[2]⚫ $text\[]", newline) - for file in logfiles - file.append("$(_timestamp()) [info] $text\n") - -func debug(text:Text, newline=yes) - say("\[32]🟢 $text\[]", newline) - for file in logfiles - file.append("$(_timestamp()) [debug] $text\n") - -func warn(text:Text, newline=yes) - say("\[33;1]🟡 $text\[]", newline) - for file in logfiles - file.append("$(_timestamp()) [warn] $text\n") - -func error(text:Text, newline=yes) - say("\[31;1]🔴 $text\[]", newline) - for file in logfiles - file.append("$(_timestamp()) [error] $text\n") - -func add_logfile(file:Path) - logfiles.add(file) - -func remove_logfile(file:Path) - logfiles.remove(file) - -func main() - add_logfile((./log.txt)) - >> info("Hello") - >> debug("Hello") - >> warn("Hello") - >> error("Hello") - diff --git a/examples/vectors/CHANGES.md b/examples/vectors/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/vectors/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/vectors/vectors.tm b/examples/vectors/vectors.tm deleted file mode 100644 index 1fd0c4fa..00000000 --- a/examples/vectors/vectors.tm +++ /dev/null @@ -1,136 +0,0 @@ -# A math vector library for 2D and 3D vectors of Nums or Ints - -struct Vec2(x,y:Num) - ZERO := Vec2(0, 0) - func plus(a,b:Vec2->Vec2; inline) - return Vec2(a.x+b.x, a.y+b.y) - func minus(a,b:Vec2->Vec2; inline) - return Vec2(a.x-b.x, a.y-b.y) - func times(a,b:Vec2->Vec2; inline) - return Vec2(a.x*b.x, a.y*b.y) - func negative(v:Vec2->Vec2; inline) - return Vec2(-v.x, -v.y) - func dot(a,b:Vec2->Num; inline) - return (a.x-b.x)*(a.x-b.x) + (a.y-b.y)*(a.y-b.y) - func cross(a,b:Vec2->Num; inline) - return a.x*b.y - a.y*b.x - func scaled_by(v:Vec2, k:Num->Vec2; inline) - return Vec2(v.x*k, v.y*k) - func divided_by(v:Vec2, divisor:Num->Vec2; inline) - return Vec2(v.x/divisor, v.y/divisor) - func length(v:Vec2->Num; inline) - return (v.x*v.x + v.y*v.y).sqrt() - func dist(a,b:Vec2->Num; inline) - return a.minus(b).length() - func angle(v:Vec2->Num; inline) - return Num.atan2(v.y, v.x) - func norm(v:Vec2->Vec2; inline) - if v.x == 0 and v.y == 0 - return v - len := v.length() - return Vec2(v.x/len, v.y/len) - func rotated(v:Vec2, radians:Num -> Vec2) - cos := radians.cos() or return v - sin := radians.sin() or return v - return Vec2(cos*v.x - sin*v.y, sin*v.x + cos*v.y) - func mix(a,b:Vec2, amount:Num -> Vec2) - return Vec2( - amount.mix(a.x, b.x), - amount.mix(a.y, b.y), - ) - -struct Vec3(x,y,z:Num) - ZERO := Vec3(0, 0, 0) - func plus(a,b:Vec3->Vec3; inline) - return Vec3(a.x+b.x, a.y+b.y, a.z+b.z) - func minus(a,b:Vec3->Vec3; inline) - return Vec3(a.x-b.x, a.y-b.y, a.z-b.z) - func times(a,b:Vec3->Vec3; inline) - return Vec3(a.x*b.x, a.y*b.y, a.z*b.z) - func negative(v:Vec3->Vec3; inline) - return Vec3(-v.x, -v.y, -v.z) - func dot(a,b:Vec3->Num; inline) - return (a.x-b.x)*(a.x-b.x) + (a.y-b.y)*(a.y-b.y) + (a.z-b.z)*(a.z-b.z) - func cross(a,b:Vec3->Vec3; inline) - return Vec3(a.y*b.z - a.z-b.y, a.z*b.x - a.x*b.z, a.x*b.y - a.y*b.x) - func scaled_by(v:Vec3, k:Num->Vec3; inline) - return Vec3(v.x*k, v.y*k, v.z*k) - func divided_by(v:Vec3, divisor:Num->Vec3; inline) - return Vec3(v.x/divisor, v.y/divisor, v.z/divisor) - func length(v:Vec3->Num; inline) - return (v.x*v.x + v.y*v.y + v.z*v.z).sqrt() - func dist(a,b:Vec3->Num; inline) - return a.minus(b).length() - func norm(v:Vec3->Vec3; inline) - if v.x == 0 and v.y == 0 and v.z == 0 - return v - len := v.length() - return Vec3(v.x/len, v.y/len, v.z/len) - func mix(a,b:Vec3, amount:Num -> Vec3) - return Vec3( - amount.mix(a.x, b.x), - amount.mix(a.y, b.y), - amount.mix(a.z, b.z), - ) - - -struct IVec2(x,y:Int) - ZERO := IVec2(0, 0) - func plus(a,b:IVec2->IVec2; inline) - return IVec2(a.x+b.x, a.y+b.y) - func minus(a,b:IVec2->IVec2; inline) - return IVec2(a.x-b.x, a.y-b.y) - func times(a,b:IVec2->IVec2; inline) - return IVec2(a.x*b.x, a.y*b.y) - func negative(v:IVec2->IVec2; inline) - return IVec2(-v.x, -v.y) - func dot(a,b:IVec2->Int; inline) - return (a.x-b.x)*(a.x-b.x) + (a.y-b.y)*(a.y-b.y) - func cross(a,b:IVec2->Int; inline) - return a.x*b.y - a.y*b.x - func scaled_by(v:IVec2, k:Int->IVec2; inline) - return IVec2(v.x*k, v.y*k) - func divided_by(v:IVec2, divisor:Int->IVec2; inline) - return IVec2(v.x/divisor, v.y/divisor) - func length(v:IVec2->Num; inline) - x := Num(v.x) - y := Num(v.y) - return Num.sqrt(x*x + y*y) - func dist(a,b:IVec2->Num; inline) - return a.minus(b).length() - func angle(v:IVec2->Num; inline) - return Num.atan2(Num(v.y), Num(v.x)) - -struct IVec3(x,y,z:Int) - ZERO := IVec3(0, 0, 0) - func plus(a,b:IVec3->IVec3; inline) - return IVec3(a.x+b.x, a.y+b.y, a.z+b.z) - func minus(a,b:IVec3->IVec3; inline) - return IVec3(a.x-b.x, a.y-b.y, a.z-b.z) - func times(a,b:IVec3->IVec3; inline) - return IVec3(a.x*b.x, a.y*b.y, a.z*b.z) - func negative(v:IVec3->IVec3; inline) - return IVec3(-v.x, -v.y, -v.z) - func dot(a,b:IVec3->Int; inline) - return (a.x-b.x)*(a.x-b.x) + (a.y-b.y)*(a.y-b.y) + (a.z-b.z)*(a.z-b.z) - func cross(a,b:IVec3->IVec3; inline) - return IVec3(a.y*b.z - a.z-b.y, a.z*b.x - a.x*b.z, a.x*b.y - a.y*b.x) - func scaled_by(v:IVec3, k:Int->IVec3; inline) - return IVec3(v.x*k, v.y*k, v.z*k) - func divided_by(v:IVec3, divisor:Int->IVec3; inline) - return IVec3(v.x/divisor, v.y/divisor, v.z/divisor) - func length(v:IVec3->Num; inline) - x := Num(v.x) - y := Num(v.y) - z := Num(v.z) - return Num.sqrt(x*x + y*y + z*z) - func dist(a,b:IVec3->Num; inline) - return a.minus(b).length() - -func main() - >> Vec2(10, 20) - >> Vec2(10, 20) + Vec2(100, 100) - = Vec2(x=110, y=120) - >> Vec3(10, 20, 30) - >> Vec2.ZERO - diff --git a/examples/wrap/CHANGES.md b/examples/wrap/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/examples/wrap/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/examples/wrap/wrap.tm b/examples/wrap/wrap.tm deleted file mode 100644 index bae01739..00000000 --- a/examples/wrap/wrap.tm +++ /dev/null @@ -1,102 +0,0 @@ -HELP := " - wrap: A tool for wrapping lines of text - - usage: wrap [--help] [files...] [--width=80] [--inplace=no] [--min_split=3] [--no-rewrap] [--hyphen='-'] - --help: Print this message and exit - [files...]: The files to wrap (stdin is used if no files are provided) - --width=N: The width to wrap the text - --inplace: Whether or not to perform the modification in-place or print the output - --min-split=N: The minimum amount of text on either end of a hyphenation split - --rewrap|--no-rewrap: Whether to rewrap text that is already wrapped or only split long lines - --hyphen='-': The text to use for hyphenation -" - -UNICODE_HYPHEN := "\{hyphen}" - -func unwrap(text:Text, preserve_paragraphs=yes, hyphen=UNICODE_HYPHEN -> Text) - if preserve_paragraphs - paragraphs := text.split($/{2+ nl}/) - if paragraphs.length > 1 - return "\n\n".join([unwrap(p, hyphen=hyphen, preserve_paragraphs=no) for p in paragraphs]) - - return text.replace("$(hyphen)\n", "") - -func wrap(text:Text, width:Int, min_split=3, hyphen="-" -> Text) - if width <= 0 - fail("Width must be a positive integer, not $width") - - if 2*min_split - hyphen.length > width - fail(" - Minimum word split length ($min_split) is too small for the given wrap width ($width)! - - I can't fit a $(2*min_split - hyphen.length)-wide word on a line without splitting it, - ... and I can't split it without splitting into chunks smaller than $min_split. - ") - - lines : @[Text] - line := "" - for word in text.split($/{whitespace}/) - letters := word.split() - skip if letters.length == 0 - - while not _can_fit_word(line, letters, width) - line_space := width - line.length - if line != "" then line_space -= 1 - - if min_split > 0 and line_space >= min_split + hyphen.length and letters.length >= 2*min_split - # Split word with a hyphen: - split := line_space - hyphen.length - split = split _max_ min_split - split = split _min_ (letters.length - min_split) - if line != "" then line ++= " " - line ++= ((++: letters.to(split)) or "") ++ hyphen - letters = letters.from(split + 1) - else if line == "" - # Force split word without hyphenation: - if line != "" then line ++= " " - line ++= (++: letters.to(line_space)) or "" - letters = letters.from(line_space + 1) - else - pass # Move to next line - - lines.insert(line) - line = "" - - if letters.length > 0 - if line != "" then line ++= " " - line ++= (++: letters) or "" - - if line != "" - lines.insert(line) - - return "\n".join(lines) - -func _can_fit_word(line:Text, letters:[Text], width:Int -> Bool; inline) - if line == "" - return letters.length <= width - else - return line.length + 1 + letters.length <= width - -func main(files:[Path], width=80, inplace=no, min_split=3, rewrap=yes, hyphen=UNICODE_HYPHEN) - if files.length == 0 - files = [(/dev/stdin)] - - for file in files - text := file.read() or exit("Could not read file: $file") - - if rewrap - text = unwrap(text) - - out := if file.is_file() and inplace - file - else - (/dev/stdout) - - first := yes - wrapped_paragraphs : @[Text] - for paragraph in text.split($/{2+ nl}/) - wrapped_paragraphs.insert( - wrap(paragraph, width=width, min_split=min_split, hyphen=hyphen) - ) - - out.write("\n\n".join(wrapped_paragraphs[]) ++ "\n") diff --git a/lib/README.md b/lib/README.md deleted file mode 100644 index 31de5b92..00000000 --- a/lib/README.md +++ /dev/null @@ -1,19 +0,0 @@ -# Core Libraries - -This folder contains some libraries that are installed by default when -installing Tomo. - -## Libraries - -Libraries can be installed with `tomo -IL ./library-folder` - -- [base64](base64/): A base64 encoding/decoding library. -- [commands](commands/): A library for running commands. -- [core](core/): Bundling up commonly used libraries into a single library. -- [json](json/): JSON parsing and encoding. -- [patterns](patterns/): Pattern matching for text. -- [pthreads](pthreads/): A POSIX threads library. -- [random](random/): Pseudorandom number generators. -- [shell](shell/): A DSL for running shell commands. -- [time](time/): A module for working with dates and times. -- [uuid](uuid/): A universally unique identifier library. diff --git a/lib/base64/CHANGES.md b/lib/base64/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/base64/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/base64/README.md b/lib/base64/README.md deleted file mode 100644 index 8db0f8ec..00000000 --- a/lib/base64/README.md +++ /dev/null @@ -1,3 +0,0 @@ -# Base64 - -This is a library for encoding/decoding Base64 values. diff --git a/lib/base64/base64.tm b/lib/base64/base64.tm deleted file mode 100644 index bf512a83..00000000 --- a/lib/base64/base64.tm +++ /dev/null @@ -1,96 +0,0 @@ -# Base 64 encoding and decoding - -_enc := "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/".bytes() - -_EQUAL_BYTE := Byte(0x3D) - -_dec : [Byte] = [ - 255, 255, 255, 255, 255, 255, 255, 255, - 255, 255, 255, 255, 255, 255, 255, 255, - 255, 255, 255, 255, 255, 255, 255, 255, - 255, 255, 255, 255, 255, 255, 255, 255, - 255, 255, 255, 255, 255, 255, 255, 255, - 255, 255, 255, 62, 255, 255, 255, 63, - 52, 53, 54, 55, 56, 57, 58, 59, - 60, 61, 255, 255, 255, 255, 255, 255, - 255, 0, 1, 2, 3, 4, 5, 6, - 7, 8, 9, 10, 11, 12, 13, 14, - 15, 16, 17, 18, 19, 20, 21, 22, - 23, 24, 25, 255, 255, 255, 255, 255, - 255, 26, 27, 28, 29, 30, 31, 32, - 33, 34, 35, 36, 37, 38, 39, 40, - 41, 42, 43, 44, 45, 46, 47, 48, - 49, 50, 51, 255, 255, 255, 255, 255, -] - -lang Base64 - func parse(text:Text -> Base64?) - return Base64.from_bytes(text.bytes()) - - func from_bytes(bytes:[Byte] -> Base64?) - output := &[Byte(0) for _ in bytes.length * 4 / 3 + 4] - src := Int64(1) - dest := Int64(1) - while src + 2 <= Int64(bytes.length) - chunk24 := ( - (Int32(bytes[src]) <<< 16) or (Int32(bytes[src+1]) <<< 8) or Int32(bytes[src+2]) - ) - src += 3 - - output[dest] = _enc[1 + ((chunk24 >>> 18) and 0b111111)] - output[dest+1] = _enc[1 + ((chunk24 >>> 12) and 0b111111)] - output[dest+2] = _enc[1 + ((chunk24 >>> 6) and 0b111111)] - output[dest+3] = _enc[1 + (chunk24 and 0b111111)] - dest += 4 - - if src + 1 == bytes.length - chunk16 := ( - (Int32(bytes[src]) <<< 8) or Int32(bytes[src+1]) - ) - output[dest] = _enc[1 + ((chunk16 >>> 10) and 0b11111)] - output[dest+1] = _enc[1 + ((chunk16 >>> 4) and 0b111111)] - output[dest+2] = _enc[1 + ((chunk16 <<< 2)and 0b111111)] - output[dest+3] = _EQUAL_BYTE - else if src == bytes.length - chunk8 := Int32(bytes[src]) - output[dest] = _enc[1 + ((chunk8 >>> 2) and 0b111111)] - output[dest+1] = _enc[1 + ((chunk8 <<< 4) and 0b111111)] - output[dest+2] = _EQUAL_BYTE - output[dest+3] = _EQUAL_BYTE - - return Base64.from_text(Text.from_bytes(output[]) or return none) - - func decode_text(b64:Base64 -> Text?) - return Text.from_bytes(b64.decode_bytes() or return none) - - func decode_bytes(b64:Base64 -> [Byte]?) - bytes := b64.text.bytes() - output := &[Byte(0) for _ in bytes.length/4 * 3] - src := Int64(1) - dest := Int64(1) - while src + 3 <= Int64(bytes.length) - chunk24 := ( - (Int32(_dec[1+bytes[src]]) <<< 18) or - (Int32(_dec[1+bytes[src+1]]) <<< 12) or - (Int32(_dec[1+bytes[src+2]]) <<< 6) or - Int32(_dec[1+bytes[src+3]]) - ) - src += 4 - - output[dest] = Byte((chunk24 >>> 16) and 0xFF) - output[dest+1] = Byte((chunk24 >>> 8) and 0xFF) - output[dest+2] = Byte(chunk24 and 0xFF) - dest += 3 - - while output[-1] == 0xFF - output[] = output.to(-2) - - return output[] - -func main(input=(/dev/stdin), decode=no) - if decode - b := Base64.from_text(input.read()!) - say(b.decode_text()!) - else - text := input.read()! - say(Base64.parse(text)!.text) diff --git a/lib/commands/CHANGES.md b/lib/commands/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/commands/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/commands/README.md b/lib/commands/README.md deleted file mode 100644 index 040f4bd5..00000000 --- a/lib/commands/README.md +++ /dev/null @@ -1,5 +0,0 @@ -# Commands - -This module provides a way to run executable programs and get their output. You -can also feed in text to the programs' `stdin`. Think of it as `popen()` on -steroids. diff --git a/lib/commands/commands.c b/lib/commands/commands.c deleted file mode 100644 index 110ae696..00000000 --- a/lib/commands/commands.c +++ /dev/null @@ -1,295 +0,0 @@ -// Logic for running system commands - -#pragma once - -#include <errno.h> -#include <gc.h> -#include <poll.h> -#include <pthread.h> -#include <signal.h> -#include <spawn.h> -#include <stdbool.h> -#include <stdint.h> -#include <string.h> -#include <sys/param.h> -#include <sys/wait.h> -#include <unistd.h> - -// This is a workaround fix for an issue on some systems that don't have `__GLIBC__` defined -// and run into problems with <unistr.h> - -#ifndef __GLIBC__ -#define __GLIBC__ 2 -#include <unistr.h> // IWYU pragma: export -#undef __GLIBC__ -#else -#include <unistr.h> // IWYU pragma: export -#endif - -#define READ_END 0 -#define WRITE_END 1 - -static void xpipe(int fd[2]) { - if (pipe(fd) != 0) fail("Failed to create pipe: ", strerror(errno)); -} - -int run_command(Text_t exe, List_t arg_list, Table_t env_table, OptionalList_t input_bytes, List_t *output_bytes, - List_t *error_bytes) { - pthread_testcancel(); - - struct sigaction sa = {.sa_handler = SIG_IGN}, oldint, oldquit; - sigaction(SIGINT, &sa, &oldint); - sigaction(SIGQUIT, &sa, &oldquit); - sigaddset(&sa.sa_mask, SIGCHLD); - sigset_t old, reset; - sigprocmask(SIG_BLOCK, &sa.sa_mask, &old); - sigemptyset(&reset); - if (oldint.sa_handler != SIG_IGN) sigaddset(&reset, SIGINT); - if (oldquit.sa_handler != SIG_IGN) sigaddset(&reset, SIGQUIT); - posix_spawnattr_t attr; - posix_spawnattr_init(&attr); - posix_spawnattr_setsigmask(&attr, &old); - posix_spawnattr_setsigdefault(&attr, &reset); - posix_spawnattr_setflags(&attr, POSIX_SPAWN_SETSIGDEF | POSIX_SPAWN_SETSIGMASK); - - int child_inpipe[2], child_outpipe[2], child_errpipe[2]; - if (input_bytes.length >= 0) xpipe(child_inpipe); - if (output_bytes) xpipe(child_outpipe); - if (error_bytes) xpipe(child_errpipe); - - posix_spawn_file_actions_t actions; - posix_spawn_file_actions_init(&actions); - if (input_bytes.length >= 0) { - posix_spawn_file_actions_adddup2(&actions, child_inpipe[READ_END], STDIN_FILENO); - posix_spawn_file_actions_addclose(&actions, child_inpipe[WRITE_END]); - } - if (output_bytes) { - posix_spawn_file_actions_adddup2(&actions, child_outpipe[WRITE_END], STDOUT_FILENO); - posix_spawn_file_actions_addclose(&actions, child_outpipe[READ_END]); - } - if (error_bytes) { - posix_spawn_file_actions_adddup2(&actions, child_errpipe[WRITE_END], STDERR_FILENO); - posix_spawn_file_actions_addclose(&actions, child_errpipe[READ_END]); - } - - const char *exe_str = Text$as_c_string(exe); - - List_t arg_strs = {}; - List$insert_value(&arg_strs, exe_str, I(0), sizeof(char *)); - for (int64_t i = 0; i < arg_list.length; i++) - List$insert_value(&arg_strs, Text$as_c_string(*(Text_t *)(arg_list.data + i * arg_list.stride)), I(0), - sizeof(char *)); - List$insert_value(&arg_strs, NULL, I(0), sizeof(char *)); - char **args = arg_strs.data; - - extern char **environ; - char **env = environ; - if (env_table.entries.length > 0) { - List_t env_list = {}; // List of const char* - for (char **e = environ; *e; e++) - List$insert(&env_list, e, I(0), sizeof(char *)); - - for (int64_t i = 0; i < env_table.entries.length; i++) { - struct { - Text_t key, value; - } *entry = env_table.entries.data + env_table.entries.stride * i; - const char *env_entry = String(entry->key, "=", entry->value); - List$insert(&env_list, &env_entry, I(0), sizeof(char *)); - } - List$insert_value(&env_list, NULL, I(0), sizeof(char *)); - assert(env_list.stride == sizeof(char *)); - env = env_list.data; - } - - pid_t pid; - int ret = exe_str[0] == '/' ? posix_spawn(&pid, exe_str, &actions, &attr, args, env) - : posix_spawnp(&pid, exe_str, &actions, &attr, args, env); - if (ret != 0) return -1; - - posix_spawnattr_destroy(&attr); - posix_spawn_file_actions_destroy(&actions); - - if (input_bytes.length >= 0) close(child_inpipe[READ_END]); - if (output_bytes) close(child_outpipe[WRITE_END]); - if (error_bytes) close(child_errpipe[WRITE_END]); - - struct pollfd pollfds[3] = {}; - if (input_bytes.length >= 0) pollfds[0] = (struct pollfd){.fd = child_inpipe[WRITE_END], .events = POLLOUT}; - if (output_bytes) pollfds[1] = (struct pollfd){.fd = child_outpipe[WRITE_END], .events = POLLIN}; - if (error_bytes) pollfds[2] = (struct pollfd){.fd = child_errpipe[WRITE_END], .events = POLLIN}; - - if (input_bytes.length > 0 && input_bytes.stride != 1) List$compact(&input_bytes, sizeof(char)); - if (output_bytes) *output_bytes = (List_t){.atomic = 1, .stride = 1, .length = 0}; - if (error_bytes) *error_bytes = (List_t){.atomic = 1, .stride = 1, .length = 0}; - - while (input_bytes.length > 0 || output_bytes || error_bytes) { - (void)poll(pollfds, sizeof(pollfds) / sizeof(pollfds[0]), -1); // Wait for data or readiness - bool did_something = false; - if (input_bytes.length >= 0 && pollfds[0].revents) { - if (input_bytes.length > 0) { - ssize_t written = write(child_inpipe[WRITE_END], input_bytes.data, (size_t)input_bytes.length); - if (written > 0) { - input_bytes.data += written; - input_bytes.length -= (int64_t)written; - did_something = true; - } else if (written < 0) { - close(child_inpipe[WRITE_END]); - pollfds[0].events = 0; - } - } - if (input_bytes.length <= 0) { - close(child_inpipe[WRITE_END]); - pollfds[0].events = 0; - } - } - char buf[256]; - if (output_bytes && pollfds[1].revents) { - ssize_t n = read(child_outpipe[READ_END], buf, sizeof(buf)); - did_something = did_something || (n > 0); - if (n <= 0) { - close(child_outpipe[READ_END]); - pollfds[1].events = 0; - } else if (n > 0) { - if (output_bytes->free < n) { - output_bytes->data = GC_REALLOC(output_bytes->data, (size_t)(output_bytes->length + n)); - output_bytes->free = 0; - } - memcpy(output_bytes->data + output_bytes->length, buf, (size_t)n); - output_bytes->length += n; - } - } - if (error_bytes && pollfds[2].revents) { - ssize_t n = read(child_errpipe[READ_END], buf, sizeof(buf)); - did_something = did_something || (n > 0); - if (n <= 0) { - close(child_errpipe[READ_END]); - pollfds[2].events = 0; - } else if (n > 0) { - if (error_bytes->free < n) { - error_bytes->data = GC_REALLOC(error_bytes->data, (size_t)(error_bytes->length + n)); - error_bytes->free = 0; - } - memcpy(error_bytes->data + error_bytes->length, buf, (size_t)n); - error_bytes->length += n; - } - } - if (!did_something) break; - } - - int status = 0; - if (ret == 0) { - while (waitpid(pid, &status, 0) < 0 && errno == EINTR) { - if (WIFEXITED(status) || WIFSIGNALED(status)) break; - else if (WIFSTOPPED(status)) kill(pid, SIGCONT); - } - } - - if (input_bytes.length >= 0) close(child_inpipe[WRITE_END]); - if (output_bytes) close(child_outpipe[READ_END]); - if (error_bytes) close(child_errpipe[READ_END]); - - sigaction(SIGINT, &oldint, NULL); - sigaction(SIGQUIT, &oldquit, NULL); - sigprocmask(SIG_SETMASK, &old, NULL); - - if (ret) errno = ret; - return status; -} - -typedef struct { - pid_t pid; - FILE *out; -} child_info_t; - -static void _line_reader_cleanup(child_info_t *child) { - if (child && child->out) { - fclose(child->out); - child->out = NULL; - } - if (child->pid) { - kill(child->pid, SIGTERM); - child->pid = 0; - } -} - -static Text_t _next_line(child_info_t *child) { - if (!child || !child->out) return NONE_TEXT; - - char *line = NULL; - size_t size = 0; - ssize_t len = getline(&line, &size, child->out); - if (len <= 0) { - _line_reader_cleanup(child); - return NONE_TEXT; - } - - while (len > 0 && (line[len - 1] == '\r' || line[len - 1] == '\n')) - --len; - - if (u8_check((uint8_t *)line, (size_t)len) != NULL) fail("Invalid UTF8!"); - - Text_t line_text = Text$from_strn(line, len); - free(line); - return line_text; -} - -OptionalClosure_t command_by_line(Text_t exe, List_t arg_list, Table_t env_table) { - posix_spawnattr_t attr; - posix_spawnattr_init(&attr); - - int child_outpipe[2]; - xpipe(child_outpipe); - - posix_spawn_file_actions_t actions; - posix_spawn_file_actions_init(&actions); - posix_spawn_file_actions_adddup2(&actions, child_outpipe[WRITE_END], STDOUT_FILENO); - posix_spawn_file_actions_addclose(&actions, child_outpipe[READ_END]); - - const char *exe_str = Text$as_c_string(exe); - - List_t arg_strs = {}; - List$insert_value(&arg_strs, exe_str, I(0), sizeof(char *)); - for (int64_t i = 0; i < arg_list.length; i++) - List$insert_value(&arg_strs, Text$as_c_string(*(Text_t *)(arg_list.data + i * arg_list.stride)), I(0), - sizeof(char *)); - List$insert_value(&arg_strs, NULL, I(0), sizeof(char *)); - char **args = arg_strs.data; - - extern char **environ; - char **env = environ; - if (env_table.entries.length > 0) { - List_t env_list = {}; // List of const char* - for (char **e = environ; *e; e++) - List$insert(&env_list, e, I(0), sizeof(char *)); - - for (int64_t i = 0; i < env_table.entries.length; i++) { - struct { - Text_t key, value; - } *entry = env_table.entries.data + env_table.entries.stride * i; - const char *env_entry = String(entry->key, "=", entry->value); - List$insert(&env_list, &env_entry, I(0), sizeof(char *)); - } - List$insert_value(&env_list, NULL, I(0), sizeof(char *)); - assert(env_list.stride == sizeof(char *)); - env = env_list.data; - } - - pid_t pid; - int ret = exe_str[0] == '/' ? posix_spawn(&pid, exe_str, &actions, &attr, args, env) - : posix_spawnp(&pid, exe_str, &actions, &attr, args, env); - if (ret != 0) return NONE_CLOSURE; - - posix_spawnattr_destroy(&attr); - posix_spawn_file_actions_destroy(&actions); - - close(child_outpipe[WRITE_END]); - - child_info_t *child_info = GC_MALLOC(sizeof(child_info_t)); - child_info->out = fdopen(child_outpipe[READ_END], "r"); - child_info->pid = pid; - GC_register_finalizer(child_info, (void *)_line_reader_cleanup, NULL, NULL, NULL); - return (Closure_t){.fn = (void *)_next_line, .userdata = child_info}; -} - -#undef READ_END -#undef WRITE_END diff --git a/lib/commands/commands.tm b/lib/commands/commands.tm deleted file mode 100644 index 8a131042..00000000 --- a/lib/commands/commands.tm +++ /dev/null @@ -1,90 +0,0 @@ -# Functions for running system commands - -use ./commands.c - -extern run_command : func(exe:Text, args:[Text], env:{Text=Text}, input:[Byte]?, output:&[Byte]?, error:&[Byte]? -> Int32) -extern command_by_line : func(exe:Text, args:[Text], env:{Text=Text} -> func(->Text?)?) - -enum ExitType(Exited(status:Int32), Signaled(signal:Int32), Failed) - func succeeded(e:ExitType -> Bool) - when e is Exited(status) return (status == 0) - else return no - - func or_fail(e:ExitType, message:Text?=none) - if not e.succeeded() - fail(message or "Program failed: $e") - -struct ProgramResult(output:[Byte], errors:[Byte], exit_type:ExitType) - func or_fail(r:ProgramResult, message:Text?=none -> ProgramResult) - when r.exit_type is Exited(status) - if status != 0 - fail(message or "Program failed: $r") - else fail(message or "Program failed: $r") - return r - - func output_text(r:ProgramResult, trim_newline=yes -> Text?) - when r.exit_type is Exited(status) - if status == 0 - if text := Text.from_bytes(r.output) - if trim_newline - text = text.without_suffix("\n") - return text - else return none - return none - - func error_text(r:ProgramResult -> Text?) - when r.exit_type is Exited(status) - if status == 0 - return Text.from_bytes(r.errors) - else return none - return none - - func succeeded(r:ProgramResult -> Bool) - when r.exit_type is Exited(status) - return (status == 0) - else - return no - -struct Command(command:Text, args:[Text]=[], env:{Text=Text}={}) - func from_path(path:Path, args:[Text]=[], env:{Text=Text}={} -> Command) - return Command(Text(path), args, env) - - func result(command:Command, input="", input_bytes:[Byte]=[] -> ProgramResult) - if input.length > 0 - (&input_bytes).insert_all(input.bytes()) - - output : [Byte] - errors : [Byte] - status := run_command(command.command, command.args, command.env, input_bytes, &output, &errors) - - if C_code:Bool(WIFEXITED(@status)) - return ProgramResult(output, errors, ExitType.Exited(C_code:Int32(WEXITSTATUS(@status)))) - - if C_code:Bool(WIFSIGNALED(@status)) - return ProgramResult(output, errors, ExitType.Signaled(C_code:Int32(WTERMSIG(@status)))) - - return ProgramResult(output, errors, ExitType.Failed) - - func run(command:Command, -> ExitType) - status := run_command(command.command, command.args, command.env, none, none, none) - - if C_code:Bool(WIFEXITED(@status)) - return ExitType.Exited(C_code:Int32(WEXITSTATUS(@status))) - - if C_code:Bool(WIFSIGNALED(@status)) - return ExitType.Signaled(C_code:Int32(WTERMSIG(@status))) - - return ExitType.Failed - - func get_output(command:Command, input="", trim_newline=yes -> Text?) - return command.result(input=input).output_text(trim_newline=trim_newline) - - func get_output_bytes(command:Command, input="", input_bytes:[Byte]=[] -> [Byte]?) - result := command.result(input=input, input_bytes=input_bytes) - when result.exit_type is Exited(status) - if status == 0 return result.output - return none - else return none - - func by_line(command:Command -> func(->Text?)?) - return command_by_line(command.command, command.args, command.env) diff --git a/lib/core/CHANGES.md b/lib/core/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/core/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/core/core.tm b/lib/core/core.tm deleted file mode 100644 index 5ed77756..00000000 --- a/lib/core/core.tm +++ /dev/null @@ -1,9 +0,0 @@ -# This file just uses all the most commonly used standard -# library modules so you don't have to import them one-by-one - -use patterns_v1.1 -use commands_v1.0 -use shell_v1.0 -use pthreads_v1.0 -use random_v1.0 -use time_v1.0 diff --git a/lib/json/CHANGES.md b/lib/json/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/json/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/json/README.md b/lib/json/README.md deleted file mode 100644 index 33e2101a..00000000 --- a/lib/json/README.md +++ /dev/null @@ -1,18 +0,0 @@ -# JSON - -This is a library for encoding/decoding JSON values. - -## Usage - -```tomo ->> j := JSON({"key1"=123, "key2"=[yes, {"ok"="inner"}, JSON.Null]}) -= JSON.Object({"key1"=Number(123), "key2"=Array([Boolean(yes), Object({"ok"=String("inner")}), Null])}) - -say("$(j.encode())") -say("$(j.pretty_print())") - -when JSON.parse("[1, null, true]") is Success(obj) - >> obj -is Failure(msg) - fail("Failed to parse JSON: $msg") -``` diff --git a/lib/json/json.tm b/lib/json/json.tm deleted file mode 100644 index 8127ce52..00000000 --- a/lib/json/json.tm +++ /dev/null @@ -1,173 +0,0 @@ -# Base 64 encoding and decoding -use patterns - -enum JSONDecodeResult( - Success(json:JSON) - Failure(reason:Text) -) - func invalid(text:Text -> JSONDecodeResult) - return Failure("Unrecognized JSON: $(text.quoted())") - -extend Text - func json_quoted(text:Text -> Text) - return '"' ++ text.translate({ - "\\"="\\\\", - '"'='\\"', - "\f"="\\f", - "\r"="\\r", - "\n"="\\n", - "\b"="\\b", - "\t"="\\t", - }) ++ '"' - -enum JSON( - Object(items:{Text=JSON}) - Array(items:[JSON]) - Boolean(value:Bool) - String(text:Text) - Number(n:Num) - Null -) - func encode(j:JSON -> Text) - when j is Object(items) - return "{" ++ ", ".join([ - '$(k.json_quoted()): $(v.encode())' - for k,v in items - ]) ++ "}" - is Array(items) - return "[" ++ ", ".join([item.encode() for item in items]) ++ "]" - is Boolean(value) - return (if value then "true" else "false") - is String(text) - return text.json_quoted() - is Number(n) - return "$n" - is Null - return "null" - - func pretty_print(j:JSON, max_line:Int=80, indent:Text=" ", current_indent:Text="" -> Text) - inline := j.encode() - if inline.length > max_line - next_indent := current_indent ++ indent - when j is Object(items) - return "{\n$next_indent" ++ ",\n$next_indent".join([ - '$(k.json_quoted()): $(v.pretty_print(max_line, indent, next_indent))' - for k,v in items - ]) ++ "\n$current_indent}" - is Array(items) - return "[\n$next_indent" ++ ",\n$next_indent".join([item.pretty_print(max_line, indent, next_indent) for item in items]) ++ "\n$current_indent]" - else pass - - return inline - - func parse_text(text:Text, remainder:&Text? = none -> JSONDecodeResult) - if text.starts_with('"') - string := "" - pos := 2 - escapes := {"n"="\n", "t"="\t", "r"="\r", '"'='"', "\\"="\\", "/"="/", "b"="\b", "f"="\f"} - while pos <= text.length - c := text[pos] - if c == '"' - if remainder - remainder[] = text.from(pos + 1) - return Success(JSON.String(string)) - - if c == "\\" - stop if pos + 1 > text.length - - if esc := escapes[text[pos+1]] - string ++= esc - pos += 2 - else if m := text.matching_pattern($Pat/u{4 digit}/) - string ++= Text.from_codepoints([Int32.parse(m.captures[1])!]) - pos += 1 + m.text.length - else - if remainder - remainder[] = text - return JSONDecodeResult.invalid(text) - else - string ++= c - pos += 1 - - if remainder - remainder[] = text - return JSONDecodeResult.invalid(text) - - func parse(text:Text, remainder:&Text? = none, trailing_commas:Bool=no -> JSONDecodeResult) - if text.starts_with("true", remainder) - return Success(JSON.Boolean(yes)) - else if text.starts_with("false", remainder) - return Success(JSON.Boolean(no)) - else if text.starts_with("null", remainder) - return Success(JSON.Null) - else if n := Num.parse(text, remainder) - return Success(JSON.Number(n)) - else if text.starts_with('"') - return JSON.parse_text(text, remainder) - else if text.starts_with("[") - elements : &[JSON] - text = text.from(2).trim_pattern($Pat"{whitespace}", right=no) - repeat - when JSON.parse(text, &text) is Success(elem) - elements.insert(elem) - else stop - - if delim := text.matching_pattern($Pat'{0+ ws},{0+ ws}') - text = text.from(delim.text.length + 1) - else stop - - if trailing_commas - if delim := text.matching_pattern($Pat'{0+ ws},{0+ ws}') - text = text.from(delim.text.length + 1) - - if terminator := text.matching_pattern($Pat'{0+ ws}]') - if remainder - remainder[] = text.from(terminator.text.length + 1) - return Success(JSON.Array(elements)) - else if text.starts_with("{") - object : &{Text=JSON} - text = text.from(2).trim_pattern($Pat"{whitespace}", right=no) - repeat - key_text := text - when JSON.parse_text(text, &text) is Success(key) - if separator := text.matching_pattern($Pat'{0+ ws}:{0+ ws}') - text = text.from(separator.text.length + 1) - else - return JSONDecodeResult.invalid(text) - - when JSON.parse(text, &text) is Success(value) - when key is String(str) - object[str] = value - else - return JSONDecodeResult.invalid(key_text) - else - return JSONDecodeResult.invalid(text) - else stop - - if delim := text.matching_pattern($Pat'{0+ ws},{0+ ws}') - text = text.from(delim.text.length + 1) - else stop - - if trailing_commas - if delim := text.matching_pattern($Pat'{0+ ws},{0+ ws}') - text = text.from(delim.text.length + 1) - - if terminator := text.matching_pattern($Pat'{0+ ws}{}}') - if remainder - remainder[] = text.from(terminator.text.length + 1) - return Success(JSON.Object(object)) - - return JSONDecodeResult.invalid(text) - -func main(input=(/dev/stdin), pretty_print:Bool = no, trailing_commas:Bool = yes) - text := (input.read() or exit("Invalid file: $input")).trim_pattern($Pat"{whitespace}") - while text.length > 0 - when JSON.parse(text, remainder=&text, trailing_commas=trailing_commas) is Success(json) - if pretty_print - say(json.pretty_print()) - else - say(json.encode()) - is Failure(msg) - exit("\033[31;1m$msg\033[m", code=1) - - text = text.trim_pattern($Pat"{whitespace}") diff --git a/lib/patterns/CHANGES.md b/lib/patterns/CHANGES.md deleted file mode 100644 index cf6254cb..00000000 --- a/lib/patterns/CHANGES.md +++ /dev/null @@ -1,8 +0,0 @@ -# Version History - -## v1.1 -- Added `Text.matching_pattern(text:Text, pattern:Pattern, pos:Int = 1 -> PatternMatch?)` - -## v1.0 - -Initial version diff --git a/lib/patterns/README.md b/lib/patterns/README.md deleted file mode 100644 index faf2854e..00000000 --- a/lib/patterns/README.md +++ /dev/null @@ -1,444 +0,0 @@ -# Text Pattern Matching - -As an alternative to full regular expressions, Tomo provides a limited text -matching pattern syntax that is intended to solve 80% of use cases in under 1% -of the code size (PCRE's codebase is roughly 150k lines of code, and Tomo's -pattern matching code is a bit under 1k lines of code). Tomo's pattern matching -syntax is highly readable and works well for matching literal text without -getting [leaning toothpick syndrome](https://en.wikipedia.org/wiki/Leaning_toothpick_syndrome). - -For more advanced use cases, consider linking against a C library for regular -expressions or pattern matching. - -`Pat` is a [domain-specific language](docs/langs.md), in other words, it's -like a `Text`, but it has a distinct type. - -Patterns are used in a small, but very powerful API that handles many text -functions that would normally be handled by a more extensive API: - -- [`by_pattern(text:Text, pattern:Pat -> func(->PatternMatch?))`](#by_pattern) -- [`by_pattern_split(text:Text, pattern:Pat -> func(->Text?))`](#by_pattern_split) -- [`each_pattern(text:Text, pattern:Pat, fn:func(m:PatternMatch), recursive=yes)`](#each_pattern) -- [`find_patterns(text:Text, pattern:Pat -> [PatternMatch])`](#find_patterns) -- [`has_pattern(text:Text, pattern:Pat -> Bool)`](#has_pattern) -- [`map_pattern(text:Text, pattern:Pat, fn:func(m:PatternMatch -> Text), recursive=yes -> Text)`](#map_pattern) -- [`matches_pattern(text:Text, pattern:Pat -> Bool)`](#matches_pattern) -- [`pattern_captures(text:Text, pattern:Pat -> [Text]?)`](#pattern_captures) -- [`replace_pattern(text:Text, pattern:Pat, replacement:Text, backref="@", recursive=yes -> Text)`](#replace_pattern) -- [`split_pattern(text:Text, pattern:Pat -> [Text])`](#split_pattern) -- [`translate_patterns(text:Text, replacements:{Pat,Text}, backref="@", recursive=yes -> Text)`](#translate_patterns) -- [`trim_pattern(text:Text, pattern=$Pat"{space}", left=yes, right=yes -> Text)`](#trim_pattern) - -## Matches - -Pattern matching functions work with a type called `PatternMatch` that has three fields: - -- `text`: The full text of the match. -- `index`: The index in the text where the match was found. -- `captures`: A list containing the matching text of each non-literal pattern group. - -See [Text Functions](text.md#Text-Functions) for the full API documentation. - -## Syntax - -Patterns have three types of syntax: - -- `{` followed by an optional count (`n`, `n-m`, or `n+`), followed by an - optional `!` to negate the pattern, followed by an optional pattern name or - Unicode character name, followed by a required `}`. - -- Any matching pair of quotes or parentheses or braces with a `?` in the middle - (e.g. `"?"` or `(?)`). - -- Any other character is treated as a literal to be matched exactly. - -## Named Patterns - -Named patterns match certain pre-defined patterns that are commonly useful. To -use a named pattern, use the syntax `{name}`. Names are case-insensitive and -mostly ignore spaces, underscores, and dashes. - -- `..` - Any character (note that a single `.` would mean the literal period - character). -- `digit` - A unicode digit -- `email` - an email address -- `emoji` - an emoji -- `end` - the very end of the text -- `id` - A unicode identifier -- `int` - One or more digits with an optional `-` (minus sign) in front -- `ip` - an IP address (IPv4 or IPv6) -- `ipv4` - an IPv4 address -- `ipv6` - an IPv6 address -- `nl`/`newline`/`crlf` - A line break (either `\r\n` or `\n`) -- `num` - One or more digits with an optional `-` (minus sign) in front and an optional `.` and more digits after -- `start` - the very start of the text -- `uri` - a URI -- `url` - a URL (URI that specifically starts with `http://`, `https://`, `ws://`, `wss://`, or `ftp://`) -- `word` - A unicode identifier (same as `id`) - -For non-alphabetic characters, any single character is treated as matching -exactly that character. For example, `{1{}` matches exactly one `{` -character. Or, `{1.}` matches exactly one `.` character. - -Patterns can also use any Unicode property name. Some helpful ones are: - -- `hex` - Hexidecimal digits -- `lower` - Lowercase letters -- `space` - The space character -- `upper` - Uppercase letters -- `whitespace` - Whitespace characters - -Patterns may also use exact Unicode codepoint names. For example: `{1 latin -small letter A}` matches `a`. - -## Negating Patterns - -If an exclamation mark (`!`) is placed before a pattern's name, then characters -are matched only when they _don't_ match the pattern. For example, `{!alpha}` -will match all characters _except_ alphabetic ones. - -## Interpolating Text and Escaping - -To escape a character in a pattern (e.g. if you want to match the literal -character `?`), you can use the syntax `{1 ?}`. This is almost never necessary -unless you have text that looks like a Tomo text pattern and has something like -`{` or `(?)` inside it. - -However, if you're trying to do an exact match of arbitrary text values, you'll -want to have the text automatically escaped. Fortunately, Tomo's injection-safe -DSL text interpolation supports automatic text escaping. This means that if you -use text interpolation with the `$` sign to insert a text value, the value will -be automatically escaped using the `{1 ?}` rule described above: - -```tomo -# Risk of code injection (would cause an error because 'xxx' is not a valid -# pattern name: ->> user_input := get_user_input() -= "{xxx}" - -# Interpolation automatically escapes: ->> $/$user_input/ -= $/{1{}..xxx}/ - -# This is: `{ 1{ }` (one open brace) followed by the literal text "..xxx}" - -# No error: ->> some_text.find($/$user_input/) -= 0 -``` - -If you prefer, you can also use this to insert literal characters: - -```tomo ->> $/literal $"{..}"/ -= $/literal {1{}..}/ -``` - -## Repetitions - -By default, named patterns match 1 or more repetitions, but you can specify how -many repetitions you want by putting a number or range of numbers first using -`n` (exactly `n` repetitions), `n-m` (between `n` and `m` repetitions), or `n+` -(`n` or more repetitions): - -``` -{4-5 alpha} -0x{hex} -{4 digit}-{2 digit}-{2 digit} -{2+ space} -{0-1 question mark} -``` - - -# Methods - -### `by_pattern` -Returns an iterator function that yields `PatternMatch` objects for each occurrence. - -```tomo -func by_pattern(text:Text, pattern:Pat -> func(->PatternMatch?)) -``` - -- `text`: The text to search. -- `pattern`: The pattern to match. - -**Returns:** -An iterator function that yields `PatternMatch` objects one at a time. - -**Example:** -```tomo -text := "one, two, three" -for word in text.by_pattern($Pat"{id}"): - say(word.text) -``` - ---- - -### `by_pattern_split` -Returns an iterator function that yields text segments split by a pattern. - -```tomo -func by_pattern_split(text:Text, pattern:Pat -> func(->Text?)) -``` - -- `text`: The text to split. -- `pattern`: The pattern to use as a separator. - -**Returns:** -An iterator function that yields text segments. - -**Example:** -```tomo -text := "one two three" -for word in text.by_pattern_split($Pat"{whitespace}"): - say(word.text) -``` - ---- - -### `each_pattern` -Applies a function to each occurrence of a pattern in the text. - -```tomo -func each_pattern(text:Text, pattern:Pat, fn:func(m:PatternMatch), recursive=yes) -``` - -- `text`: The text to search. -- `pattern`: The pattern to match. -- `fn`: The function to apply to each match. -- `recursive`: If `yes`, applies the function recursively on modified text. - -**Example:** -```tomo -text := "one two three" -text.each_pattern($Pat"{id}", func(m:PatternMatch): - say(m.txt) -) -``` - ---- - -### `find_patterns` -Finds all occurrences of a pattern in a text and returns them as `PatternMatch` objects. - -```tomo -func find_patterns(text:Text, pattern:Pat -> [PatternMatch]) -``` - -- `text`: The text to search. -- `pattern`: The pattern to match. - -**Returns:** -A list of `PatternMatch` objects. - -**Example:** -```tomo -text := "one! two three!" ->> text.find_patterns($Pat"{id}!") -= [PatternMatch(text="one!", index=1, captures=["one"]), PatternMatch(text="three!", index=10, captures=["three"])] -``` - ---- - -### `has_pattern` -Checks whether a given pattern appears in the text. - -```tomo -func has_pattern(text:Text, pattern:Pat -> Bool) -``` - -- `text`: The text to search. -- `pattern`: The pattern to check for. - -**Returns:** -`yes` if a match is found, otherwise `no`. - -**Example:** -```tomo -text := "...okay..." ->> text.has_pattern($Pat"{id}") -= yes -``` - ---- - -### `map_pattern` -Transforms matches of a pattern using a mapping function. - -```tomo -func map_pattern(text:Text, pattern:Pat, fn:func(m:PatternMatch -> Text), recursive=yes -> Text) -``` - -- `text`: The text to modify. -- `pattern`: The pattern to match. -- `fn`: A function that transforms matches. -- `recursive`: If `yes`, applies transformations recursively. - -**Returns:** -A new text with the transformed matches. - -**Example:** -```tomo -text := "I have #apples and #oranges and #plums" -fruits := {"apples"=4, "oranges"=5} ->> text.map_pattern($Pat'#{id}', func(match:PatternMatch): - fruit := match.captures[1] - "$(fruits[fruit] or 0) $fruit" -) -= "I have 4 apples and 5 oranges and 0 plums" -``` - ---- - -### `matches_pattern` -Returns whether or not text matches a pattern completely. - -```tomo -func matches_pattern(text:Text, pattern:Pat -> Bool) -``` - -- `text`: The text to match against. -- `pattern`: The pattern to match. - -**Returns:** -`yes` if the whole text matches the pattern, otherwise `no`. - -**Example:** -```tomo ->> "Hello!!!".matches_pattern($Pat"{id}") -= no ->> "Hello".matches_pattern($Pat"{id}") -= yes -``` - ---- - -### `pattern_captures` -Returns a list of pattern captures for the given pattern. - -```tomo -func pattern_captures(text:Text, pattern:Pat -> [Text]?) -``` - -- `text`: The text to match against. -- `pattern`: The pattern to match. - -**Returns:** -An optional list of matched pattern captures. Returns `none` if the text does -not match the pattern. - -**Example:** -```tomo ->> "123 boxes".pattern_captures($Pat"{int} {id}") -= ["123", "boxes"]? ->> "xxx".pattern_captures($Pat"{int} {id}") -= none -``` - ---- - -### `replace_pattern` -Replaces occurrences of a pattern with a replacement text, supporting backreferences. - -```tomo -func replace_pattern(text:Text, pattern:Pat, replacement:Text, backref="@", recursive=yes -> Text) -``` - -- `text`: The text to modify. -- `pattern`: The pattern to match. -- `replacement`: The text to replace matches with. -- `backref`: The symbol for backreferences in the replacement. -- `recursive`: If `yes`, applies replacements recursively. - -**Returns:** -A new text with replacements applied. - -**Example:** -```tomo ->> "I have 123 apples and 456 oranges".replace_pattern($Pat"{int}", "some") -= "I have some apples and some oranges" - ->> "I have 123 apples and 456 oranges".replace_pattern($Pat"{int}", "(@1)") -= "I have (123) apples and (456) oranges" - ->> "I have 123 apples and 456 oranges".replace_pattern($Pat"{int}", "(?1)", backref="?") -= "I have (123) apples and (456) oranges" - ->> "bad(fn(), bad(notbad))".replace_pattern($Pat"bad(?)", "good(@1)") -= "good(fn(), good(notbad))" - ->> "bad(fn(), bad(notbad))".replace_pattern($Pat"bad(?)", "good(@1)", recursive=no) -= "good(fn(), bad(notbad))" -``` - ---- - -### `split_pattern` -Splits a text into segments using a pattern as the delimiter. - -```tomo -func split_pattern(text:Text, pattern:Pat -> [Text]) -``` - -- `text`: The text to split. -- `pattern`: The pattern to use as a separator. - -**Returns:** -A list of text segments. - -**Example:** -```tomo ->> "one two three".split_pattern($Pat"{whitespace}") -= ["one", "two", "three"] -``` - ---- - -### `translate_patterns` -Replaces multiple patterns using a mapping of patterns to replacement texts. - -```tomo -func translate_patterns(text:Text, replacements:{Pat,Text}, backref="@", recursive=yes -> Text) -``` - -- `text`: The text to modify. -- `replacements`: A table mapping patterns to their replacements. -- `backref`: The symbol for backreferences in replacements. -- `recursive`: If `yes`, applies replacements recursively. - -**Returns:** -A new text with all specified replacements applied. - -**Example:** -```tomo ->> text := "foo(x, baz(1))" ->> text.translate_patterns({ - $Pat"{id}(?)"="call(fn('@1'), @2)", - $Pat"{id}"="var('@1')", - $Pat"{int}"="int(@1)", -}) -= "call(fn('foo'), var('x'), call(fn('baz'), int(1)))" -``` - ---- - -### `trim_pattern` -Removes matching patterns from the beginning and/or end of a text. - -```tomo -func trim_pattern(text:Text, pattern=$Pat"{space}", left=yes, right=yes -> Text) -``` - -- `text`: The text to trim. -- `pattern`: The pattern to trim (defaults to whitespace). -- `left`: If `yes`, trims from the beginning. -- `right`: If `yes`, trims from the end. - -**Returns:** -The trimmed text. - -**Example:** -```tomo ->> "123abc456".trim_pattern($Pat"{digit}") -= "abc" -``` diff --git a/lib/patterns/_test.tm b/lib/patterns/_test.tm deleted file mode 100644 index 26c23628..00000000 --- a/lib/patterns/_test.tm +++ /dev/null @@ -1,256 +0,0 @@ -use patterns_v1.0 - -func main() - amelie := "Am\{UE9}lie" - amelie2 := "Am\{U65}\{U301}lie" - >> "Hello".replace_pattern($Pat/e/, "X") - = "HXllo" - - >> "Hello".has_pattern($Pat/l/) - = yes - >> "Hello".has_pattern($Pat/l{end}/) - = no - >> "Hello".has_pattern($Pat/{start}l/) - = no - - >> "Hello".has_pattern($Pat/o/) - = yes - >> "Hello".has_pattern($Pat/o{end}/) - = yes - >> "Hello".has_pattern($Pat/{start}o/) - = no - - >> "Hello".has_pattern($Pat/H/) - = yes - >> "Hello".has_pattern($Pat/H{end}/) - = no - >> "Hello".has_pattern($Pat/{start}H/) - = yes - - >> "Hello".replace_pattern($Pat/l/, "") - = "Heo" - >> "xxxx".replace_pattern($Pat/x/, "") - = "" - >> "xxxx".replace_pattern($Pat/y/, "") - = "xxxx" - >> "One two three four five six".replace_pattern($Pat/e /, "") - = "Ontwo threfour fivsix" - - >> " one ".replace_pattern($Pat/{start}{space}/, "") - = "one " - >> " one ".replace_pattern($Pat/{space}{end}/, "") - = " one" - - >> amelie.has_pattern($Pat/$amelie2/) - = yes - - >> "one two three".replace_pattern($Pat/{alpha}/, "") - = " " - >> "one two three".replace_pattern($Pat/{alpha}/, "word") - = "word word word" - - say("Test splitting and joining text:") - - >> "one two three".split_pattern($Pat/ /) - = ["one", "two", "three"] - - >> "one,two,three,".split_pattern($Pat/,/) - = ["one", "two", "three", ""] - - >> "one two three".split_pattern($Pat/{space}/) - = ["one", "two", "three"] - - >> "abc".split_pattern($Pat//) - = ["a", "b", "c"] - - >> ", ".join(["one", "two", "three"]) - = "one, two, three" - - >> "".join(["one", "two", "three"]) - = "onetwothree" - - >> "+".join(["one"]) - = "one" - - >> "+".join([]) - = "" - - say("Test text.find_patterns()") - >> " #one #two #three ".find_patterns($Pat/#{alpha}/) - = [PatternMatch(text="#one", index=2, captures=["one"]), PatternMatch(text="#two", index=8, captures=["two"]), PatternMatch(text="#three", index=13, captures=["three"])] - - >> " #one #two #three ".find_patterns($Pat/#{!space}/) - = [PatternMatch(text="#one", index=2, captures=["one"]), PatternMatch(text="#two", index=8, captures=["two"]), PatternMatch(text="#three", index=13, captures=["three"])] - - >> " ".find_patterns($Pat/{alpha}/) - = [] - - >> " foo(baz(), 1) doop() ".find_patterns($Pat/{id}(?)/) - = [PatternMatch(text="foo(baz(), 1)", index=2, captures=["foo", "baz(), 1"]), PatternMatch(text="doop()", index=17, captures=["doop", ""])] - - >> "".find_patterns($Pat'') - = [] - - >> "Hello".find_patterns($Pat'') - = [] - - say("Test text slicing:") - >> "abcdef".slice() - = "abcdef" - >> "abcdef".slice(from=3) - = "cdef" - >> "abcdef".slice(to=-2) - = "abcde" - >> "abcdef".slice(from=2, to=4) - = "bcd" - >> "abcdef".slice(from=5, to=1) - = "" - - >> house := "家" - = "家" - >> house.length - = 1 - >> house.codepoint_names() - = ["CJK Unified Ideographs-5BB6"] - >> house.utf32_codepoints() - = [23478] - - >> "🐧".codepoint_names() - = ["PENGUIN"] - - >> Text.from_codepoint_names(["not a valid name here buddy"]) - = none - - >> "one two; three four".find_patterns($Pat/; {..}/) - = [PatternMatch(text="; three four", index=8, captures=["three four"])] - - malicious := "{xxx}" - >> $Pat/$malicious/ - = $Pat/{1{}xxx}/ - - >> "Hello".replace_pattern($Pat/{lower}/, "(@0)") - = "H(ello)" - - >> " foo(xyz) foo(yyy) foo(z()) ".replace_pattern($Pat/foo(?)/, "baz(@1)") - = " baz(xyz) baz(yyy) baz(z()) " - - >> "<tag>".translate_patterns({$Pat/</="<", $Pat/>/=">"}) - = "<tag>" - - >> " BAD(x, fn(y), BAD(z), w) ".replace_pattern($Pat/BAD(?)/, "good(@1)", recursive=yes) - = " good(x, fn(y), good(z), w) " - - >> " BAD(x, fn(y), BAD(z), w) ".replace_pattern($Pat/BAD(?)/, "good(@1)", recursive=no) - = " good(x, fn(y), BAD(z), w) " - - >> "Hello".matches_pattern($Pat/{id}/) - = yes - >> "Hello".matches_pattern($Pat/{lower}/) - = no - >> "Hello".matches_pattern($Pat/{upper}/) - = no - >> "Hello...".matches_pattern($Pat/{id}/) - = no - - >> "hello world".map_pattern($Pat/world/, func(m:PatternMatch) m.text.upper()) - = "hello WORLD" - - >> "Abc".repeat(3) - = "AbcAbcAbc" - - >> " abc def ".trim_pattern() - = "abc def" - >> " abc123def ".trim_pattern($Pat/{!digit}/) - = "123" - >> " abc123def ".trim_pattern($Pat/{!digit}/, left=no) - = " abc123" - >> " abc123def ".trim_pattern($Pat/{!digit}/, right=no) - = "123def " - # Only trim single whole matches that bookend the text: - >> "AbcAbcxxxxxxxxAbcAbc".trim_pattern($Pat/Abc/) - = "AbcxxxxxxxxAbc" - - >> "A=B=C=D".replace_pattern($Pat/{..}={..}/, "1:(@1) 2:(@2)") - = "1:(A) 2:(B=C=D)" - - >> "abcde".starts_with("ab") - = yes - >> "abcde".starts_with("bc") - = no - - >> "abcde".ends_with("de") - = yes - >> "abcde".starts_with("cd") - = no - - >> ("hello" ++ " " ++ "Amélie").reversed() - = "eilémA olleh" - - do - say("Testing concatenation-stability:") - ab := Text.from_codepoint_names(["LATIN SMALL LETTER E", "COMBINING VERTICAL LINE BELOW"])! - >> ab.codepoint_names() - = ["LATIN SMALL LETTER E", "COMBINING VERTICAL LINE BELOW"] - >> ab.length - = 1 - - a := Text.from_codepoint_names(["LATIN SMALL LETTER E"])! - b := Text.from_codepoint_names(["COMBINING VERTICAL LINE BELOW"])! - >> (a++b).codepoint_names() - = ["LATIN SMALL LETTER E", "COMBINING VERTICAL LINE BELOW"] - >> (a++b) == ab - = yes - >> (a++b).length - = 1 - - - do - concat := "e" ++ Text.from_codepoints([Int32(0x300)]) - >> concat.length - = 1 - - concat2 := concat ++ Text.from_codepoints([Int32(0x302)]) - >> concat2.length - = 1 - - concat3 := concat2 ++ Text.from_codepoints([Int32(0x303)]) - >> concat3.length - = 1 - - final := Text.from_codepoints([Int32(0x65), Int32(0x300), Int32(0x302), Int32(0x303)]) - >> final.length - = 1 - >> concat3 == final - = yes - - concat4 := Text.from_codepoints([Int32(0x65), Int32(0x300)]) ++ Text.from_codepoints([Int32(0x302), Int32(0x303)]) - >> concat4.length - = 1 - >> concat4 == final - = yes - - >> "x".left_pad(5) - = " x" - >> "x".right_pad(5) - = "x " - >> "x".middle_pad(5) - = " x " - >> "1234".left_pad(8, "XYZ") - = "XYZX1234" - >> "1234".right_pad(8, "XYZ") - = "1234XYZX" - >> "1234".middle_pad(9, "XYZ") - = "XY1234XYZ" - - >> amelie.width() - = 6 - cowboy := "🤠" - >> cowboy.width() - = 2 - >> cowboy.left_pad(4) - = " 🤠" - >> cowboy.right_pad(4) - = "🤠 " - >> cowboy.middle_pad(4) - = " 🤠 " - diff --git a/lib/patterns/match_type.h b/lib/patterns/match_type.h deleted file mode 100644 index f85f1862..00000000 --- a/lib/patterns/match_type.h +++ /dev/null @@ -1,9 +0,0 @@ -// A datatype used for pattern match results - -#pragma once - -typedef struct { - Text_t text; - Int_t index; - List_t captures; -} XMatch; diff --git a/lib/patterns/patterns.c b/lib/patterns/patterns.c deleted file mode 100644 index 27799c40..00000000 --- a/lib/patterns/patterns.c +++ /dev/null @@ -1,1212 +0,0 @@ -// Logic for text pattern matching - -#include <ctype.h> -#include <string.h> -#include <strings.h> -#include <sys/param.h> -#include <unictype.h> -#include <uniname.h> -#include <unistring/version.h> - -#define MAX_BACKREFS 100 - -typedef struct { - Text_t text; - Int_t index; - List_t captures; -} PatternMatch; - -typedef struct { - Text_t text; - Int_t index; - List_t captures; - bool is_none : 1; -} OptionalPatternMatch; - -#define NONE_MATCH ((OptionalPatternMatch){.is_none = true}) - -typedef struct { - int64_t index, length; - bool occupied, recursive; -} capture_t; - -typedef struct { - enum { PAT_START, PAT_END, PAT_ANY, PAT_GRAPHEME, PAT_PROPERTY, PAT_QUOTE, PAT_PAIR, PAT_FUNCTION } tag; - bool negated, non_capturing; - int64_t min, max; - union { - int32_t grapheme; - uc_property_t property; - int64_t (*fn)(TextIter_t *, int64_t); - int32_t quote_graphemes[2]; - int32_t pair_graphemes[2]; - }; -} pat_t; - -static Text_t replace_list(Text_t text, List_t replacements, Text_t backref_pat, bool recursive); - -static INLINE void skip_whitespace(TextIter_t *state, int64_t *i) { - while (*i < state->stack[0].text.length) { - int32_t grapheme = Text$get_grapheme_fast(state, *i); - if (grapheme > 0 && !uc_is_property_white_space((ucs4_t)grapheme)) return; - *i += 1; - } -} - -static INLINE bool match_grapheme(TextIter_t *state, int64_t *i, int32_t grapheme) { - if (*i < state->stack[0].text.length && Text$get_grapheme_fast(state, *i) == grapheme) { - *i += 1; - return true; - } - return false; -} - -static INLINE bool match_str(TextIter_t *state, int64_t *i, const char *str) { - int64_t matched = 0; - while (matched[str]) { - if (*i + matched >= state->stack[0].text.length || Text$get_grapheme_fast(state, *i + matched) != str[matched]) - return false; - matched += 1; - } - *i += matched; - return true; -} - -static int64_t parse_int(TextIter_t *state, int64_t *i) { - int64_t value = 0; - for (;; *i += 1) { - uint32_t grapheme = Text$get_main_grapheme_fast(state, *i); - int digit = uc_digit_value(grapheme); - if (digit < 0) break; - if (value >= INT64_MAX / 10) break; - value = 10 * value + digit; - } - return value; -} - -static const char *get_property_name(TextIter_t *state, int64_t *i) { - skip_whitespace(state, i); - char *name = GC_MALLOC_ATOMIC(UNINAME_MAX); - char *dest = name; - while (*i < state->stack[0].text.length) { - int32_t grapheme = Text$get_grapheme_fast(state, *i); - if (!(grapheme & ~0xFF) && (isalnum(grapheme) || grapheme == ' ' || grapheme == '_' || grapheme == '-')) { - *dest = (char)grapheme; - ++dest; - if (dest >= name + UNINAME_MAX - 1) break; - } else { - break; - } - *i += 1; - } - - while (dest > name && dest[-1] == ' ') - *(dest--) = '\0'; - - if (dest == name) return NULL; - *dest = '\0'; - return name; -} - -#define EAT1(state, index, cond) \ - ({ \ - int32_t grapheme = Text$get_grapheme_fast(state, index); \ - bool success = (cond); \ - if (success) index += 1; \ - success; \ - }) - -#define EAT2(state, index, cond1, cond2) \ - ({ \ - int32_t grapheme = Text$get_grapheme_fast(state, index); \ - bool success = (cond1); \ - if (success) { \ - grapheme = Text$get_grapheme_fast(state, index + 1); \ - success = (cond2); \ - if (success) index += 2; \ - } \ - success; \ - }) - -#define EAT_MANY(state, index, cond) \ - ({ \ - int64_t _n = 0; \ - while (EAT1(state, index, cond)) { \ - _n += 1; \ - } \ - _n; \ - }) - -static int64_t match_email(TextIter_t *state, int64_t index) { - // email = local "@" domain - // local = 1-64 ([a-zA-Z0-9!#$%&‘*+–/=?^_`.{|}~] | non-ascii) - // domain = dns-label ("." dns-label)* - // dns-label = 1-63 ([a-zA-Z0-9-] | non-ascii) - - if (index > 0) { - uint32_t prev_codepoint = Text$get_main_grapheme_fast(state, index - 1); - if (uc_is_property_alphabetic(prev_codepoint)) return -1; - } - - int64_t start_index = index; - - // Local part: - int64_t local_len = 0; - static const char *allowed_local = "!#$%&‘*+–/=?^_`.{|}~"; - while (EAT1(state, index, (grapheme & ~0x7F) || isalnum((char)grapheme) || strchr(allowed_local, (char)grapheme))) { - local_len += 1; - if (local_len > 64) return -1; - } - - if (!EAT1(state, index, grapheme == '@')) return -1; - - // Host - int64_t host_len = 0; - do { - int64_t label_len = 0; - while (EAT1(state, index, (grapheme & ~0x7F) || isalnum((char)grapheme) || grapheme == '-')) { - label_len += 1; - if (label_len > 63) return -1; - } - - if (label_len == 0) return -1; - - host_len += label_len; - if (host_len > 255) return -1; - host_len += 1; - } while (EAT1(state, index, grapheme == '.')); - - return index - start_index; -} - -static int64_t match_ipv6(TextIter_t *state, int64_t index) { - if (index > 0) { - int32_t prev_codepoint = Text$get_grapheme_fast(state, index - 1); - if ((prev_codepoint & ~0x7F) && (isxdigit(prev_codepoint) || prev_codepoint == ':')) return -1; - } - int64_t start_index = index; - const int NUM_CLUSTERS = 8; - bool double_colon_used = false; - for (int cluster = 0; cluster < NUM_CLUSTERS; cluster++) { - for (int digits = 0; digits < 4; digits++) { - if (!EAT1(state, index, ~(grapheme & ~0x7F) && isxdigit((char)grapheme))) break; - } - if (EAT1(state, index, ~(grapheme & ~0x7F) && isxdigit((char)grapheme))) return -1; // Too many digits - - if (cluster == NUM_CLUSTERS - 1) { - break; - } else if (!EAT1(state, index, grapheme == ':')) { - if (double_colon_used) break; - return -1; - } - - if (EAT1(state, index, grapheme == ':')) { - if (double_colon_used) return -1; - double_colon_used = true; - } - } - return index - start_index; -} - -static int64_t match_ipv4(TextIter_t *state, int64_t index) { - if (index > 0) { - int32_t prev_codepoint = Text$get_grapheme_fast(state, index - 1); - if ((prev_codepoint & ~0x7F) && (isdigit(prev_codepoint) || prev_codepoint == '.')) return -1; - } - int64_t start_index = index; - - const int NUM_CLUSTERS = 4; - for (int cluster = 0; cluster < NUM_CLUSTERS; cluster++) { - for (int digits = 0; digits < 3; digits++) { - if (!EAT1(state, index, ~(grapheme & ~0x7F) && isdigit((char)grapheme))) { - if (digits == 0) return -1; - break; - } - } - - if (EAT1(state, index, ~(grapheme & ~0x7F) && isdigit((char)grapheme))) return -1; // Too many digits - - if (cluster == NUM_CLUSTERS - 1) break; - else if (!EAT1(state, index, grapheme == '.')) return -1; - } - return (index - start_index); -} - -static int64_t match_ip(TextIter_t *state, int64_t index) { - int64_t len = match_ipv6(state, index); - if (len >= 0) return len; - len = match_ipv4(state, index); - return (len >= 0) ? len : -1; -} - -static int64_t match_host(TextIter_t *state, int64_t index) { - int64_t ip_len = match_ip(state, index); - if (ip_len > 0) return ip_len; - - int64_t start_index = index; - if (match_grapheme(state, &index, '[')) { - ip_len = match_ip(state, index); - if (ip_len <= 0) return -1; - index += ip_len; - if (match_grapheme(state, &index, ']')) return (index - start_index); - return -1; - } - - if (!EAT1(state, index, isalpha(grapheme))) return -1; - - static const char *non_host_chars = "/#?:@ \t\r\n<>[]{}\\^|\"`"; - EAT_MANY(state, index, (grapheme & ~0x7F) || !strchr(non_host_chars, (char)grapheme)); - return (index - start_index); -} - -static int64_t match_authority(TextIter_t *state, int64_t index) { - int64_t authority_start = index; - static const char *non_segment_chars = "/#?:@ \t\r\n<>[]{}\\^|\"`."; - - // Optional user@ prefix: - int64_t username_len = EAT_MANY(state, index, (grapheme & ~0x7F) || !strchr(non_segment_chars, (char)grapheme)); - if (username_len < 1 || !EAT1(state, index, grapheme == '@')) index = authority_start; // No user@ part - - // Host: - int64_t host_len = match_host(state, index); - if (host_len <= 0) return -1; - index += host_len; - - // Port: - if (EAT1(state, index, grapheme == ':')) { - if (EAT_MANY(state, index, !(grapheme & ~0x7F) && isdigit(grapheme)) == 0) return -1; - } - return (index - authority_start); -} - -static int64_t match_uri(TextIter_t *state, int64_t index) { - // URI = scheme ":" ["//" authority] path ["?" query] ["#" fragment] - // scheme = [a-zA-Z] [a-zA-Z0-9+.-] - // authority = [userinfo "@"] host [":" port] - - if (index > 0) { - // Don't match if we're not at a word edge: - uint32_t prev_codepoint = Text$get_main_grapheme_fast(state, index - 1); - if (uc_is_property_alphabetic(prev_codepoint)) return -1; - } - - int64_t start_index = index; - - // Scheme: - if (!EAT1(state, index, isalpha(grapheme))) return -1; - EAT_MANY(state, index, - !(grapheme & ~0x7F) && (isalnum(grapheme) || grapheme == '+' || grapheme == '.' || grapheme == '-')); - if (!match_grapheme(state, &index, ':')) return -1; - - // Authority: - int64_t authority_len; - if (match_str(state, &index, "//")) { - authority_len = match_authority(state, index); - if (authority_len > 0) index += authority_len; - } else { - authority_len = 0; - } - - // Path: - int64_t path_start = index; - if (EAT1(state, index, grapheme == '/') || authority_len <= 0) { - static const char *non_path = " \"#?<>[]{}\\^`|"; - EAT_MANY(state, index, (grapheme & ~0x7F) || !strchr(non_path, (char)grapheme)); - - if (EAT1(state, index, grapheme == '?')) { // Query - static const char *non_query = " \"#<>[]{}\\^`|"; - EAT_MANY(state, index, (grapheme & ~0x7F) || !strchr(non_query, (char)grapheme)); - } - - if (EAT1(state, index, grapheme == '#')) { // Fragment - static const char *non_fragment = " \"#<>[]{}\\^`|"; - EAT_MANY(state, index, (grapheme & ~0x7F) || !strchr(non_fragment, (char)grapheme)); - } - } - - if (authority_len <= 0 && index == path_start) return -1; - - return index - start_index; -} - -static int64_t match_url(TextIter_t *state, int64_t index) { - int64_t lookahead = index; - if (!(match_str(state, &lookahead, "https:") || match_str(state, &lookahead, "http:") - || match_str(state, &lookahead, "ftp:") || match_str(state, &lookahead, "wss:") - || match_str(state, &lookahead, "ws:"))) - return -1; - - return match_uri(state, index); -} - -static int64_t match_id(TextIter_t *state, int64_t index) { - if (!EAT1(state, index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_XID_START))) return -1; - return 1 + EAT_MANY(state, index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_XID_CONTINUE)); -} - -static int64_t match_int(TextIter_t *state, int64_t index) { - int64_t negative = EAT1(state, index, grapheme == '-') ? 1 : 0; - int64_t len = EAT_MANY(state, index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_DECIMAL_DIGIT)); - return len > 0 ? negative + len : -1; -} - -static int64_t match_alphanumeric(TextIter_t *state, int64_t index) { - return EAT1(state, index, uc_is_property_alphabetic((ucs4_t)grapheme) || uc_is_property_numeric((ucs4_t)grapheme)) - ? 1 - : -1; -} - -static int64_t match_num(TextIter_t *state, int64_t index) { - bool negative = EAT1(state, index, grapheme == '-') ? 1 : 0; - int64_t pre_decimal = EAT_MANY(state, index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_DECIMAL_DIGIT)); - bool decimal = (EAT1(state, index, grapheme == '.') == 1); - int64_t post_decimal = - decimal ? EAT_MANY(state, index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_DECIMAL_DIGIT)) : 0; - if (pre_decimal == 0 && post_decimal == 0) return -1; - return negative + pre_decimal + decimal + post_decimal; -} - -static int64_t match_newline(TextIter_t *state, int64_t index) { - if (index >= state->stack[0].text.length) return -1; - - uint32_t grapheme = index >= state->stack[0].text.length ? 0 : Text$get_main_grapheme_fast(state, index); - if (grapheme == '\n') return 1; - if (grapheme == '\r' && Text$get_grapheme_fast(state, index + 1) == '\n') return 2; - return -1; -} - -static int64_t match_pat(TextIter_t *state, int64_t index, pat_t pat) { - Text_t text = state->stack[0].text; - int32_t grapheme = index >= text.length ? 0 : Text$get_grapheme_fast(state, index); - - switch (pat.tag) { - case PAT_START: { - if (index == 0) return pat.negated ? -1 : 0; - return pat.negated ? 0 : -1; - } - case PAT_END: { - if (index >= text.length) return pat.negated ? -1 : 0; - return pat.negated ? 0 : -1; - } - case PAT_ANY: { - assert(!pat.negated); - return (index < text.length) ? 1 : -1; - } - case PAT_GRAPHEME: { - if (index >= text.length) return -1; - else if (grapheme == pat.grapheme) return pat.negated ? -1 : 1; - return pat.negated ? 1 : -1; - } - case PAT_PROPERTY: { - if (index >= text.length) return -1; - else if (uc_is_property((ucs4_t)grapheme, pat.property)) return pat.negated ? -1 : 1; - return pat.negated ? 1 : -1; - } - case PAT_PAIR: { - // Nested punctuation: (?), [?], etc - if (index >= text.length) return -1; - - int32_t open = pat.pair_graphemes[0]; - if (grapheme != open) return pat.negated ? 1 : -1; - - int32_t close = pat.pair_graphemes[1]; - int64_t depth = 1; - int64_t match_len = 1; - for (; depth > 0; match_len++) { - if (index + match_len >= text.length) return pat.negated ? 1 : -1; - - int32_t c = Text$get_grapheme_fast(state, index + match_len); - if (c == open) depth += 1; - else if (c == close) depth -= 1; - } - return pat.negated ? -1 : match_len; - } - case PAT_QUOTE: { - // Nested quotes: "?", '?', etc - if (index >= text.length) return -1; - - int32_t open = pat.quote_graphemes[0]; - if (grapheme != open) return pat.negated ? 1 : -1; - - int32_t close = pat.quote_graphemes[1]; - for (int64_t i = index + 1; i < text.length; i++) { - int32_t c = Text$get_grapheme_fast(state, i); - if (c == close) { - return pat.negated ? -1 : (i - index) + 1; - } else if (c == '\\' && index + 1 < text.length) { - i += 1; // Skip ahead an extra step - } - } - return pat.negated ? 1 : -1; - } - case PAT_FUNCTION: { - int64_t match_len = pat.fn(state, index); - if (match_len >= 0) return pat.negated ? -1 : match_len; - return pat.negated ? 1 : -1; - } - default: errx(1, "Invalid pattern"); - } - errx(1, "Unreachable"); - return 0; -} - -static pat_t parse_next_pat(TextIter_t *state, int64_t *index) { - if (EAT2(state, *index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_QUOTATION_MARK), grapheme == '?')) { - // Quotations: "?", '?', etc - int32_t open = Text$get_grapheme_fast(state, *index - 2); - int32_t close = open; - uc_mirror_char((ucs4_t)open, (ucs4_t *)&close); - if (!match_grapheme(state, index, close)) fail("Pattern's closing quote is missing: ", state->stack[0].text); - - return (pat_t){ - .tag = PAT_QUOTE, - .min = 1, - .max = 1, - .quote_graphemes = {open, close}, - }; - } else if (EAT2(state, *index, uc_is_property((ucs4_t)grapheme, UC_PROPERTY_PAIRED_PUNCTUATION), grapheme == '?')) { - // Nested punctuation: (?), [?], etc - int32_t open = Text$get_grapheme_fast(state, *index - 2); - int32_t close = open; - uc_mirror_char((ucs4_t)open, (ucs4_t *)&close); - if (!match_grapheme(state, index, close)) fail("Pattern's closing brace is missing: ", state->stack[0].text); - - return (pat_t){ - .tag = PAT_PAIR, - .min = 1, - .max = 1, - .pair_graphemes = {open, close}, - }; - } else if (EAT1(state, *index, grapheme == '{')) { // named patterns {id}, {2-3 hex}, etc. - skip_whitespace(state, index); - int64_t min, max; - if (uc_is_digit((ucs4_t)Text$get_grapheme_fast(state, *index))) { - min = parse_int(state, index); - skip_whitespace(state, index); - if (match_grapheme(state, index, '+')) { - max = INT64_MAX; - } else if (match_grapheme(state, index, '-')) { - max = parse_int(state, index); - } else { - max = min; - } - if (min > max) fail("Minimum repetitions (", min, ") is less than the maximum (", max, ")"); - } else { - min = -1, max = -1; - } - - skip_whitespace(state, index); - - bool negated = match_grapheme(state, index, '!'); -#define PAT(_tag, ...) ((pat_t){.min = min, .max = max, .negated = negated, .tag = _tag, __VA_ARGS__}) - const char *prop_name; - if (match_str(state, index, "..")) prop_name = ".."; - else prop_name = get_property_name(state, index); - - if (!prop_name) { - // Literal character, e.g. {1?} - skip_whitespace(state, index); - int32_t grapheme = Text$get_grapheme_fast(state, (*index)++); - if (!match_grapheme(state, index, '}')) fail("Missing closing '}' in pattern: ", state->stack[0].text); - return PAT(PAT_GRAPHEME, .grapheme = grapheme); - } else if (strlen(prop_name) == 1) { - // Single letter names: {1+ A} - skip_whitespace(state, index); - if (!match_grapheme(state, index, '}')) fail("Missing closing '}' in pattern: ", state->stack[0].text); - return PAT(PAT_GRAPHEME, .grapheme = prop_name[0]); - } - - skip_whitespace(state, index); - if (!match_grapheme(state, index, '}')) fail("Missing closing '}' in pattern: ", state->stack[0].text); - - switch (tolower(prop_name[0])) { - case '.': - if (prop_name[1] == '.') { - if (negated) return ((pat_t){.tag = PAT_END, .min = min, .max = max, .non_capturing = true}); - else return PAT(PAT_ANY); - } - break; - case 'a': - if (strcasecmp(prop_name, "authority") == 0) { - return PAT(PAT_FUNCTION, .fn = match_authority); - } else if (strcasecmp(prop_name, "alphanum") == 0 || strcasecmp(prop_name, "anum") == 0 - || strcasecmp(prop_name, "alphanumeric") == 0) { - return PAT(PAT_FUNCTION, .fn = match_alphanumeric); - } - break; - case 'c': - if (strcasecmp(prop_name, "crlf") == 0) return PAT(PAT_FUNCTION, .fn = match_newline); - break; - case 'd': - if (strcasecmp(prop_name, "digit") == 0) { - return PAT(PAT_PROPERTY, .property = UC_PROPERTY_DECIMAL_DIGIT); - } - break; - case 'e': - if (strcasecmp(prop_name, "end") == 0) { - return PAT(PAT_END, .non_capturing = !negated); - } else if (strcasecmp(prop_name, "email") == 0) { - return PAT(PAT_FUNCTION, .fn = match_email); - } -#if _LIBUNISTRING_VERSION >= 0x0100000 - else if (strcasecmp(prop_name, "emoji") == 0) { - return PAT(PAT_PROPERTY, .property = UC_PROPERTY_EMOJI); - } -#endif - break; - case 'h': - if (strcasecmp(prop_name, "host") == 0) { - return PAT(PAT_FUNCTION, .fn = match_host); - } - break; - case 'i': - if (strcasecmp(prop_name, "id") == 0) { - return PAT(PAT_FUNCTION, .fn = match_id); - } else if (strcasecmp(prop_name, "int") == 0) { - return PAT(PAT_FUNCTION, .fn = match_int); - } else if (strcasecmp(prop_name, "ipv4") == 0) { - return PAT(PAT_FUNCTION, .fn = match_ipv4); - } else if (strcasecmp(prop_name, "ipv6") == 0) { - return PAT(PAT_FUNCTION, .fn = match_ipv6); - } else if (strcasecmp(prop_name, "ip") == 0) { - return PAT(PAT_FUNCTION, .fn = match_ip); - } - break; - case 'n': - if (strcasecmp(prop_name, "nl") == 0 || strcasecmp(prop_name, "newline") == 0) { - return PAT(PAT_FUNCTION, .fn = match_newline); - } else if (strcasecmp(prop_name, "num") == 0) { - return PAT(PAT_FUNCTION, .fn = match_num); - } - break; - case 's': - if (strcasecmp(prop_name, "start") == 0) { - return PAT(PAT_START, .non_capturing = !negated); - } - break; - case 'u': - if (strcasecmp(prop_name, "uri") == 0) { - return PAT(PAT_FUNCTION, .fn = match_uri); - } else if (strcasecmp(prop_name, "url") == 0) { - return PAT(PAT_FUNCTION, .fn = match_url); - } - break; - case 'w': - if (strcasecmp(prop_name, "word") == 0) { - return PAT(PAT_FUNCTION, .fn = match_id); - } else if (strcasecmp(prop_name, "ws") == 0 || strcasecmp(prop_name, "whitespace") == 0) { - return PAT(PAT_PROPERTY, .property = UC_PROPERTY_WHITE_SPACE); - } - break; - default: break; - } - - uc_property_t prop = uc_property_byname(prop_name); - if (uc_property_is_valid(prop)) return PAT(PAT_PROPERTY, .property = prop); - - ucs4_t grapheme = unicode_name_character(prop_name); - if (grapheme == UNINAME_INVALID) fail("Not a valid property or character name: ", prop_name); - return PAT(PAT_GRAPHEME, .grapheme = (int32_t)grapheme); -#undef PAT - } else { - return (pat_t){.tag = PAT_GRAPHEME, - .non_capturing = true, - .min = 1, - .max = 1, - .grapheme = Text$get_grapheme_fast(state, (*index)++)}; - } -} - -static int64_t match(Text_t text, int64_t text_index, Text_t pattern, int64_t pattern_index, capture_t *captures, - int64_t capture_index) { - if (pattern_index >= pattern.length) // End of the pattern - return 0; - - int64_t start_index = text_index; - TextIter_t pattern_state = NEW_TEXT_ITER_STATE(pattern), text_state = NEW_TEXT_ITER_STATE(text); - pat_t pat = parse_next_pat(&pattern_state, &pattern_index); - - if (pat.min == -1 && pat.max == -1) { - if (pat.tag == PAT_ANY && pattern_index >= pattern.length) { - pat.min = pat.max = MAX(1, text.length - text_index); - } else { - pat.min = 1; - pat.max = INT64_MAX; - } - } - - int64_t capture_start = text_index; - int64_t count = 0, capture_len = 0, next_match_len = 0; - - if (pat.tag == PAT_ANY && pattern_index >= pattern.length) { - int64_t remaining = text.length - text_index; - capture_len = remaining >= pat.min ? MIN(remaining, pat.max) : -1; - text_index += capture_len; - goto success; - } - - if (pat.min == 0 && pattern_index < pattern.length) { - next_match_len = - match(text, text_index, pattern, pattern_index, captures, capture_index + (pat.non_capturing ? 0 : 1)); - if (next_match_len >= 0) { - capture_len = 0; - goto success; - } - } - - while (count < pat.max) { - int64_t match_len = match_pat(&text_state, text_index, pat); - if (match_len < 0) break; - capture_len += match_len; - text_index += match_len; - count += 1; - - if (pattern_index < pattern.length) { // More stuff after this - if (count < pat.min) next_match_len = -1; - else - next_match_len = match(text, text_index, pattern, pattern_index, captures, - capture_index + (pat.non_capturing ? 0 : 1)); - } else { - next_match_len = 0; - } - - if (match_len == 0) { - if (next_match_len >= 0) { - // If we're good to go, no need to keep re-matching zero-length - // matches till we hit max: - count = pat.max; - break; - } else { - return -1; - } - } - - if (pattern_index < pattern.length && next_match_len >= 0) break; // Next guy exists and wants to stop here - - if (text_index >= text.length) break; - } - - if (count < pat.min || next_match_len < 0) return -1; - -success: - if (captures && capture_index < MAX_BACKREFS && !pat.non_capturing) { - if (pat.tag == PAT_PAIR || pat.tag == PAT_QUOTE) { - assert(capture_len > 0); - captures[capture_index] = (capture_t){ - .index = capture_start + 1, // Skip leading quote/paren - .length = capture_len - 2, // Skip open/close - .occupied = true, - .recursive = (pat.tag == PAT_PAIR), - }; - } else { - captures[capture_index] = (capture_t){ - .index = capture_start, - .length = capture_len, - .occupied = true, - .recursive = false, - }; - } - } - return (text_index - start_index) + next_match_len; -} - -#undef EAT1 -#undef EAT2 -#undef EAT_MANY - -static int64_t _find(Text_t text, Text_t pattern, int64_t first, int64_t last, int64_t *match_length, - capture_t *captures) { - int32_t first_grapheme = Text$get_grapheme(pattern, 0); - bool find_first = (first_grapheme != '{' && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_QUOTATION_MARK) - && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_PAIRED_PUNCTUATION)); - - TextIter_t text_state = NEW_TEXT_ITER_STATE(text); - for (int64_t i = first; i <= last; i++) { - // Optimization: quickly skip ahead to first char in pattern: - if (find_first) { - while (i < text.length && Text$get_grapheme_fast(&text_state, i) != first_grapheme) - ++i; - } - - int64_t m = match(text, i, pattern, 0, captures, 0); - if (m >= 0) { - if (match_length) *match_length = m; - return i; - } - } - if (match_length) *match_length = -1; - return -1; -} - -static OptionalPatternMatch find(Text_t text, Text_t pattern, Int_t from_index) { - int64_t first = Int64$from_int(from_index, false); - if (first == 0) fail("Invalid index: 0"); - if (first < 0) first = text.length + first + 1; - if (first > text.length || first < 1) return NONE_MATCH; - - capture_t captures[MAX_BACKREFS] = {}; - int64_t len = 0; - int64_t found = _find(text, pattern, first - 1, text.length - 1, &len, captures); - if (found == -1) return NONE_MATCH; - - List_t capture_list = {}; - for (int i = 0; captures[i].occupied; i++) { - Text_t capture = Text$slice(text, I(captures[i].index + 1), I(captures[i].index + captures[i].length)); - List$insert(&capture_list, &capture, I(0), sizeof(Text_t)); - } - return (OptionalPatternMatch){ - .text = Text$slice(text, I(found + 1), I(found + len)), - .index = I(found + 1), - .captures = capture_list, - }; -} - -PUREFUNC static bool Pattern$has(Text_t text, Text_t pattern) { - if (Text$starts_with(pattern, Text("{start}"), &pattern)) { - int64_t m = match(text, 0, pattern, 0, NULL, 0); - return m >= 0; - } else if (Text$ends_with(text, Text("{end}"), NULL)) { - for (int64_t i = text.length - 1; i >= 0; i--) { - int64_t match_len = match(text, i, pattern, 0, NULL, 0); - if (match_len >= 0 && i + match_len == text.length) return true; - } - return false; - } else { - int64_t found = _find(text, pattern, 0, text.length - 1, NULL, NULL); - return (found >= 0); - } -} - -static bool Pattern$matches(Text_t text, Text_t pattern) { - capture_t captures[MAX_BACKREFS] = {}; - int64_t match_len = match(text, 0, pattern, 0, NULL, 0); - return (match_len == text.length); -} - -static bool Pattern$match_at(Text_t text, Text_t pattern, Int_t pos, PatternMatch *dest) { - int64_t start = Int64$from_int(pos, false) - 1; - capture_t captures[MAX_BACKREFS] = {}; - int64_t match_len = match(text, start, pattern, 0, captures, 0); - if (match_len < 0) return false; - - List_t capture_list = {}; - for (int i = 0; captures[i].occupied; i++) { - Text_t capture = Text$slice(text, I(captures[i].index + 1), I(captures[i].index + captures[i].length)); - List$insert(&capture_list, &capture, I(0), sizeof(Text_t)); - } - dest->text = Text$slice(text, I(start + 1), I(start + match_len)); - dest->index = I(start + 1); - dest->captures = capture_list; - return true; -} - -static OptionalList_t Pattern$captures(Text_t text, Text_t pattern) { - capture_t captures[MAX_BACKREFS] = {}; - int64_t match_len = match(text, 0, pattern, 0, captures, 0); - if (match_len != text.length) return NONE_LIST; - - List_t capture_list = {}; - for (int i = 0; captures[i].occupied; i++) { - Text_t capture = Text$slice(text, I(captures[i].index + 1), I(captures[i].index + captures[i].length)); - List$insert(&capture_list, &capture, I(0), sizeof(Text_t)); - } - return capture_list; -} - -static List_t Pattern$find_all(Text_t text, Text_t pattern) { - if (pattern.length == 0) // special case - return (List_t){.length = 0}; - - List_t matches = {}; - for (int64_t i = 1;;) { - OptionalPatternMatch m = find(text, pattern, I(i)); - if (m.is_none) break; - i = Int64$from_int(m.index, false) + m.text.length; - List$insert(&matches, &m, I_small(0), sizeof(PatternMatch)); - } - return matches; -} - -typedef struct { - TextIter_t state; - Int_t i; - Text_t pattern; -} match_iter_state_t; - -static OptionalPatternMatch next_match(match_iter_state_t *state) { - if (Int64$from_int(state->i, false) > state->state.stack[0].text.length) return NONE_MATCH; - - OptionalPatternMatch m = find(state->state.stack[0].text, state->pattern, state->i); - if (m.is_none) // No match - state->i = I(state->state.stack[0].text.length + 1); - else state->i = Int$plus(m.index, I(MAX(1, m.text.length))); - return m; -} - -static Closure_t Pattern$by_match(Text_t text, Text_t pattern) { - return (Closure_t){ - .fn = (void *)next_match, - .userdata = new (match_iter_state_t, .state = NEW_TEXT_ITER_STATE(text), .i = I_small(1), .pattern = pattern), - }; -} - -static Text_t apply_backrefs(Text_t text, List_t recursive_replacements, Text_t replacement, Text_t backref_pat, - capture_t *captures) { - if (backref_pat.length == 0) return replacement; - - int32_t first_grapheme = Text$get_grapheme(backref_pat, 0); - bool find_first = (first_grapheme != '{' && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_QUOTATION_MARK) - && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_PAIRED_PUNCTUATION)); - - Text_t ret = Text(""); - TextIter_t replacement_state = NEW_TEXT_ITER_STATE(replacement); - int64_t nonmatching_pos = 0; - for (int64_t pos = 0; pos < replacement.length;) { - // Optimization: quickly skip ahead to first char in the backref pattern: - if (find_first) { - while (pos < replacement.length && Text$get_grapheme_fast(&replacement_state, pos) != first_grapheme) - ++pos; - } - - int64_t backref_len = match(replacement, pos, backref_pat, 0, NULL, 0); - if (backref_len < 0) { - pos += 1; - continue; - } - - int64_t after_backref = pos + backref_len; - int64_t backref = parse_int(&replacement_state, &after_backref); - if (after_backref == pos + backref_len) { // Not actually a backref if there's no number - pos += 1; - continue; - } - if (backref < 0 || backref > 9) - fail("Invalid backref index: ", backref, " (only 0-", MAX_BACKREFS - 1, " are allowed)"); - backref_len = (after_backref - pos); - - if (Text$get_grapheme_fast(&replacement_state, pos + backref_len) == ';') - backref_len += 1; // skip optional semicolon - - if (!captures[backref].occupied) fail("There is no capture number ", backref, "!"); - - Text_t backref_text = - Text$slice(text, I(captures[backref].index + 1), I(captures[backref].index + captures[backref].length)); - - if (captures[backref].recursive && recursive_replacements.length > 0) - backref_text = replace_list(backref_text, recursive_replacements, backref_pat, true); - - if (pos > nonmatching_pos) { - Text_t before_slice = Text$slice(replacement, I(nonmatching_pos + 1), I(pos)); - ret = Text$concat(ret, before_slice, backref_text); - } else { - ret = Text$concat(ret, backref_text); - } - - pos += backref_len; - nonmatching_pos = pos; - } - if (nonmatching_pos < replacement.length) { - Text_t last_slice = Text$slice(replacement, I(nonmatching_pos + 1), I(replacement.length)); - ret = Text$concat(ret, last_slice); - } - return ret; -} - -static Text_t Pattern$replace(Text_t text, Text_t pattern, Text_t replacement, Text_t backref_pat, bool recursive) { - Text_t ret = EMPTY_TEXT; - - int32_t first_grapheme = Text$get_grapheme(pattern, 0); - bool find_first = (first_grapheme != '{' && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_QUOTATION_MARK) - && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_PAIRED_PUNCTUATION)); - - Text_t entries[2] = {pattern, replacement}; - List_t replacements = { - .data = entries, - .length = 1, - .stride = sizeof(entries), - }; - - TextIter_t text_state = NEW_TEXT_ITER_STATE(text); - int64_t nonmatching_pos = 0; - for (int64_t pos = 0; pos < text.length;) { - // Optimization: quickly skip ahead to first char in pattern: - if (find_first) { - while (pos < text.length && Text$get_grapheme_fast(&text_state, pos) != first_grapheme) - ++pos; - } - - capture_t captures[MAX_BACKREFS] = {}; - int64_t match_len = match(text, pos, pattern, 0, captures, 1); - if (match_len < 0) { - pos += 1; - continue; - } - captures[0] = (capture_t){ - .index = pos, - .length = match_len, - .occupied = true, - .recursive = false, - }; - - Text_t replacement_text = - apply_backrefs(text, recursive ? replacements : (List_t){}, replacement, backref_pat, captures); - if (pos > nonmatching_pos) { - Text_t before_slice = Text$slice(text, I(nonmatching_pos + 1), I(pos)); - ret = Text$concat(ret, before_slice, replacement_text); - } else { - ret = Text$concat(ret, replacement_text); - } - nonmatching_pos = pos + match_len; - pos += MAX(match_len, 1); - } - if (nonmatching_pos < text.length) { - Text_t last_slice = Text$slice(text, I(nonmatching_pos + 1), I(text.length)); - ret = Text$concat(ret, last_slice); - } - return ret; -} - -static Text_t Pattern$trim(Text_t text, Text_t pattern, bool trim_left, bool trim_right) { - int64_t first = 0, last = text.length - 1; - if (trim_left) { - int64_t match_len = match(text, 0, pattern, 0, NULL, 0); - if (match_len > 0) first = match_len; - } - - if (trim_right) { - for (int64_t i = text.length - 1; i >= first; i--) { - int64_t match_len = match(text, i, pattern, 0, NULL, 0); - if (match_len > 0 && i + match_len == text.length) last = i - 1; - } - } - return Text$slice(text, I(first + 1), I(last + 1)); -} - -static Text_t Pattern$map(Text_t text, Text_t pattern, Closure_t fn, bool recursive) { - Text_t ret = EMPTY_TEXT; - - int32_t first_grapheme = Text$get_grapheme(pattern, 0); - bool find_first = (first_grapheme != '{' && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_QUOTATION_MARK) - && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_PAIRED_PUNCTUATION)); - - TextIter_t text_state = NEW_TEXT_ITER_STATE(text); - int64_t nonmatching_pos = 0; - - Text_t (*text_mapper)(PatternMatch, void *) = fn.fn; - for (int64_t pos = 0; pos < text.length; pos++) { - // Optimization: quickly skip ahead to first char in pattern: - if (find_first) { - while (pos < text.length && Text$get_grapheme_fast(&text_state, pos) != first_grapheme) - ++pos; - } - - capture_t captures[MAX_BACKREFS] = {}; - int64_t match_len = match(text, pos, pattern, 0, captures, 0); - if (match_len < 0) continue; - - PatternMatch m = { - .text = Text$slice(text, I(pos + 1), I(pos + match_len)), - .index = I(pos + 1), - .captures = {}, - }; - for (int i = 0; captures[i].occupied; i++) { - Text_t capture = Text$slice(text, I(captures[i].index + 1), I(captures[i].index + captures[i].length)); - if (recursive) capture = Pattern$map(capture, pattern, fn, recursive); - List$insert(&m.captures, &capture, I(0), sizeof(Text_t)); - } - - Text_t replacement = text_mapper(m, fn.userdata); - if (pos > nonmatching_pos) { - Text_t before_slice = Text$slice(text, I(nonmatching_pos + 1), I(pos)); - ret = Text$concat(ret, before_slice, replacement); - } else { - ret = Text$concat(ret, replacement); - } - nonmatching_pos = pos + match_len; - pos += (match_len - 1); - } - if (nonmatching_pos < text.length) { - Text_t last_slice = Text$slice(text, I(nonmatching_pos + 1), I(text.length)); - ret = Text$concat(ret, last_slice); - } - return ret; -} - -static void Pattern$each(Text_t text, Text_t pattern, Closure_t fn, bool recursive) { - int32_t first_grapheme = Text$get_grapheme(pattern, 0); - bool find_first = (first_grapheme != '{' && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_QUOTATION_MARK) - && !uc_is_property((ucs4_t)first_grapheme, UC_PROPERTY_PAIRED_PUNCTUATION)); - - TextIter_t text_state = NEW_TEXT_ITER_STATE(text); - void (*action)(PatternMatch, void *) = fn.fn; - for (int64_t pos = 0; pos < text.length; pos++) { - // Optimization: quickly skip ahead to first char in pattern: - if (find_first) { - while (pos < text.length && Text$get_grapheme_fast(&text_state, pos) != first_grapheme) - ++pos; - } - - capture_t captures[MAX_BACKREFS] = {}; - int64_t match_len = match(text, pos, pattern, 0, captures, 0); - if (match_len < 0) continue; - - PatternMatch m = { - .text = Text$slice(text, I(pos + 1), I(pos + match_len)), - .index = I(pos + 1), - .captures = {}, - }; - for (int i = 0; captures[i].occupied; i++) { - Text_t capture = Text$slice(text, I(captures[i].index + 1), I(captures[i].index + captures[i].length)); - if (recursive) Pattern$each(capture, pattern, fn, recursive); - List$insert(&m.captures, &capture, I(0), sizeof(Text_t)); - } - - action(m, fn.userdata); - pos += (match_len - 1); - } -} - -Text_t replace_list(Text_t text, List_t replacements, Text_t backref_pat, bool recursive) { - if (replacements.length == 0) return text; - - Text_t ret = EMPTY_TEXT; - - int64_t nonmatch_pos = 0; - for (int64_t pos = 0; pos < text.length;) { - // Find the first matching pattern at this position: - for (int64_t i = 0; i < replacements.length; i++) { - Text_t pattern = *(Text_t *)(replacements.data + i * replacements.stride); - capture_t captures[MAX_BACKREFS] = {}; - int64_t len = match(text, pos, pattern, 0, captures, 1); - if (len < 0) continue; - captures[0].index = pos; - captures[0].length = len; - - // If we skipped over some non-matching text before finding a match, insert it here: - if (pos > nonmatch_pos) { - Text_t before_slice = Text$slice(text, I(nonmatch_pos + 1), I(pos)); - ret = Text$concat(ret, before_slice); - } - - // Concatenate the replacement: - Text_t replacement = *(Text_t *)(replacements.data + i * replacements.stride + sizeof(Text_t)); - Text_t replacement_text = - apply_backrefs(text, recursive ? replacements : (List_t){}, replacement, backref_pat, captures); - ret = Text$concat(ret, replacement_text); - pos += MAX(len, 1); - nonmatch_pos = pos; - goto next_pos; - } - - pos += 1; - next_pos: - continue; - } - - if (nonmatch_pos <= text.length) { - Text_t last_slice = Text$slice(text, I(nonmatch_pos + 1), I(text.length)); - ret = Text$concat(ret, last_slice); - } - return ret; -} - -static Text_t Pattern$replace_all(Text_t text, Table_t replacements, Text_t backref_pat, bool recursive) { - return replace_list(text, replacements.entries, backref_pat, recursive); -} - -static List_t Pattern$split(Text_t text, Text_t pattern) { - if (text.length == 0) // special case - return (List_t){.length = 0}; - - if (pattern.length == 0) // special case - return Text$clusters(text); - - List_t chunks = {}; - - int64_t i = 0; - for (;;) { - int64_t len = 0; - int64_t found = _find(text, pattern, i, text.length - 1, &len, NULL); - if (found == i && len == 0) found = _find(text, pattern, i + 1, text.length - 1, &len, NULL); - if (found < 0) break; - Text_t chunk = Text$slice(text, I(i + 1), I(found)); - List$insert(&chunks, &chunk, I_small(0), sizeof(Text_t)); - i = MAX(found + len, i + 1); - } - - Text_t last_chunk = Text$slice(text, I(i + 1), I(text.length)); - List$insert(&chunks, &last_chunk, I_small(0), sizeof(Text_t)); - - return chunks; -} - -typedef struct { - TextIter_t state; - int64_t i; - Text_t pattern; -} split_iter_state_t; - -static OptionalText_t next_split(split_iter_state_t *state) { - Text_t text = state->state.stack[0].text; - if (state->i >= text.length) { - if (state->pattern.length > 0 && state->i == text.length) { // special case - state->i = text.length + 1; - return EMPTY_TEXT; - } - return NONE_TEXT; - } - - if (state->pattern.length == 0) { // special case - Text_t ret = Text$cluster(text, I(state->i + 1)); - state->i += 1; - return ret; - } - - int64_t start = state->i; - int64_t len = 0; - int64_t found = _find(text, state->pattern, start, text.length - 1, &len, NULL); - - if (found == start && len == 0) found = _find(text, state->pattern, start + 1, text.length - 1, &len, NULL); - - if (found >= 0) { - state->i = MAX(found + len, state->i + 1); - return Text$slice(text, I(start + 1), I(found)); - } else { - state->i = state->state.stack[0].text.length + 1; - return Text$slice(text, I(start + 1), I(text.length)); - } -} - -static Closure_t Pattern$by_split(Text_t text, Text_t pattern) { - return (Closure_t){ - .fn = (void *)next_split, - .userdata = new (split_iter_state_t, .state = NEW_TEXT_ITER_STATE(text), .i = 0, .pattern = pattern), - }; -} - -static Text_t Pattern$escape_text(Text_t text) { - // TODO: optimize for spans of non-escaped text - Text_t ret = EMPTY_TEXT; - TextIter_t state = NEW_TEXT_ITER_STATE(text); - for (int64_t i = 0; i < text.length; i++) { - uint32_t g = Text$get_main_grapheme_fast(&state, i); - if (g == '{') { - ret = Text$concat(ret, Text("{1{}")); - } else if (g == '?' || uc_is_property_quotation_mark(g) - || (uc_is_property_paired_punctuation(g) && uc_is_property_left_of_pair(g))) { - ret = Text$concat(ret, Text("{1"), Text$slice(text, I(i + 1), I(i + 1)), Text("}")); - } else { - ret = Text$concat(ret, Text$slice(text, I(i + 1), I(i + 1))); - } - } - return ret; -} - -static Text_t Pattern$as_text(const void *obj, bool colorize, const TypeInfo_t *info) { - (void)info; - if (!obj) return Text("Pattern"); - - Text_t pat = *(Text_t *)obj; - Text_t quote = Pattern$has(pat, Text("/")) && !Pattern$has(pat, Text("|")) ? Text("|") : Text("/"); - return Text$concat(colorize ? Text("\x1b[1m$\033[m") : Text("$"), Text$quoted(pat, colorize, quote)); -} diff --git a/lib/patterns/patterns.tm b/lib/patterns/patterns.tm deleted file mode 100644 index c5444b86..00000000 --- a/lib/patterns/patterns.tm +++ /dev/null @@ -1,66 +0,0 @@ -use ./patterns.c - -struct PatternMatch(text:Text, index:Int, captures:[Text]) - -lang Pat - convert(text:Text -> Pat) - return C_code:Pat(Pattern$escape_text(@text)) - - convert(n:Int -> Pat) - return Pat.from_text("$n") - -extend Text - func matching_pattern(text:Text, pattern:Pat, pos:Int = 1 -> PatternMatch?) - result : PatternMatch - if C_code:Bool(Pattern$match_at(@text, @pattern, @pos, (void*)&@result)) - return result - return none - - func matches_pattern(text:Text, pattern:Pat -> Bool) - return C_code:Bool(Pattern$matches(@text, @pattern)) - - func pattern_captures(text:Text, pattern:Pat -> [Text]?) - return C_code:[Text]?(Pattern$captures(@text, @pattern)) - - func replace_pattern(text:Text, pattern:Pat, replacement:Text, backref="@", recursive=yes -> Text) - return C_code:Text(Pattern$replace(@text, @pattern, @replacement, @backref, @recursive)) - - func translate_patterns(text:Text, replacements:{Pat=Text}, backref="@", recursive=yes -> Text) - return C_code:Text(Pattern$replace_all(@text, @replacements, @backref, @recursive)) - - func has_pattern(text:Text, pattern:Pat -> Bool) - return C_code:Bool(Pattern$has(@text, @pattern)) - - func find_patterns(text:Text, pattern:Pat -> [PatternMatch]) - return C_code:[PatternMatch](Pattern$find_all(@text, @pattern)) - - func by_pattern(text:Text, pattern:Pat -> func(->PatternMatch?)) - return C_code:func(->PatternMatch?)(Pattern$by_match(@text, @pattern)) - - func each_pattern(text:Text, pattern:Pat, fn:func(m:PatternMatch), recursive=yes) - C_code { Pattern$each(@text, @pattern, @fn, @recursive); } - - func map_pattern(text:Text, pattern:Pat, fn:func(m:PatternMatch -> Text), recursive=yes -> Text) - return C_code:Text(Pattern$map(@text, @pattern, @fn, @recursive)) - - func split_pattern(text:Text, pattern:Pat -> [Text]) - return C_code:[Text](Pattern$split(@text, @pattern)) - - func by_pattern_split(text:Text, pattern:Pat -> func(->Text?)) - return C_code:func(->Text?)(Pattern$by_split(@text, @pattern)) - - func trim_pattern(text:Text, pattern=$Pat"{space}", left=yes, right=yes -> Text) - return C_code:Text(Pattern$trim(@text, @pattern, @left, @right)) - -func main() - >> "Hello world".matching_pattern($Pat'{id}') - >> "...Hello world".matching_pattern($Pat'{id}') -# func main(pattern:Pat, input=(/dev/stdin)) -# for line in input.by_line()! -# skip if not line.has_pattern(pattern) -# pos := 1 -# for match in line.by_pattern(pattern) -# say(line.slice(pos, match.index-1), newline=no) -# say("\033[34;1m$(match.text)\033[m", newline=no) -# pos = match.index + match.text.length -# say(line.from(pos), newline=yes) diff --git a/lib/pthreads/CHANGES.md b/lib/pthreads/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/pthreads/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/pthreads/pthreads.tm b/lib/pthreads/pthreads.tm deleted file mode 100644 index 8fca6bd6..00000000 --- a/lib/pthreads/pthreads.tm +++ /dev/null @@ -1,117 +0,0 @@ -# A Posix Threads (pthreads) wrapper -use <pthread.h> - -struct pthread_mutex_t(; extern, opaque) - func new(->@pthread_mutex_t) - return C_code : @pthread_mutex_t( - pthread_mutex_t *mutex = GC_MALLOC(sizeof(pthread_mutex_t)); - pthread_mutex_init(mutex, NULL); - GC_register_finalizer(mutex, (void*)pthread_mutex_destroy, NULL, NULL, NULL); - mutex - ) - - func lock(m:&pthread_mutex_t) - fail("Failed to lock mutex") unless C_code:Int32(pthread_mutex_lock(@m)) == 0 - - func unlock(m:&pthread_mutex_t) - fail("Failed to unlock mutex") unless C_code:Int32(pthread_mutex_unlock(@m)) == 0 - -struct pthread_cond_t(; extern, opaque) - func new(->@pthread_cond_t) - return C_code : @pthread_cond_t( - pthread_cond_t *cond = GC_MALLOC(sizeof(pthread_cond_t)); - pthread_cond_init(cond, NULL); - GC_register_finalizer(cond, (void*)pthread_cond_destroy, NULL, NULL, NULL); - cond - ) - - func wait(cond:&pthread_cond_t, mutex:&pthread_mutex_t) - fail("Failed to wait on condition") unless C_code:Int32(pthread_cond_wait(@cond, @mutex)) == 0 - - func signal(cond:&pthread_cond_t) - fail("Failed to signal pthread_cond_t") unless C_code:Int32(pthread_cond_signal(@cond)) == 0 - - func broadcast(cond:&pthread_cond_t) - fail("Failed to broadcast pthread_cond_t") unless C_code:Int32(pthread_cond_broadcast(@cond)) == 0 - -struct pthread_rwlock_t(; extern, opaque) - func new(->@pthread_rwlock_t) - return C_code : @pthread_rwlock_t ( - pthread_rwlock_t *lock = GC_MALLOC(sizeof(pthread_rwlock_t)); - pthread_rwlock_init(lock, NULL); - GC_register_finalizer(lock, (void*)pthread_rwlock_destroy, NULL, NULL, NULL); - lock - ) - - func read_lock(lock:&pthread_rwlock_t) - C_code { pthread_rwlock_rdlock(@lock); } - - func write_lock(lock:&pthread_rwlock_t) - C_code { pthread_rwlock_wrlock(@lock); } - - func unlock(lock:&pthread_rwlock_t) - C_code { pthread_rwlock_unlock(@lock); } - -struct pthread_t(; extern, opaque) - func new(fn:func() -> @pthread_t) - return C_code:@pthread_t( - pthread_t *thread = GC_MALLOC(sizeof(pthread_t)); - pthread_create(thread, NULL, @fn.fn, @fn.userdata); - thread - ) - - func join(p:pthread_t) C_code { pthread_join(@p, NULL); } - func cancel(p:pthread_t) C_code { pthread_cancel(@p); } - func detatch(p:pthread_t) C_code { pthread_detach(@p); } - -struct IntQueue(_queue:@[Int], _mutex:@pthread_mutex_t, _cond:@pthread_cond_t) - func new(initial:[Int]=[] -> IntQueue) - return IntQueue(@initial, pthread_mutex_t.new(), pthread_cond_t.new()) - - func give(q:IntQueue, n:Int) - do q._mutex.lock() - q._queue.insert(n) - q._mutex.unlock() - q._cond.signal() - - func take(q:IntQueue -> Int) - do q._mutex.lock() - n := q._queue.pop(1) - while not n - q._cond.wait(q._mutex) - n = q._queue.pop(1) - q._mutex.unlock() - return n! - -func main() - jobs := IntQueue.new() - results := IntQueue.new() - - say_mutex := pthread_mutex_t.new() - announce := func(speaker:Text, text:Text) - do say_mutex.lock() - say("\[2][$speaker]\[] $text") - say_mutex.unlock() - - worker := pthread_t.new(func() - say("I'm in the thread!") - repeat - announce("worker", "waiting for job") - job := jobs.take() - result := job * 10 - announce("worker", "Jobbing $job into $result") - results.give(result) - announce("worker", "Signaled $result") - ) - - for i in 10 - announce("boss", "Pushing job $i") - jobs.give(i) - announce("boss", "Gave job $i") - - for i in 10 - announce("boss", "Getting result...") - result := results.take() - announce("boss", "Got result $result") - - >> worker.cancel() diff --git a/lib/random/CHANGES.md b/lib/random/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/random/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/random/README.md b/lib/random/README.md deleted file mode 100644 index 183b9d0b..00000000 --- a/lib/random/README.md +++ /dev/null @@ -1,196 +0,0 @@ -# Random Number Generators (RNG) - -This library provides an `RNG` type (Random Number Generator). This type -represents a self-contained piece of data that encapsulates the state of a -relatively fast and relatively secure pseudo-random number generator. The -current implementation is based on the [ChaCha20 stream -cipher,](https://en.wikipedia.org/wiki/Salsa20#ChaCha_variant) inspired by -[`arc4random` in OpenBSD.](https://man.openbsd.org/arc4random.3) - -An `RNG` object can be used for deterministic, repeatable generation of -pseudorandom numbers (for example, to be used in a video game for creating -seeded levels). The default random number generator for Tomo is called `random` -and is, by default, initialized with random data from the operating system when -a Tomo program launches. - -## RNG Functions - -This documentation provides details on RNG functions available in the API. -Lists also have some methods which use RNG values: -`list.shuffle()`, `list.shuffled()`, `list.random()`, and `list.sample()`. - -- [`func bool(rng: RNG, p: Num = 0.5 -> Bool)`](#bool) -- [`func byte(rng: RNG -> Byte)`](#byte) -- [`func bytes(rng: RNG, count: Int -> [Byte])`](#bytes) -- [`func copy(rng: RNG -> RNG)`](#copy) -- [`func int(rng: RNG, min: Int, max: Int -> Int)`](#int`, `int64`, `int32`, `int16`, `int8) -- [`func new(seed: [Byte] = (/dev/urandom).read_bytes(40)! -> RNG)`](#new) -- [`func num(rng: RNG, min: Num = 0.0, max: Num = 1.0 -> Num)`](#num`, `num32) - -------------- - -### `bool` -Generate a random boolean value with a given probability. - -```tomo -func bool(rng: RNG, p: Num = 0.5 -> Bool) -``` - -- `rng`: The random number generator to use. -- `p`: The probability of returning a `yes` value. Values less than zero and - `NaN` values are treated as equal to zero and values larger than zero are - treated as equal to one. - -**Returns:** -`yes` with probability `p` and `no` with probability `1-p`. - -**Example:** -```tomo ->> random.bool() -= no ->> random.bool(1.0) -= yes -``` - ---- - -### `byte` -Generate a random byte with uniform probability. - -```tomo -func byte(rng: RNG -> Byte) -``` - -- `rng`: The random number generator to use. - -**Returns:** -A random byte (0-255). - -**Example:** -```tomo ->> random.byte() -= 103[B] -``` - ---- - -### `bytes` -Generate a list of uniformly random bytes with the given length. - -```tomo -func bytes(rng: RNG, count: Int -> [Byte]) -``` - -- `rng`: The random number generator to use. -- `count`: The number of random bytes to return. - -**Returns:** -A list of length `count` random bytes with uniform random distribution (0-255). - -**Example:** -```tomo ->> random.bytes(4) -= [135[B], 169[B], 103[B], 212[B]] -``` - ---- - -### `copy` -Return a copy of a random number generator. This copy will be a parallel version of -the given RNG with its own internal state. - -```tomo -func copy(rng: RNG -> RNG) -``` - -- `rng`: The random number generator to copy. - -**Returns:** -A copy of the given RNG. - -**Example:** -```tomo ->> rng := RNG.new([]) ->> copy := rng.copy() - ->> rng.bytes(10) -= [224[B], 102[B], 190[B], 59[B], 251[B], 50[B], 217[B], 170[B], 15[B], 221[B]] - -# The copy runs in parallel to the original RNG: ->> copy.bytes(10) -= [224[B], 102[B], 190[B], 59[B], 251[B], 50[B], 217[B], 170[B], 15[B], 221[B]] -``` - ---- - -### `int`, `int64`, `int32`, `int16`, `int8` -Generate a random integer value with the given range. - -```tomo -func int(rng: RNG, min: Int, max: Int -> Int) -func int64(rng: RNG, min: Int64 = Int64.min, max: Int64 = Int64.max -> Int) -func int32(rng: RNG, min: Int32 = Int32.min, max: Int32 = Int32.max -> Int) -func int16(rng: RNG, min: Int16 = Int16.min, max: Int16 = Int16.max -> Int) -func int8(rng: RNG, min: Int8 = Int8.min, max: Int8 = Int8.max -> Int) -``` - -- `rng`: The random number generator to use. -- `min`: The minimum value to be returned. -- `max`: The maximum value to be returned. - -**Returns:** -An integer uniformly chosen from the range `[min, max]` (inclusive). If `min` -is greater than `max`, an error will be raised. - -**Example:** -```tomo ->> random.int(1, 10) -= 8 -``` - ---- - -### `new` -Return a new random number generator. - -```tomo -func new(seed: [Byte] = (/dev/urandom).read_bytes(40)! -> RNG) -``` - -- `seed`: The seed use for the random number generator. A seed length of 40 - bytes is recommended. Seed lengths of less than 40 bytes are padded with - zeroes. - -**Returns:** -A new random number generator. - -**Example:** -```tomo ->> my_rng := RNG.new([1[B], 2[B], 3[B], 4[B]]) ->> my_rng.bool() -= yes -``` - ---- - -### `num`, `num32` -Generate a random floating point value with the given range. - -```tomo -func num(rng: RNG, min: Num = 0.0, max: Num = 1.0 -> Int) -func num32(rng: RNG, min: Num = 0.0_f32, max: Num = 1.0_f32 -> Int) -``` - -- `rng`: The random number generator to use. -- `min`: The minimum value to be returned. -- `max`: The maximum value to be returned. - -**Returns:** -A floating point number uniformly chosen from the range `[min, max]` -(inclusive). If `min` is greater than `max`, an error will be raised. - -**Example:** -```tomo ->> random.num(1, 10) -= 9.512830439975572 -``` diff --git a/lib/random/chacha.h b/lib/random/chacha.h deleted file mode 100644 index 22015819..00000000 --- a/lib/random/chacha.h +++ /dev/null @@ -1,192 +0,0 @@ -// The ChaCha stream cipher used for pseudorandom number generation - -#pragma once -/* -chacha-merged.c version 20080118 -D. J. Bernstein -Public domain. -*/ - -/* $OpenBSD: chacha_private.h,v 1.3 2022/02/28 21:56:29 dtucker Exp $ */ -/* Tomo: chacha.h,v 1.0 2024/11/03 Bruce Hill */ - -typedef unsigned char u8; -typedef unsigned int u32; - -typedef struct { - u32 input[16]; /* could be compressed */ -} chacha_ctx; - -#define KEYSZ 32 -#define IVSZ 8 - -#define U8C(v) (v##U) -#define U32C(v) (v##U) - -#define U8V(v) ((u8)(v) & U8C(0xFF)) -#define U32V(v) ((u32)(v) & U32C(0xFFFFFFFF)) - -#define ROTL32(v, n) (U32V((v) << (n)) | ((v) >> (32 - (n)))) - -#define U8TO32_LITTLE(p) (((u32)((p)[0])) | ((u32)((p)[1]) << 8) | ((u32)((p)[2]) << 16) | ((u32)((p)[3]) << 24)) - -#define U32TO8_LITTLE(p, v) \ - do { \ - (p)[0] = U8V((v)); \ - (p)[1] = U8V((v) >> 8); \ - (p)[2] = U8V((v) >> 16); \ - (p)[3] = U8V((v) >> 24); \ - } while (0) - -#define ROTATE(v, c) (ROTL32(v, c)) -#define XOR(v, w) ((v) ^ (w)) -#define PLUS(v, w) (U32V((v) + (w))) -#define PLUSONE(v) (PLUS((v), 1)) - -#define QUARTERROUND(a, b, c, d) \ - a = PLUS(a, b); \ - d = ROTATE(XOR(d, a), 16); \ - c = PLUS(c, d); \ - b = ROTATE(XOR(b, c), 12); \ - a = PLUS(a, b); \ - d = ROTATE(XOR(d, a), 8); \ - c = PLUS(c, d); \ - b = ROTATE(XOR(b, c), 7); - -static const char sigma[16] = "expand 32-byte k"; - -static void chacha_keysetup(chacha_ctx *chacha, const u8 *k) { - chacha->input[0] = U8TO32_LITTLE(sigma + 0); - chacha->input[1] = U8TO32_LITTLE(sigma + 4); - chacha->input[2] = U8TO32_LITTLE(sigma + 8); - chacha->input[3] = U8TO32_LITTLE(sigma + 12); - chacha->input[4] = U8TO32_LITTLE(k + 0); - chacha->input[5] = U8TO32_LITTLE(k + 4); - chacha->input[6] = U8TO32_LITTLE(k + 8); - chacha->input[7] = U8TO32_LITTLE(k + 12); - chacha->input[8] = U8TO32_LITTLE(k + 16); - chacha->input[9] = U8TO32_LITTLE(k + 20); - chacha->input[10] = U8TO32_LITTLE(k + 24); - chacha->input[11] = U8TO32_LITTLE(k + 28); -} - -static void chacha_ivsetup(chacha_ctx *chacha, const u8 *iv) { - chacha->input[12] = 0; - chacha->input[13] = 0; - chacha->input[14] = U8TO32_LITTLE(iv + 0); - chacha->input[15] = U8TO32_LITTLE(iv + 4); -} - -static void chacha_encrypt_bytes(chacha_ctx *chacha, const u8 *m, u8 *c, u32 bytes) { - u32 x0, x1, x2, x3, x4, x5, x6, x7, x8, x9, x10, x11, x12, x13, x14, x15; - u32 j0, j1, j2, j3, j4, j5, j6, j7, j8, j9, j10, j11, j12, j13, j14, j15; - u8 *ctarget = (u8 *)0; - u8 tmp[64]; - unsigned int i; - - if (!bytes) return; - - j0 = chacha->input[0]; - j1 = chacha->input[1]; - j2 = chacha->input[2]; - j3 = chacha->input[3]; - j4 = chacha->input[4]; - j5 = chacha->input[5]; - j6 = chacha->input[6]; - j7 = chacha->input[7]; - j8 = chacha->input[8]; - j9 = chacha->input[9]; - j10 = chacha->input[10]; - j11 = chacha->input[11]; - j12 = chacha->input[12]; - j13 = chacha->input[13]; - j14 = chacha->input[14]; - j15 = chacha->input[15]; - - for (;;) { - if (bytes < 64) { - for (i = 0; i < bytes; ++i) - tmp[i] = m[i]; - m = tmp; - ctarget = c; - c = tmp; - } - x0 = j0; - x1 = j1; - x2 = j2; - x3 = j3; - x4 = j4; - x5 = j5; - x6 = j6; - x7 = j7; - x8 = j8; - x9 = j9; - x10 = j10; - x11 = j11; - x12 = j12; - x13 = j13; - x14 = j14; - x15 = j15; - for (i = 20; i > 0; i -= 2) { - QUARTERROUND(x0, x4, x8, x12) - QUARTERROUND(x1, x5, x9, x13) - QUARTERROUND(x2, x6, x10, x14) - QUARTERROUND(x3, x7, x11, x15) - QUARTERROUND(x0, x5, x10, x15) - QUARTERROUND(x1, x6, x11, x12) - QUARTERROUND(x2, x7, x8, x13) - QUARTERROUND(x3, x4, x9, x14) - } - x0 = PLUS(x0, j0); - x1 = PLUS(x1, j1); - x2 = PLUS(x2, j2); - x3 = PLUS(x3, j3); - x4 = PLUS(x4, j4); - x5 = PLUS(x5, j5); - x6 = PLUS(x6, j6); - x7 = PLUS(x7, j7); - x8 = PLUS(x8, j8); - x9 = PLUS(x9, j9); - x10 = PLUS(x10, j10); - x11 = PLUS(x11, j11); - x12 = PLUS(x12, j12); - x13 = PLUS(x13, j13); - x14 = PLUS(x14, j14); - x15 = PLUS(x15, j15); - - j12 = PLUSONE(j12); - if (!j12) { - j13 = PLUSONE(j13); - /* stopping at 2^70 bytes per nonce is user's responsibility */ - } - - U32TO8_LITTLE(c + 0, x0); - U32TO8_LITTLE(c + 4, x1); - U32TO8_LITTLE(c + 8, x2); - U32TO8_LITTLE(c + 12, x3); - U32TO8_LITTLE(c + 16, x4); - U32TO8_LITTLE(c + 20, x5); - U32TO8_LITTLE(c + 24, x6); - U32TO8_LITTLE(c + 28, x7); - U32TO8_LITTLE(c + 32, x8); - U32TO8_LITTLE(c + 36, x9); - U32TO8_LITTLE(c + 40, x10); - U32TO8_LITTLE(c + 44, x11); - U32TO8_LITTLE(c + 48, x12); - U32TO8_LITTLE(c + 52, x13); - U32TO8_LITTLE(c + 56, x14); - U32TO8_LITTLE(c + 60, x15); - - if (bytes <= 64) { - if (bytes < 64) { - for (i = 0; i < bytes; ++i) - ctarget[i] = c[i]; - } - chacha->input[12] = j12; - chacha->input[13] = j13; - return; - } - bytes -= 64; - c += 64; - } -} diff --git a/lib/random/random.tm b/lib/random/random.tm deleted file mode 100644 index 107fad66..00000000 --- a/lib/random/random.tm +++ /dev/null @@ -1,234 +0,0 @@ -# Random Number Generator (RNG) implementation based on ChaCha - -use <assert.h> -use ./sysrandom.h -use ./chacha.h - -struct chacha_ctx(j0,j1,j2,j3,j4,j5,j6,j7,j8,j9,j10,j11,j12,j13,j14,j15:Int32; extern, secret) - func from_seed(seed:[Byte]=[] -> chacha_ctx) - return C_code:chacha_ctx( - chacha_ctx ctx; - uint8_t seed_bytes[KEYSZ + IVSZ] = {}; - for (int64_t i = 0; i < (int64_t)sizeof(seed_bytes); i++) - seed_bytes[i] = i < @seed.length ? *(uint8_t*)(@seed.data + i*@seed.stride) : 0; - chacha_keysetup(&ctx, seed_bytes); - chacha_ivsetup(&ctx, seed_bytes + KEYSZ); - ctx - ) - -random := RandomNumberGenerator.new() - -func _os_random_bytes(count:Int64 -> [Byte]) - return C_code:[Byte]( - uint8_t *random_bytes = GC_MALLOC_ATOMIC(@count); - assert(getrandom(random_bytes, (size_t)@count, 0) == (size_t)@count); - (List_t){.length=@count, .data=random_bytes, .stride=1, .atomic=1} - ) -struct RandomNumberGenerator(_chacha:chacha_ctx, _random_bytes:[Byte]=[]; secret) - func new(seed:[Byte]?=none, -> @RandomNumberGenerator) - ctx := chacha_ctx.from_seed(seed or _os_random_bytes(40)) - return @RandomNumberGenerator(ctx, []) - - func _rekey(rng:&RandomNumberGenerator) - rng._random_bytes = C_code:[Byte]( - Byte_t new_keystream[KEYSZ + IVSZ] = {}; - // Fill the buffer with the keystream - chacha_encrypt_bytes(&@rng->_chacha, new_keystream, new_keystream, sizeof(new_keystream)); - // Immediately reinitialize for backtracking resistance - chacha_keysetup(&@rng->_chacha, new_keystream); - chacha_ivsetup(&@rng->_chacha, new_keystream + KEYSZ); - List_t new_bytes = (List_t){.data=GC_MALLOC_ATOMIC(1024), .length=1024, .stride=1, .atomic=1}; - memset(new_bytes.data, 0, new_bytes.length); - chacha_encrypt_bytes(&@rng->_chacha, new_bytes.data, new_bytes.data, new_bytes.length); - new_bytes - ) - - func _fill_bytes(rng:&RandomNumberGenerator, dest:&Memory, needed:Int64) - C_code { - while (@needed > 0) { - if (@rng->_random_bytes.length == 0) - @(rng._rekey()); - - assert(@rng->_random_bytes.stride == 1); - - int64_t batch_size = MIN(@needed, @rng->_random_bytes.length); - uint8_t *batch_src = @rng->_random_bytes.data; - memcpy(@dest, batch_src, batch_size); - memset(batch_src, 0, batch_size); - @rng->_random_bytes.data += batch_size; - @rng->_random_bytes.length -= batch_size; - @dest += batch_size; - @needed -= batch_size; - } - } - - func bytes(rng:&RandomNumberGenerator, count:Int -> [Byte]) - count64 := Int64(count) - buf := C_code:@Memory(GC_MALLOC_ATOMIC(@count64)) - rng._fill_bytes(buf, count64) - return C_code:[Byte]((List_t){.data=@buf, .stride=1, .atomic=1, .length=@count64}) - - func byte(rng:&RandomNumberGenerator -> Byte) - byte : &Byte - rng._fill_bytes(byte, 1) - return byte[] - - func bool(rng:&RandomNumberGenerator, probability=0.5 -> Bool) - if probability == 0.5 - return rng.byte() < 0x80 - else - return rng.num(0., 1.) < 0.5 - - func int64(rng:&RandomNumberGenerator, min=Int64.min, max=Int64.max -> Int64) - fail("Random minimum value $min is larger than the maximum value $max") if min > max - return min if min == max - random_int64 : &Int64 - rng._fill_bytes(random_int64, 8) - if min == Int64.min and max == Int64.max - return random_int64 - - return C_code:Int64( - uint64_t range = (uint64_t)@max - (uint64_t)@min + 1; - uint64_t min_r = -range % range; - uint64_t r; - @random_int64 = (int64_t*)&r; - for (;;) { - @(rng._fill_bytes(random_int64, 8)); - if (r >= min_r) break; - } - (int64_t)((uint64_t)@min + (r % range)) - ) - - func int32(rng:&RandomNumberGenerator, min=Int32.min, max=Int32.max -> Int32) - fail("Random minimum value $min is larger than the maximum value $max") if min > max - return min if min == max - random_int32 : &Int32 - rng._fill_bytes(random_int32, 8) - if min == Int32.min and max == Int32.max - return random_int32 - - return C_code:Int32( - uint32_t range = (uint32_t)@max - (uint32_t)@min + 1; - uint32_t min_r = -range % range; - uint32_t r; - @random_int32 = (int32_t*)&r; - for (;;) { - @(rng._fill_bytes(random_int32, 8)); - if (r >= min_r) break; - } - (int32_t)((uint32_t)@min + (r % range)) - ) - - func int16(rng:&RandomNumberGenerator, min=Int16.min, max=Int16.max -> Int16) - fail("Random minimum value $min is larger than the maximum value $max") if min > max - return min if min == max - random_int16 : &Int16 - rng._fill_bytes(random_int16, 8) - if min == Int16.min and max == Int16.max - return random_int16 - - return C_code:Int16( - uint16_t range = (uint16_t)@max - (uint16_t)@min + 1; - uint16_t min_r = -range % range; - uint16_t r; - @random_int16 = (int16_t*)&r; - for (;;) { - @(rng._fill_bytes(random_int16, 8)); - if (r >= min_r) break; - } - (int16_t)((uint16_t)@min + (r % range)) - ) - - func int8(rng:&RandomNumberGenerator, min=Int8.min, max=Int8.max -> Int8) - fail("Random minimum value $min is larger than the maximum value $max") if min > max - return min if min == max - random_int8 : &Int8 - rng._fill_bytes(random_int8, 8) - if min == Int8.min and max == Int8.max - return random_int8 - - return C_code:Int8( - uint8_t range = (uint8_t)@max - (uint8_t)@min + 1; - uint8_t min_r = -range % range; - uint8_t r; - @random_int8 = (int8_t*)&r; - for (;;) { - @(rng._fill_bytes(random_int8, 8)); - if (r >= min_r) break; - } - (int8_t)((uint8_t)@min + (r % range)) - ) - - func num(rng:&RandomNumberGenerator, min=0., max=1. -> Num) - num_buf : &Num - return C_code:Num( - if (@min > @max) fail("Random minimum value (", @min, ") is larger than the maximum value (", @max, ")"); - if (@min == @max) return @min; - - union { - Num_t num; - uint64_t bits; - } r = {.bits=0}, one = {.num=1.0}; - @num_buf = &r.num; - @(rng._fill_bytes(num_buf, 8)); - - // Set r.num to 1.<random-bits> - r.bits &= ~(0xFFFULL << 52); - r.bits |= (one.bits & (0xFFFULL << 52)); - - r.num -= 1.0; - - (@min == 0.0 && @max == 1.0) ? r.num : ((1.0-r.num)*@min + r.num*@max) - ) - - func num32(rng:&RandomNumberGenerator, min=Num32(0.), max=Num32(1.) -> Num32) - return Num32(rng.num(Num(min), Num(max))) - - func int(rng:&RandomNumberGenerator, min:Int, max:Int -> Int) - return C_code:Int( - if (likely(((@min.small & @max.small) & 1) != 0)) { - int32_t r = @(rng.int32(Int32(min), Int32(max))); - return I_small(r); - } - - int32_t cmp = @(min <> max); - if (cmp > 0) - fail("Random minimum value (", @min, ") is larger than the maximum value (", @max, ")"); - if (cmp == 0) return @min; - - mpz_t range_size; - mpz_init_set_int(range_size, @max); - if (@min.small & 1) { - mpz_t min_mpz; - mpz_init_set_si(min_mpz, @min.small >> 2); - mpz_sub(range_size, range_size, min_mpz); - } else { - mpz_sub(range_size, range_size, *@min.big); - } - - gmp_randstate_t gmp_rng; - gmp_randinit_default(gmp_rng); - int64_t seed = @(rng.int64()); - gmp_randseed_ui(gmp_rng, (unsigned long)seed); - - mpz_t r; - mpz_init(r); - mpz_urandomm(r, gmp_rng, range_size); - - gmp_randclear(gmp_rng); - Int$plus(@min, Int$from_mpz(r)) - ) - - -func main() - >> rng := RandomNumberGenerator.new() - >> rng.num() - >> rng.num() - >> rng.num() - >> rng.num(0, 100) - >> rng.byte() - >> rng.bytes(20) - # >> rng.int(1, 100) - # >> rng.int(1, 100) - # >> rng.int(1, 100) - # >> rng.int(1, 100) diff --git a/lib/random/sysrandom.h b/lib/random/sysrandom.h deleted file mode 100644 index 3d6b800f..00000000 --- a/lib/random/sysrandom.h +++ /dev/null @@ -1,17 +0,0 @@ -// Logic for using system random numbers - -#pragma once - -#if defined(__FreeBSD__) || defined(__OpenBSD__) || defined(__NetBSD__) || defined(__APPLE__) -#include <stdlib.h> -static ssize_t getrandom(void *buf, size_t buflen, unsigned int flags) { - (void)flags; - arc4random_buf(buf, buflen); - return buflen; -} -#elif defined(__linux__) -// Use getrandom() -#include <sys/random.h> -#else -#error "Unsupported platform for secure random number generation" -#endif diff --git a/lib/shell/CHANGES.md b/lib/shell/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/shell/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/shell/README.md b/lib/shell/README.md deleted file mode 100644 index d4bcd42e..00000000 --- a/lib/shell/README.md +++ /dev/null @@ -1,13 +0,0 @@ -# Shell - -This module defines a `lang` for running shell scripts: - -```tomo -use shell_v1.0 - ->> $Shell" - seq 5 - echo DONE -":get_output() -= "1$\n2$\n3$\n4$\n5$\nDONE" -``` diff --git a/lib/shell/shell.tm b/lib/shell/shell.tm deleted file mode 100644 index f9476161..00000000 --- a/lib/shell/shell.tm +++ /dev/null @@ -1,44 +0,0 @@ -use commands_v1.0 - -lang Shell - convert(text:Text -> Shell) - return Shell.from_text("'" ++ text.replace($/'/, `'"'"'`) ++ "'") - - convert(texts:[Text] -> Shell) - return Shell.from_text(" ".join([Shell(t).text for t in texts])) - - convert(path:Path -> Shell) - return Shell(Text(path.expand_home())) - - convert(paths:[Path] -> Shell) - return Shell.from_text(" ".join([Shell(Text(p)).text for p in paths])) - - convert(n:Int -> Shell) return Shell.from_text(Text(n)) - convert(n:Int64 -> Shell) return Shell.from_text(Text(n)) - convert(n:Int32 -> Shell) return Shell.from_text(Text(n)) - convert(n:Int16 -> Shell) return Shell.from_text(Text(n)) - convert(n:Int8 -> Shell) return Shell.from_text(Text(n)) - convert(n:Num -> Shell) return Shell.from_text(Text(n)) - convert(n:Num32 -> Shell) return Shell.from_text(Text(n)) - - func command(shell:Shell -> Command) - return Command("sh", ["-c", shell.text]) - - func result(shell:Shell, input="", input_bytes:[Byte]=[] -> ProgramResult) - return shell.command().result(input=input, input_bytes=input_bytes) - - func run(shell:Shell -> ExitType) - return shell.command().run() - - func get_output(shell:Shell, input="", trim_newline=yes -> Text?) - return shell.command().get_output(input=input, trim_newline=trim_newline) - - func get_output_bytes(shell:Shell, input="", input_bytes:[Byte]=[] -> [Byte]?) - return shell.command().get_output_bytes(input=input, input_bytes=input_bytes) - - func by_line(shell:Shell -> func(->Text?)?) - return shell.command().by_line() - -func main(command:Shell) - for line in command.by_line()! - >> line diff --git a/lib/time/CHANGES.md b/lib/time/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/time/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/time/README.md b/lib/time/README.md deleted file mode 100644 index 55f725f1..00000000 --- a/lib/time/README.md +++ /dev/null @@ -1,5 +0,0 @@ -# Time - -This is a Tomo library for working with dates and times. The `Time` type that -is provided is a date-and-time datatype that refers to a specific moment in -time. diff --git a/lib/time/time.tm b/lib/time/time.tm deleted file mode 100644 index d26d7218..00000000 --- a/lib/time/time.tm +++ /dev/null @@ -1,214 +0,0 @@ -# Time - a module for dealing with dates and times -use <math.h> -use ./time_defs.h - -enum Weekday(Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, Saturday) - -struct TimeInfo(year,month,day,hour,minute,second,nanosecond:Int, weekday:Weekday, day_of_year:Int, timezone:Text) - -struct Time(tv_sec:Int64, tv_usec:Int64; extern) - func now(->Time) - return C_code : Time ( - struct timespec ts; - if (clock_gettime(CLOCK_REALTIME, &ts) != 0) - fail("Couldn't get the time!"); - (Time){.tv_sec=ts.tv_sec, .tv_usec=ts.tv_nsec/1000} - ) - - func local_timezone(->Text) - C_code { - if (_local_timezone.length < 0) { - static char buf[PATH_MAX]; - ssize_t len = readlink("/etc/localtime", buf, sizeof(buf)); - if (len < 0) - fail("Could not get local tz!"); - - char *zoneinfo = strstr(buf, "/zoneinfo/"); - if (zoneinfo) - _local_timezone = Text$from_str(zoneinfo + strlen("/zoneinfo/")); - else - fail("Could not resolve local tz!"); - } - } - return C_code:Text(_local_timezone) - - func set_local_timezone(timezone:Text) - C_code { - setenv("TZ", @(CString(timezone)), 1); - _local_timezone = @timezone; - tzset(); - } - - func format(t:Time, format="%c", timezone=Time.local_timezone() -> Text) - return C_code : Text ( - struct tm result; - time_t time = @t.tv_sec; - struct tm *final_info; - WITH_TIMEZONE(@timezone, final_info = localtime_r(&time, &result)); - static char buf[256]; - size_t len = strftime(buf, sizeof(buf), String(@format), final_info); - Text$from_strn(buf, len) - ) - - func new(year,month,day:Int, hour=0, minute=0, second=0.0, timezone=Time.local_timezone() -> Time) - return C_code : Time( - struct tm info = { - .tm_min=Int32$from_int(@minute, false), - .tm_hour=Int32$from_int(@hour, false), - .tm_mday=Int32$from_int(@day, false), - .tm_mon=Int32$from_int(@month, false) - 1, - .tm_year=Int32$from_int(@year, false) - 1900, - .tm_isdst=-1, - }; - - time_t t; - WITH_TIMEZONE(@timezone, t = mktime(&info)); - (Time){.tv_sec=t + (time_t)@second, .tv_usec=(suseconds_t)(fmod(@second, 1.0) * 1e9)} - ) - - func unix_timestamp(t:Time -> Int64) - return C_code:Int64((int64_t)@t.tv_sec) - - func from_unix_timestamp(timestamp:Int64 -> Time) - return C_code:Time((Time){.tv_sec=@timestamp};) - - func seconds_till(t:Time, target:Time -> Num) - seconds := Num(target.tv_sec - t.tv_sec) - seconds += 1e-9*Num(target.tv_usec - t.tv_usec) - return seconds - - func minutes_till(t:Time, target:Time -> Num) - return t.seconds_till(target)/60. - - func hours_till(t:Time, target:Time -> Num) - return t.seconds_till(target)/3600. - - func relative(t:Time, relative_to=Time.now(), timezone=Time.local_timezone() -> Text) - C_code { - struct tm info = {}; - struct tm relative_info = {}; - WITH_TIMEZONE(@timezone, { - localtime_r(&@t.tv_sec, &info); - localtime_r(&@relative_to.tv_sec, &relative_info); - }); - double second_diff = @(relative_to.seconds_till(t)); - if (info.tm_year != relative_info.tm_year && fabs(second_diff) > 365.*24.*60.*60.) - return num_format((long)info.tm_year - (long)relative_info.tm_year, "year"); - else if (info.tm_mon != relative_info.tm_mon && fabs(second_diff) > 31.*24.*60.*60.) - return num_format(12*((long)info.tm_year - (long)relative_info.tm_year) + (long)info.tm_mon - (long)relative_info.tm_mon, "month"); - else if (info.tm_yday != relative_info.tm_yday && fabs(second_diff) > 24.*60.*60.) - return num_format(round(second_diff/(24.*60.*60.)), "day"); - else if (info.tm_hour != relative_info.tm_hour && fabs(second_diff) > 60.*60.) - return num_format(round(second_diff/(60.*60.)), "hour"); - else if (info.tm_min != relative_info.tm_min && fabs(second_diff) > 60.) - return num_format(round(second_diff/(60.)), "minute"); - else { - if (fabs(second_diff) < 1e-6) - return num_format((long)(second_diff*1e9), "nanosecond"); - else if (fabs(second_diff) < 1e-3) - return num_format((long)(second_diff*1e6), "microsecond"); - else if (fabs(second_diff) < 1.0) - return num_format((long)(second_diff*1e3), "millisecond"); - else - return num_format((long)(second_diff), "second"); - } - } - fail("Unreachable") - - func time(t:Time, seconds=no, am_pm=yes, timezone=Time.local_timezone() -> Text) - time := if seconds and am_pm - t.format("%l:%M:%S%P") - else if seconds and not am_pm - t.format("%T") - else if not seconds and am_pm - t.format("%l:%M%P") - else - t.format("%H:%M") - return time.trim() - - func date(t:Time, timezone=Time.local_timezone() -> Text) - return t.format("%F") - - func info(t:Time, timezone=Time.local_timezone() -> TimeInfo) - ret : TimeInfo - C_code { - struct tm info = {}; - WITH_TIMEZONE(@timezone, localtime_r(&@t.tv_sec, &info)); - @ret.year = I(info.tm_year + 1900); - @ret.month = I(info.tm_mon + 1); - @ret.day = I(info.tm_mday); - @ret.hour = I(info.tm_hour); - @ret.minute = I(info.tm_min); - @ret.second = I(info.tm_sec); - @ret.nanosecond = I(@t.tv_usec); - @ret.weekday = info.tm_wday + 1; - @ret.day_of_year = I(info.tm_yday); - @ret.timezone = @timezone; - } - return ret - - func after(t:Time, seconds=0.0, minutes=0.0, hours=0.0, days=0, weeks=0, months=0, years=0, timezone=Time.local_timezone() -> Time) - return C_code : Time ( - double offset = @seconds + 60.*@minutes + 3600.*@hours ; - @t.tv_sec += (time_t)offset; - - struct tm info = {}; - WITH_TIMEZONE(@timezone, localtime_r(&@t.tv_sec, &info)); - - info.tm_mday += Int32$from_int(@days, false) + 7*Int32$from_int(@weeks, false); - info.tm_mon += Int32$from_int(@months, false); - info.tm_year += Int32$from_int(@years, false); - - time_t t = mktime(&info); - (Time){ - .tv_sec=t, - .tv_usec=@t.tv_usec + (suseconds_t)(fmod(offset, 1.0) * 1e9), - } - ) - - func parse(text:Text, format="%Y-%m-%dT%H:%M:%S%z", timezone=Time.local_timezone() -> Time?) - ret : Time? - C_code { - struct tm info = {.tm_isdst=-1}; - const char *str = Text$as_c_string(@text); - const char *fmt = Text$as_c_string(@format); - if (strstr(fmt, "%Z")) - fail("The %Z specifier is not supported for time parsing!"); - - char *invalid; - WITH_TIMEZONE(@timezone, invalid = strptime(str, fmt, &info)); - if (!invalid || invalid[0] != '\0') { - @ret.is_none = true; - } else { - long offset = info.tm_gmtoff; // Need to cache this because mktime() mutates it to local tz >:( - time_t t; - WITH_TIMEZONE(@timezone, t = mktime(&info)); - @ret.value.tv_sec = t + offset - info.tm_gmtoff; - } - } - return ret - -func _run_tests() - >> Time.now().format() - >> Time.set_local_timezone("Europe/Paris") - >> Time.now().format() - >> Time.set_local_timezone("America/New_York") - >> Time.now().format() - # >> Time.now().format(timezone="Europe/Paris") - # >> Time.now().format() - # >> Time.now().format("%Y-%m-%d") - # >> Time.new(2023, 11, 5).format() - # >> Time.local_timezone() - - # >> Time.new(2023, 11, 5).seconds_till(Time.now()) - # >> Time.new(2023, 11, 5).relative() - - # >> Time.now().info() - # >> Time.now().time() - # >> Time.now().date() - - # >> Time.parse("2023-11-05 01:01", "%Y-%m-%d %H:%M") - # >> Time.parse("2023-11-05 01:01", "%Y-%m-%d %H:%M", timezone="Europe/Paris") - -func main() - _run_tests() diff --git a/lib/time/time_defs.h b/lib/time/time_defs.h deleted file mode 100644 index eac3f23a..00000000 --- a/lib/time/time_defs.h +++ /dev/null @@ -1,34 +0,0 @@ -// Some helper logic for working with times. - -#pragma once -#include <stdlib.h> -#include <sys/time.h> -#include <time.h> - -typedef struct timeval Time; - -static OptionalText_t _local_timezone = NONE_TEXT; - -static INLINE Text_t num_format(long n, const char *unit) { - if (n == 0) return Text("now"); - return Text$from_str( - String((int64_t)labs(n), " ", unit, (n == -1 || n == 1) ? "" : "s", n <= 0 ? " ago" : " later")); -} - -static void set_local_timezone(Text_t tz) { - setenv("TZ", Text$as_c_string(tz), 1); - _local_timezone = tz; - tzset(); -} - -#define WITH_TIMEZONE(tz, body) \ - ({ \ - if (tz.length >= 0) { \ - OptionalText_t old_timezone = _local_timezone; \ - set_local_timezone(tz); \ - body; \ - set_local_timezone(old_timezone); \ - } else { \ - body; \ - } \ - }) diff --git a/lib/uuid/CHANGES.md b/lib/uuid/CHANGES.md deleted file mode 100644 index 42ae752c..00000000 --- a/lib/uuid/CHANGES.md +++ /dev/null @@ -1,5 +0,0 @@ -# Version History - -## v1.0 - -Initial version diff --git a/lib/uuid/uuid.tm b/lib/uuid/uuid.tm deleted file mode 100644 index f2be618e..00000000 --- a/lib/uuid/uuid.tm +++ /dev/null @@ -1,39 +0,0 @@ -use random_v1.0 -use time_v1.0 - -lang UUID - func v4(-> UUID) # Random UUID - bytes := &random.bytes(16) - bytes[7; unchecked] = 0x40 or (bytes[7; unchecked] and 0x0F) - bytes[9; unchecked] = (Byte(random.int8(0x8, 0xB)) << 4) or (bytes[9; unchecked] and 0x0F) - hex := "".join([b.hex() for b in bytes]) - uuid := "$(hex.slice(1, 8))-$(hex.slice(9, 12))-$(hex.slice(13, 16))-$(hex.slice(17, -1))" - return UUID.from_text(uuid) - - func v7(-> UUID) # Timestamp + random UUID - n := Time.now() - timestamp := n.tv_sec*1000 + n.tv_usec/1_000 - - bytes := [ - Byte((timestamp >> 40)), - Byte((timestamp >> 32)), - Byte((timestamp >> 24)), - Byte((timestamp >> 16)), - Byte((timestamp >> 8)), - Byte(timestamp), - (random.byte() and 0x0F) or 0x70, - random.byte(), - (random.byte() and 0x3F) or 0x80, - random.byte() for _ in 7, - ] - - hex := "".join([b.hex() for b in bytes]) - uuid := "$(hex.slice(1, 8))-$(hex.slice(9, 12))-$(hex.slice(13, 16))-$(hex.slice(17, -1))" - return UUID.from_text(uuid) - -enum UUIDVersion(v4, v7) -func main(version=UUIDVersion.v7) - when version is v4 - say(UUID.v4().text) - is v7 - say(UUID.v7().text) diff --git a/src/compile/headers.c b/src/compile/headers.c index be564144..d1131bde 100644 --- a/src/compile/headers.c +++ b/src/compile/headers.c @@ -177,7 +177,7 @@ Text_t compile_statement_type_header(env_t *env, Path_t header_path, ast_t *ast) if (glob(String(TOMO_PREFIX "/lib/tomo_" TOMO_VERSION "/", folder, "/[!._0-9]*.tm"), GLOB_TILDE, NULL, &tm_files) != 0) { - if (!try_install_module(mod)) code_err(ast, "Could not find library"); + if (!try_install_module(mod, true)) code_err(ast, "Could not find library"); } Text_t includes = EMPTY_TEXT; diff --git a/src/compile/statements.c b/src/compile/statements.c index 82a91a5e..af6a5223 100644 --- a/src/compile/statements.c +++ b/src/compile/statements.c @@ -207,7 +207,7 @@ static Text_t _compile_statement(env_t *env, ast_t *ast) { if (glob(String(TOMO_PREFIX "/lib/tomo_" TOMO_VERSION "/", folder, "/[!._0-9]*.tm"), GLOB_TILDE, NULL, &tm_files) != 0) { - if (!try_install_module(mod)) code_err(ast, "Could not find library"); + if (!try_install_module(mod, true)) code_err(ast, "Could not find library"); } Text_t initialization = EMPTY_TEXT; diff --git a/src/modules.c b/src/modules.c index 5327f28b..6658f8a5 100644 --- a/src/modules.c +++ b/src/modules.c @@ -15,6 +15,7 @@ #include "stdlib/tables.h" #include "stdlib/text.h" #include "stdlib/types.h" +#include "stdlib/util.h" #define xsystem(...) \ ({ \ @@ -23,6 +24,33 @@ errx(1, "Failed to run command: %s", String(__VA_ARGS__)); \ }) +bool install_from_modules_ini(Path_t ini_file, bool ask_confirmation) { + OptionalClosure_t by_line = Path$by_line(ini_file); + if (by_line.fn == NULL) return false; + OptionalText_t (*next_line)(void *) = by_line.fn; + module_info_t info = {}; + for (Text_t line; (line = next_line(by_line.userdata)).length >= 0;) { + char *line_str = Text$as_c_string(line); + const char *next_section = NULL; + if (!strparse(line_str, "[", &next_section, "]")) { + if (info.name) { + if (!try_install_module(info, ask_confirmation)) return false; + } + print("Checking module ", next_section, "..."); + info = (module_info_t){.name = next_section}; + continue; + } + if (!strparse(line_str, "version=", &info.version) || !strparse(line_str, "url=", &info.url) + || !strparse(line_str, "git=", &info.git) || !strparse(line_str, "path=", &info.path) + || !strparse(line_str, "revision=", &info.revision)) + continue; + } + if (info.name) { + if (!try_install_module(info, ask_confirmation)) return false; + } + return true; +} + static void read_modules_ini(Path_t ini_file, module_info_t *info) { OptionalClosure_t by_line = Path$by_line(ini_file); if (by_line.fn == NULL) return; @@ -57,28 +85,35 @@ module_info_t get_module_info(ast_t *use) { return *info; } -bool try_install_module(module_info_t mod) { +bool try_install_module(module_info_t mod, bool ask_confirmation) { Path_t dest = Path$from_text( Texts(TOMO_PREFIX "/lib/tomo_" TOMO_VERSION "/", Text$from_str(mod.name), "_", Text$from_str(mod.version))); + if (Path$exists(dest)) return true; + if (mod.git) { - OptionalText_t answer = ask(Texts("The module \"", Text$from_str(mod.name), "\" ", Text$from_str(mod.version), - " is not installed.\nDo you want to install it from git URL ", - Text$from_str(mod.git), "? [Y/n] "), - true, true); - if (!(answer.length == 0 || Text$equal_values(answer, Text("Y")) || Text$equal_values(answer, Text("y")))) - return false; + if (ask_confirmation) { + OptionalText_t answer = + ask(Texts("The module \"", Text$from_str(mod.name), "\" ", Text$from_str(mod.version), + " is not installed.\nDo you want to install it from git URL ", Text$from_str(mod.git), + "? [Y/n] "), + true, true); + if (!(answer.length == 0 || Text$equal_values(answer, Text("Y")) || Text$equal_values(answer, Text("y")))) + return false; + } print("Installing ", mod.name, " from git..."); if (mod.revision) xsystem("git clone --depth=1 --revision ", mod.revision, " ", mod.git, " ", dest); else xsystem("git clone --depth=1 ", mod.git, " ", dest); xsystem("tomo -L ", dest); return true; } else if (mod.url) { - OptionalText_t answer = - ask(Texts("The module \"", Text$from_str(mod.name), "\" ", Text$from_str(mod.version), + if (ask_confirmation) { + OptionalText_t answer = ask( + Texts("The module \"", Text$from_str(mod.name), "\" ", Text$from_str(mod.version), " is not installed.\nDo you want to install it from URL ", Text$from_str(mod.url), "? [Y/n] "), true, true); - if (!(answer.length == 0 || Text$equal_values(answer, Text("Y")) || Text$equal_values(answer, Text("y")))) - return false; + if (!(answer.length == 0 || Text$equal_values(answer, Text("Y")) || Text$equal_values(answer, Text("y")))) + return false; + } print("Installing ", mod.name, " from URL..."); @@ -102,12 +137,14 @@ bool try_install_module(module_info_t mod) { Path$remove(tmpdir, true); return true; } else if (mod.path) { - OptionalText_t answer = - ask(Texts("The module \"", Text$from_str(mod.name), "\" ", Text$from_str(mod.version), + if (ask_confirmation) { + OptionalText_t answer = ask( + Texts("The module \"", Text$from_str(mod.name), "\" ", Text$from_str(mod.version), " is not installed.\nDo you want to install it from path ", Text$from_str(mod.path), "? [Y/n] "), true, true); - if (!(answer.length == 0 || Text$equal_values(answer, Text("Y")) || Text$equal_values(answer, Text("y")))) - return false; + if (!(answer.length == 0 || Text$equal_values(answer, Text("Y")) || Text$equal_values(answer, Text("y")))) + return false; + } print("Installing ", mod.name, " from path..."); xsystem("ln -s ", mod.path, " ", dest); diff --git a/src/modules.h b/src/modules.h index 1c3b2d8e..c36d96dd 100644 --- a/src/modules.h +++ b/src/modules.h @@ -11,4 +11,5 @@ typedef struct { } module_info_t; module_info_t get_module_info(ast_t *use); -bool try_install_module(module_info_t mod); +bool install_from_modules_ini(Path_t ini_file, bool ask_confirmation); +bool try_install_module(module_info_t mod, bool ask_confirmation); @@ -260,8 +260,15 @@ int main(int argc, char *argv[]) { // This *could* be done in parallel, but there may be some dependency issues. pid_t child = fork(); if (child == 0) { - build_library(*lib); - if (should_install) install_library(*lib); + if (Text$equal_values(Path$extension(*lib, false), Text("ini"))) { + if (!install_from_modules_ini(*lib, false)) { + print("Failed to install modules from file: ", *lib); + _exit(1); + } + } else { + build_library(*lib); + if (should_install) install_library(*lib); + } _exit(0); } wait_for_child_success(child); diff --git a/src/typecheck.c b/src/typecheck.c index adb267df..80c270ef 100644 --- a/src/typecheck.c +++ b/src/typecheck.c @@ -189,7 +189,7 @@ static env_t *load_module(env_t *env, ast_t *module_ast) { if (glob(String(TOMO_PREFIX "/lib/tomo_" TOMO_VERSION "/", folder, "/[!._0-9]*.tm"), GLOB_TILDE, NULL, &tm_files) != 0) { - if (!try_install_module(mod)) code_err(module_ast, "Couldn't find or install library: ", folder); + if (!try_install_module(mod, true)) code_err(module_ast, "Couldn't find or install library: ", folder); } env_t *module_env = fresh_scope(env); |