aboutsummaryrefslogtreecommitdiff
path: root/nomsu.3.peg
diff options
context:
space:
mode:
authorBruce Hill <bitbucket@bruce-hill.com>2018-09-14 14:00:48 -0700
committerBruce Hill <bitbucket@bruce-hill.com>2018-09-14 14:01:07 -0700
commit7c1f2dfd69afe119322e19b3a4061876b3dd1033 (patch)
treeee540007cdad2681b03a208c657811591da58cc9 /nomsu.3.peg
parent4d48bf359df98512a2a741d48ea222d055b733c0 (diff)
Fixes and improvements, and ported nomsu.3.peg to new tidier syntax.
Diffstat (limited to 'nomsu.3.peg')
-rw-r--r--nomsu.3.peg168
1 files changed, 99 insertions, 69 deletions
diff --git a/nomsu.3.peg b/nomsu.3.peg
index 244a621..21fa86c 100644
--- a/nomsu.3.peg
+++ b/nomsu.3.peg
@@ -3,69 +3,94 @@ file:
{:curr_indent: ' '* :}
(((action / expression / inline_block / indented_block) eol !.)
/ file_chunks / empty_block)
- %ws* (!! .+ -> "Parse error" !!)?
+ {:curr_indent: %nil :}
+ !.
-shebang: "#!" (!"nomsu" [^%nl])* "nomsu" %ws+ "-V" %ws* {:version: [0-9.]+ :} [^%nl]*
+shebang: "#!" (!"nomsu" [^%nl])* "nomsu" ws+ "-V" ws* {:version: [0-9.]+ :} [^%nl]*
file_chunks (FileChunks):
{:curr_indent: ' '* :}
shebang? comment? blank_lines?
(top_block (nl_nodent section_division top_block)*)
blank_lines?
+ ws* unexpected_chunk?
+ {:curr_indent: %nil :}
top_block (Block):
{:curr_indent: ' '* :}
comment? blank_lines? statement (nl_nodent statement)*
+ {:curr_indent: %nil :}
empty_block (Block):
{:curr_indent: ' '* :}
comment? blank_lines?
+ {:curr_indent: %nil :}
-nodent: =curr_indent !(" ")
+nodent: (unexpected_indent [^%nl]* / =curr_indent)
indent: =curr_indent " "
-blank_lines: %nl ((nodent comment / %ws*) %nl)*
-eol: %ws* eol_comment? (!. / &%nl)
+blank_lines: %nl ((nodent comment / ws*) %nl)*
+eol: ws* eol_comment? (!. / &%nl)
nl_nodent: blank_lines nodent
-nl_indent: blank_lines {:curr_indent: indent :} (comment nl_nodent)?
-
-comment:
- "#" (({} {~ [^%nl]* (%nl+ (indent -> '') [^%nl]*)* ~} %userdata) => add_comment)
-eol_comment:
- "#" (({} {[^%nl]*} %userdata) => add_comment)
+nl_indent: blank_lines {:curr_indent: indent :} (comment nl_nodent)*
+
+comment (Comment):
+ "#" {~ [^%nl]* (%nl+ (indent -> '') [^%nl]*)* ~}
+eol_comment (Comment):
+ "#" {[^%nl]*}
+
+unexpected_code: ws* _unexpected_code
+_unexpected_code (Error):
+ {:error: {~ [^%nl]+ -> "Couldn't parse this code" ~} :}
+unexpected_chunk (Error):
+ {:error: {~ .+ -> "Couldn't parse this code" ~} :}
+unexpected_indent (Error):
+ {:error: {~ (=curr_indent ws+) -> "Messed up indentation" ~} :}
+ {:hint: {~ '' -> 'Either make sure this line is aligned with the one above it, or make sure the previous line ends with something that uses indentation, like ":" or "(..)"' ~} :}
+missing_paren_err (Error):
+ {:error: {~ eol -> 'Line ended without finding a closing )-parenthesis' ~} :}
+ {:hint: {~ '' -> 'Put a ")" here' ~} :}
+missing_quote_err (Error):
+ {:error: {~ eol -> 'Line ended before finding a closing double quotation mark' ~} :}
+ {:hint: {~ "" -> "Put a quotation mark here" ~} :}
+missing_bracket_error (Error):
+ {:error: {~ eol -> "Line ended before finding a closing ]-bracket" ~} :}
+ {:hint: {~ '' -> 'Put a "]" here' ~} :}
+missing_brace_error (Error):
+ {:error: {~ eol -> "Line ended before finding a closing }-brace" ~} :}
+ {:hint: {~ '' -> 'Put a "}" here' ~} :}
section_division: ("~")^+3 eol
inline_block:
- "(" %ws* inline_block %ws* ")" / raw_inline_block
+ "(" ws* inline_block ws* ")" / raw_inline_block
raw_inline_block (Block):
- (!"::") ":" %ws* ((inline_statement (%ws* ";" %ws* inline_statement)*) / !(eol nl_indent))
+ (!"::") ":" ws* ((inline_statement (ws* ";" ws* inline_statement)*) / !(eol nl_indent))
indented_block (Block):
- ":" eol nl_indent statement (nl_nodent statement)* (%nl (%ws* %nl)* nodent comment)*
+ ":" eol nl_indent statement (nl_nodent statement)*
+ (%nl (ws* %nl)* nodent (comment / eol / unexpected_code))*
+ {:curr_indent: %nil :}
statement:
- (action / expression) (eol / (!! [^%nl]+ -> "Unexpected code while parsing line" !!))
+ (action / expression) (eol / unexpected_code)
inline_statement: (inline_action / inline_expression)
noindex_inline_expression:
number / variable / inline_text / inline_list / inline_dict / inline_nomsu
/ ( "("
- %ws* (inline_action / inline_expression) %ws*
- (%ws* ',' %ws* (inline_action / inline_expression) %ws*)*
- (")"
- / (!! eol -> 'Line ended without finding a closing )-parenthesis' !!)
- / (!! [^%nl]+ -> 'Unexpected code while parsing subexpression' !!)
- )
+ ws* (inline_action / inline_expression) ws*
+ (ws* ',' ws* (inline_action / inline_expression) ws*)*
+ (")" / missing_paren_err / unexpected_code)
)
inline_expression: index_chain / noindex_inline_expression
indented_expression:
indented_text / indented_nomsu / indented_list / indented_dict / ({|
"(..)" nl_indent
- (action / expression) (nl_nodent comment)*
- (eol / (!! [^%nl]+ -> "Unexpected code while parsing indented expression" !!))
+ (action / expression) (eol / unexpected_code)
+ (%nl (ws* %nl)* nodent (comment / eol / unexpected_code))*
+ {:curr_indent: %nil :}
|} -> unpack)
- / (nl_indent (!! [^%nl]* -> "Unexpected indentation. Perhaps you meant to put a ':' or '(..)' on the previous line?" !!) (nl_nodent [^%nl]*)*)
expression:
inline_expression / indented_expression
@@ -79,45 +104,42 @@ index_chain (IndexChain):
-- Actions need either at least 1 word, or at least 2 tokens
inline_action (Action):
!section_division
- ({:target: inline_arg :} %ws* "::" %ws*)?
- ( (inline_arg (%ws* (inline_arg / word))+)
- / (word (%ws* (inline_arg / word))*))
- (%ws* inline_block)?
+ ({:target: inline_arg :} ws* "::" ws*)?
+ ( (inline_arg (ws* (inline_arg / word))+)
+ / (word (ws* (inline_arg / word))*))
+ (ws* inline_block)?
inline_arg: inline_expression / inline_block
action (Action):
!section_division
- ({:target: arg :} (nl_nodent "..")? %ws* "::" (nl_nodent "..")? %ws*)?
- ( (arg ((nl_nodent "..")? %ws* (arg / word))+)
- / (word ((nl_nodent "..")? %ws* (arg / word))*))
+ ({:target: arg :} (nl_nodent "..")? ws* "::" (nl_nodent "..")? ws*)?
+ ( (arg ((nl_nodent "..")? ws* (arg / word))+)
+ / (word ((nl_nodent "..")? ws* (arg / word))*))
arg: expression / inline_block / indented_block
-word: !number { %operator_char+ / %ident_char+ }
+word: !number { operator_char+ / ident_char+ }
text_word (Text): word
inline_text (Text):
- !('".."' eol)
- '"'
- ({~ (('\"' -> '"') / ('\\' -> '\') / %escaped_char / [^%nl\"])+ ~}
- / inline_text_interpolation)*
- ('"'
- / (!! eol -> 'Line ended before finding a closing double quotation mark' !!)
- / (!! [^%nl]+ -> 'Unexpected code while parsing Text' !!))
+ !(indented_text)
+ ('"' _inline_text* ('"' / missing_quote_err / unexpected_code))
+_inline_text:
+ {~ (('\"' -> '"') / ('\\' -> '\') / escaped_char / [^%nl\"]+)+ ~}
+ / inline_text_interpolation
inline_text_interpolation:
"\" (
variable / inline_list / inline_dict / inline_text
/ ("("
- %ws* (inline_action / inline_expression) %ws*
- (%ws* ',' %ws* (inline_action / inline_expression) %ws*)*
- (")"
- / (!! eol -> 'Line ended without finding a closing )-parenthesis' !!)
- / (!! [^%nl]+ -> 'Unexpected code while parsing Text interpolation' !!)))
+ ws* (inline_action / inline_expression) ws*
+ (ws* ',' ws* (inline_action / inline_expression) ws*)*
+ (")" / missing_paren_err / unexpected_code))
)
indented_text (Text):
- '".."' eol %nl {%nl*} {:curr_indent: indent :}
+ '".."' eol %nl {%nl+}? {:curr_indent: indent :}
(indented_plain_text / text_interpolation / {~ %nl+ (=curr_indent -> "") ~})*
- (!! [^%nl]+ -> "Unexpected code while parsing Text" !!)?
+ unexpected_code?
+ {:curr_indent: %nil :}
-- Tracking text-lines-within-indented-text as separate objects allows for better debugging line info
indented_plain_text (Text):
{~ (("\\" -> "\") / (("\" blank_lines =curr_indent "..") -> "") / (!text_interpolation "\") / [^%nl\]+)+
@@ -129,43 +151,51 @@ number (Number): (("-"? (([0-9]+ "." [0-9]+) / ("." [0-9]+) / "0x" [0-9a-fA-F]+
-- Variables can be nameless (i.e. just %) and can only contain identifier chars.
-- This ensures you don't get weird parsings of `%x+%y` or `%'s thing`.
-variable (Var): "%" {%ident_char*}
+variable (Var): "%" {ident_char*}
inline_list (List):
!('[..]')
- "[" %ws*
- (inline_list_item (%ws* ',' %ws* inline_list_item)* (%ws* ',')?)? %ws*
- ("]" / (","? (
- (!! eol -> "Line ended before finding a closing ]-bracket" !!)
- /(!! [^%nl]+ -> "Unexpected code while parsing List" !!)
- )))
+ "[" ws*
+ (inline_list_item (ws* ',' ws* inline_list_item)* (ws* ',')?)? ws*
+ ("]" / (","? (missing_bracket_error / unexpected_code)))
indented_list (List):
"[..]" eol nl_indent
- list_line (nl_nodent list_line)* (nl_nodent comment)*
- (","? (!! [^%nl]+ -> "Unexpected code while parsing List" !!))?
+ list_line (nl_nodent list_line)*
+ (%nl (ws* %nl)* nodent (comment / eol / unexpected_code))*
+ (","? unexpected_code)?
list_line:
- (inline_list_item %ws* "," %ws*)+ eol
- / (inline_list_item %ws* "," %ws*)* (action / expression) eol
+ (inline_list_item ws* "," ws*)+ eol
+ / (inline_list_item ws* "," ws*)* (action / expression) eol
inline_list_item: inline_action / inline_expression
inline_dict (Dict):
!('{..}')
- "{" %ws*
- (inline_dict_entry (%ws* ',' %ws* inline_dict_entry)*)? %ws*
- ("}" / (","? (
- (!! eol -> "Line ended before finding a closing }-brace" !!)
- / (!! [^%nl]* -> "Unexpected code while parsing Dictionary" !!)
- )))
+ "{" ws*
+ (inline_dict_entry (ws* ',' ws* inline_dict_entry)*)? ws*
+ ("}" / (","? (missing_brace_error / unexpected_code)))
indented_dict (Dict):
"{..}" eol nl_indent
- dict_line (nl_nodent dict_line)* (nl_nodent comment)*
- (","? (!! [^%nl]+ -> "Unexpected code while parsing Dictionary" !!))?
+ dict_line (nl_nodent dict_line)*
+ (%nl (ws* %nl)* nodent (comment / eol / unexpected_code))*
+ (","? unexpected_code)?
dict_line:
- (inline_dict_entry %ws* "," %ws*)+ eol
- / (inline_dict_entry %ws* "," %ws*)* dict_entry eol
+ (inline_dict_entry ws* "," ws*)+ eol
+ / (inline_dict_entry ws* "," ws*)* dict_entry eol
dict_entry(DictEntry):
- dict_key (%ws* ":" %ws* (action / expression))?
+ dict_key (ws* ":" ws* (action / expression))?
inline_dict_entry(DictEntry):
- dict_key (%ws* ":" %ws* (inline_action / inline_expression)?)?
+ dict_key (ws* ":" ws* (inline_action / inline_expression)?)?
dict_key:
text_word / inline_expression
+
+operator_char: ['`~!@$^&*+=|<>?/-]
+ident_char: [a-zA-Z0-9_] / %utf8_char
+ws: [ %tab]
+
+escaped_char:
+ ("\"->'') (
+ (([xX]->'') ((({[0-9a-fA-F]^2} %number_16) -> tonumber) -> tochar))
+ / ((([0-9] [0-9]^-2) -> tonumber) -> tochar)
+ / ("a"->ascii_7) / ("b"->ascii_8) / ("t"->ascii_9) / ("n"->ascii_10)
+ / ("v"->ascii_11) / ("f"->ascii_12) / ("r"->ascii_13)
+ )