AnotherFoxGuy/opensourcegames
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

grammars for entries parsing now LALR(1), which is much faster

Browse Source
This commit is contained in:
Trilarion
2020-09-02 22:31:01 +02:00
parent 60fd9ed93e
commit b3f9d3af9b
18 changed files with 147 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
code/grammar_entries.lark
View File

@ -1,21 +1,26 @@
start: title description property+ (_E note)? _E building
start: title description property+ note? building
title: "#" /(?! ).+(?<! )/ _NL // not starting or ending with a space
title: "#" /(?! ).+(?<! )/ "\n" _E // not starting or ending with a space
description: "_" /(?! ).+(?<![ _])/ "_" _NL // single line not ending with underscore
property: "-" _key ":" _values _NL
_key : /(?! ).+?(?=:)(?<! )/ // key: everything until next ":", not beginning or ending with a space
_values : [_value ("," _value)*]
_value : quoted_value | unquoted_value
quoted_value : /\".+?\"/ // with quotation marks, can contain commas
unquoted_value : /(?![ \"])[^,\n]+(?<![ \"])/ // cannot contain commas, cannot start or end with quotation mark
description: "_" /(?! ).+(?<![ _])/ "_\n" _E // single line not ending with underscore
note.3: /(?![\-#]).*\n/+ // Unstructured text, not starting with - or #
note: /(?![\-#]).*\n/* // Unstructured text, not starting with - or #
building: "## Building" _NL property* note? // the "building" section
building: "## Building\n" (_E property+)? (_E note)? // the "building" section
property: "-" _key ":" _values "\n"? // a property on a single line "- key: value"
_key: /(?! ).+?(?=:)(?<! )/ // key: everything until next ":", not beginning or ending with a space
_values: [_value ("," _value)*] // a comma separated list
_value: quoted_value | unquoted_value // quoted or unquoted values
quoted_value: /\".+?\"/ // with quotation marks, can contain commas
unquoted_value: /(?![ \"])[^,\n]+(?<![ ])/ // cannot contain commas, cannot start or end with quotation mark
_NUMBER: /[0-9]+/
_E: /^$\n/m // empty new line
CR : /\r/
LF : /\n/
_NL : CR? LF
WS : (" "|/\t/)+
_EL.2 : /^$\n/m
%import common.WS
%ignore WS
%ignore WS
%ignore _EL