-
-
Notifications
You must be signed in to change notification settings - Fork 394
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
lexers: add initial version of CUE lexer (#858)
This commit adds a decent first-cut of a lexer for CUE (https://cuelang.org). The main aspects of the language are implemented, but there are likely a number of edge cases not covered, especially when it comes to string interpolation, but this is a sufficiently decent first cut so as to warrant merging into the chroma project rather than first working in a fork. This was tested locally for visual correctness using mkdir -p _examples (command cd ./cmd/chroma; go run . --lexer=cue --html --html-inline-styles) < lexers/testdata/cue.actual > _examples/cue.html and then loading _examples/cue.html in a browser. Closes #857
- Loading branch information
Showing
3 changed files
with
224 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,85 @@ | ||
| <lexer> | ||
| <config> | ||
| <name>CUE</name> | ||
| <alias>cue</alias> | ||
| <filename>*.cue</filename> | ||
| <mime_type>text/x-cue</mime_type> | ||
| <dot_all>true</dot_all> | ||
| <ensure_nl>true</ensure_nl> | ||
| </config> | ||
| <rules> | ||
| <state name="root"> | ||
| <rule pattern="[^\S\n]+"> | ||
| <token type="Text"/> | ||
| </rule> | ||
| <rule pattern="\\\n"> | ||
| <token type="Text"/> | ||
| </rule> | ||
| <rule pattern="//[^\n\r]+"> | ||
| <token type="CommentSingle"/> | ||
| </rule> | ||
| <rule pattern="\n"> | ||
| <token type="Text"/> | ||
| </rule> | ||
| <rule pattern="(\+|&&|==|<|=|-|\|\||!=|>|:|\*|&|=~|<=|\?|\[|\]|,|/|\||!~|>=|!|_\|_|\.\.\.)"> | ||
| <token type="Operator"/> | ||
| </rule> | ||
| <rule pattern="#*"+"> | ||
| <token type="LiteralString"/> | ||
| <push state="string"/> | ||
| </rule> | ||
| <rule pattern="'(\\\\|\\'|[^'\n])*['\n]"> | ||
| <token type="LiteralString"/> | ||
| </rule> | ||
| <rule pattern="0[boxX][0-9a-fA-F][_0-9a-fA-F]*|(\.\d+|\d[_\d]*(\.\d*)?)([eE][+-]?\d+)?[KMGTP]?i?"> | ||
| <token type="LiteralNumber"/> | ||
| </rule> | ||
| <rule pattern="[~!%^&*()+=|\[\]:;,.<>/?-]"> | ||
| <token type="Punctuation"/> | ||
| </rule> | ||
| <rule pattern="[{}]"> | ||
| <token type="Punctuation"/> | ||
| </rule> | ||
| <rule pattern="(import|for|if|in|let|package)\b"> | ||
| <token type="Keyword"/> | ||
| </rule> | ||
| <rule pattern="(bool|float|int|string|uint|ulong|ushort)\b\??"> | ||
| <token type="KeywordType"/> | ||
| </rule> | ||
| <rule pattern="(true|false|null|_)\b"> | ||
| <token type="KeywordConstant"/> | ||
| </rule> | ||
| <rule pattern="[_a-zA-Z]\w*"> | ||
| <token type="Name"/> | ||
| </rule> | ||
| </state> | ||
| <state name="string"> | ||
| <rule pattern="\\#*\("> | ||
| <token type="LiteralStringInterpol"/> | ||
| <push state="string-intp"/> | ||
| </rule> | ||
| <rule pattern=""+#*"> | ||
| <token type="LiteralString"/> | ||
| <pop depth="1"/> | ||
| </rule> | ||
| <rule pattern="\\['"\\nrt]|\\x[0-9a-fA-F]{2}|\\[0-7]{1,3}|\\u[0-9a-fA-F]{4}|\\U[0-9a-fA-F]{8}"> | ||
| <token type="LiteralStringEscape"/> | ||
| </rule> | ||
| <rule pattern="[^\\"]+"> | ||
| <token type="LiteralString"/> | ||
| </rule> | ||
| <rule pattern="\\"> | ||
| <token type="LiteralString"/> | ||
| </rule> | ||
| </state> | ||
| <state name="string-intp"> | ||
| <rule pattern="\)"> | ||
| <token type="LiteralStringInterpol"/> | ||
| <pop depth="1"/> | ||
| </rule> | ||
| <rule> | ||
| <include state="root"/> | ||
| </rule> | ||
| </state> | ||
| </rules> | ||
| </lexer> |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,31 @@ | ||
| "this is a test" | ||
| 5 | ||
| false | ||
|
|
||
| // This is a comment | ||
| a: { | ||
| foo: string // foo is a string | ||
| [=~"^i"]: int // all other fields starting with i are integers | ||
| [=~"^b"]: bool // all other fields starting with b are booleans | ||
| [>"c"]: string // all other fields lexically after c are strings | ||
|
|
||
| ...string // all other fields must be a string. Note: default constraints are not yet implemented. | ||
| } | ||
|
|
||
| x: #""" | ||
|
|
||
| \#(test) | ||
|
|
||
| """ | ||
|
|
||
| b: a & { | ||
| i3: 3 | ||
| bar: true | ||
| other: "a string" | ||
| } | ||
|
|
||
| A: close({ | ||
| field1: string | ||
| field2: string | ||
| }) | ||
|
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,108 @@ | ||
| [ | ||
| {"type":"LiteralString","value":"\"this is a test\""}, | ||
| {"type":"Text","value":"\n"}, | ||
| {"type":"LiteralNumber","value":"5"}, | ||
| {"type":"Text","value":"\n"}, | ||
| {"type":"KeywordConstant","value":"false"}, | ||
| {"type":"Text","value":"\n\n"}, | ||
| {"type":"CommentSingle","value":"// This is a comment"}, | ||
| {"type":"Text","value":"\n"}, | ||
| {"type":"Name","value":"a"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"Punctuation","value":"{"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Name","value":"foo"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordType","value":"string"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"CommentSingle","value":"// foo is a string"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Operator","value":"[="}, | ||
| {"type":"Punctuation","value":"~"}, | ||
| {"type":"LiteralString","value":"\"^i\""}, | ||
| {"type":"Operator","value":"]:"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordType","value":"int"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"CommentSingle","value":"// all other fields starting with i are integers"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Operator","value":"[="}, | ||
| {"type":"Punctuation","value":"~"}, | ||
| {"type":"LiteralString","value":"\"^b\""}, | ||
| {"type":"Operator","value":"]:"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordType","value":"bool"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"CommentSingle","value":"// all other fields starting with b are booleans"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Operator","value":"[\u003e"}, | ||
| {"type":"LiteralString","value":"\"c\""}, | ||
| {"type":"Operator","value":"]:"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordType","value":"string"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"CommentSingle","value":"// all other fields lexically after c are strings"}, | ||
| {"type":"Text","value":"\n\n "}, | ||
| {"type":"Operator","value":"..."}, | ||
| {"type":"KeywordType","value":"string"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"CommentSingle","value":"// all other fields must be a string. Note: default constraints are not yet implemented."}, | ||
| {"type":"Text","value":"\n"}, | ||
| {"type":"Punctuation","value":"}"}, | ||
| {"type":"Text","value":"\n\n"}, | ||
| {"type":"Name","value":"x"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"LiteralString","value":"#\"\"\"\n\n "}, | ||
| {"type":"LiteralStringInterpol","value":"\\#("}, | ||
| {"type":"Name","value":"test"}, | ||
| {"type":"LiteralStringInterpol","value":")"}, | ||
| {"type":"LiteralString","value":"\n\n\"\"\""}, | ||
| {"type":"Text","value":"\n\n"}, | ||
| {"type":"Name","value":"b"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"Name","value":"a"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"Operator","value":"\u0026"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"Punctuation","value":"{"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Name","value":"i3"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"LiteralNumber","value":"3"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Name","value":"bar"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordConstant","value":"true"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Name","value":"other"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"LiteralString","value":"\"a string\""}, | ||
| {"type":"Text","value":"\n"}, | ||
| {"type":"Punctuation","value":"}"}, | ||
| {"type":"Text","value":"\n\n"}, | ||
| {"type":"Name","value":"A"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"Name","value":"close"}, | ||
| {"type":"Punctuation","value":"({"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Name","value":"field1"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordType","value":"string"}, | ||
| {"type":"Text","value":"\n "}, | ||
| {"type":"Name","value":"field2"}, | ||
| {"type":"Operator","value":":"}, | ||
| {"type":"Text","value":" "}, | ||
| {"type":"KeywordType","value":"string"}, | ||
| {"type":"Text","value":"\n"}, | ||
| {"type":"Punctuation","value":"})"}, | ||
| {"type":"Text","value":"\n\n"} | ||
| ] |