(yaml) improve tag support; add verbatim tags (#2487)

* YAML parse non-word characters as part of tags * adds support for verbatim tags Co-authored-by: Josh Goebel <me@joshgoebel.com>
highlightjs · Apr 27, 2020 · 7502e42 · 7502e42
1 parent f86d6a9
commit 7502e42
Show file tree

Hide file tree

Showing 4 changed files with 41 additions and 4 deletions.
diff --git a/CHANGES.md b/CHANGES.md
@@ -1,3 +1,12 @@
+## Version 10.1.0 (in progress)
+
+Language improvements:
+
+- fix(yaml) Fix tags to include non-word characters (#2486) [Peter Plantinga][]
+
+[Peter Plantinga]: https://github.com/pplantinga
+
+
 ## Version 10.0.1
 
 Parser Engine Changes:

diff --git a/src/languages/yaml.js b/src/languages/yaml.js
@@ -10,6 +10,9 @@ Category: common, config
 export default function(hljs) {
   var LITERALS = 'true false yes no null';
 
+  // YAML spec allows non-reserved URI characters in tags.
+  var URI_CHARACTERS = '[\\w#;/?:@&=+$,.~*\\\'()[\\]]+'
+
   // Define keys as starting with a word character
   // ...containing word chars, spaces, colons, forward-slashes, hyphens and periods
   // ...and ending with a colon followed immediately by a space, tab or newline.
@@ -79,13 +82,22 @@ export default function(hljs) {
         excludeEnd: true,
         relevance: 0
       },
-      { // local tags
+      { // named tags
+        className: 'type',
+        begin: '!\\w+!' + URI_CHARACTERS,
+      },
+      // https://yaml.org/spec/1.2/spec.html#id2784064
+      { // verbatim tags
+        className: 'type',
+        begin: '!<' + URI_CHARACTERS + ">",
+      },
+      { // primary tags
         className: 'type',
-        begin: '!' + hljs.UNDERSCORE_IDENT_RE,
+        begin: '!' + URI_CHARACTERS,
       },
-      { // data type
+      { // secondary tags
         className: 'type',
-        begin: '!!' + hljs.UNDERSCORE_IDENT_RE,
+        begin: '!!' + URI_CHARACTERS,
       },
       { // fragment id &ref
         className: 'meta',

diff --git a/test/markup/yaml/tag.expect.txt b/test/markup/yaml/tag.expect.txt
@@ -2,3 +2,11 @@
 <span class="hljs-attr">key:</span> <span class="hljs-type">!localtagname</span> <span class="hljs-string">test</span>
 <span class="hljs-attr">key:</span> <span class="hljs-string">"!notatag"</span>
 <span class="hljs-attr">key:</span> <span class="hljs-string">'!!notatageither'</span>
+<span class="hljs-attr">key:</span> <span class="hljs-type">!!python/dict</span> <span class="hljs-string">test</span>
+<span class="hljs-attr">key:</span> <span class="hljs-type">!!python/name:module.name</span> <span class="hljs-string">test</span>
+<span class="hljs-attr">key:</span> <span class="hljs-type">!foo2.bar</span> <span class="hljs-string">test</span>
+<span class="hljs-attr">key:</span> <span class="hljs-type">!(foo.bar?):tag</span> <span class="hljs-string">test</span>
+<span class="hljs-attr">key:</span> <span class="hljs-type">!named!tag</span> <span class="hljs-string">test</span>
+
+<span class="hljs-string">---</span> <span class="hljs-type">!&lt;tag:clarkevans.com,2002:invoice&gt;</span>
+<span class="hljs-attr">invoice:</span> <span class="hljs-number">34843</span>
diff --git a/test/markup/yaml/tag.txt b/test/markup/yaml/tag.txt
@@ -2,3 +2,11 @@ key: !!builtintagname test
 key: !localtagname test
 key: "!notatag"
 key: '!!notatageither'
+key: !!python/dict test
+key: !!python/name:module.name test
+key: !foo2.bar test
+key: !(foo.bar?):tag test
+key: !named!tag test
+
+--- !<tag:clarkevans.com,2002:invoice>
+invoice: 34843