fix(http) avoid recursive sublanguage and tighten rules (#2893)

* tweaks, add response markup test
* do not trigger autodetect * unless we are sure it is HTTP
* more specific rules
* add illegal to child modes

CHANGES.md

@@ -7,6 +7,7 @@ New Languages:
 Language improvements:
 
 - enh(makefile): Add `make` as an alias (#2883) [tripleee][]
+- fix(http) avoid recursive sublanguage and tighten rules (#2893) [Josh Goebel][]
 - enh(swift) Improved grammar for strings (#2819) [Steven Van Impe][]
 - fix(asciidoc): Handle section titles level 5 (#2868) [Vaibhav Chanana][]
 

src/languages/http.js

@@ -6,43 +6,85 @@ Category: common, protocols
 Website: https://developer.mozilla.org/en-US/docs/Web/HTTP/Overview
 */
 
+import * as regex from '../lib/regex.js';
+
 export default function(hljs) {
-  var VERSION = 'HTTP/[0-9\\.]+';
+  const VERSION = 'HTTP/(2|1\\.[01])';
+  const HEADER_NAME = /[A-Za-z][A-Za-z0-9-]*/;
+  const HEADERS_AND_BODY = [
+    {
+      className: 'attribute',
+      begin: regex.concat('^', HEADER_NAME, '(?=\\:\\s)'),
+      starts: {
+        contains: [
+          {
+            className: "punctuation",
+            begin: /: /,
+            relevance: 0,
+            starts: {
+              end: '$',
+              relevance: 0
+            }
+          }
+        ]
+      }
+    },
+    {
+      begin: '\\n\\n',
+      starts: { subLanguage: [], endsWithParent: true }
+    }
+  ];
+
   return {
     name: 'HTTP',
     aliases: ['https'],
-    illegal: '\\S',
+    illegal: /\S/,
     contains: [
+      // response
       {
-        begin: '^' + VERSION, end: '$',
-        contains: [{className: 'number', begin: '\\b\\d{3}\\b'}]
+        begin: '^(?=' + VERSION + " \\d{3})",
+        end: /$/,
+        contains: [
+          {
+            className: "meta",
+            begin: VERSION
+          },
+          {
+            className: 'number', begin: '\\b\\d{3}\\b'
+          }
+        ],
+        starts: {
+          end: /\b\B/,
+          illegal: /\S/,
+          contains: HEADERS_AND_BODY
+        }
       },
+      // request
       {
-        begin: '^[A-Z]+ (.*?) ' + VERSION + '$', returnBegin: true, end: '$',
+        begin: '(?=^[A-Z]+ (.*?) ' + VERSION + '$)',
+        end: /$/,
         contains: [
           {
             className: 'string',
-            begin: ' ', end: ' ',
-            excludeBegin: true, excludeEnd: true
+            begin: ' ',
+            end: ' ',
+            excludeBegin: true,
+            excludeEnd: true
           },
           {
+            className: "meta",
             begin: VERSION
           },
           {
             className: 'keyword',
             begin: '[A-Z]+'
           }
-        ]
-      },
-      {
-        className: 'attribute',
-        begin: '^\\w', end: ': ', excludeEnd: true,
-        illegal: '\\n|\\s|=',
-        starts: {end: '$', relevance: 0}
-      },
-      {
-        begin: '\\n\\n',
-        starts: {subLanguage: [], endsWithParent: true}
+        ],
+        starts: {
+          end: /\b\B/,
+          illegal: /\S/,
+          contains: HEADERS_AND_BODY
+        }
       }
     ]
   };

test/markup/http/default.expect.txt

@@ -1,7 +1,7 @@
-<span class="hljs-keyword">POST</span> <span class="hljs-string">/task?id=1</span> HTTP/1.1
-<span class="hljs-attribute">Host</span>: example.org
-<span class="hljs-attribute">Content-Type</span>: application/json; charset=utf-8
-<span class="hljs-attribute">Content-Length</span>: 19
+<span class="hljs-keyword">POST</span> <span class="hljs-string">/task?id=1</span> <span class="hljs-meta">HTTP/1.1</span>
+<span class="hljs-attribute">Host</span><span class="hljs-punctuation">: </span>example.org
+<span class="hljs-attribute">Content-Type</span><span class="hljs-punctuation">: </span>application/json; charset=utf-8
+<span class="hljs-attribute">Content-Length</span><span class="hljs-punctuation">: </span>19
 
 <span class="json">{<span class="hljs-attr">&quot;status&quot;</span>: <span class="hljs-string">&quot;ok&quot;</span>, <span class="hljs-attr">&quot;extended&quot;</span>: <span class="hljs-literal">true</span>}
 </span>

test/markup/http/response.expect.txt

@@ -0,0 +1,11 @@
+<span class="hljs-meta">HTTP/2</span> <span class="hljs-number">301</span>
+<span class="hljs-attribute">location</span><span class="hljs-punctuation">: </span>https://www.google.com/
+<span class="hljs-attribute">content-type</span><span class="hljs-punctuation">: </span>text/html; charset=UTF-8
+<span class="hljs-attribute">date</span><span class="hljs-punctuation">: </span>Sat, 31 Oct 2020 22:47:14 GMT
+<span class="hljs-attribute">expires</span><span class="hljs-punctuation">: </span>Mon, 30 Nov 2020 22:47:14 GMT
+<span class="hljs-attribute">cache-control</span><span class="hljs-punctuation">: </span>public, max-age=2592000
+<span class="hljs-attribute">server</span><span class="hljs-punctuation">: </span>gws
+<span class="hljs-attribute">content-length</span><span class="hljs-punctuation">: </span>220
+<span class="hljs-attribute">x-xss-protection</span><span class="hljs-punctuation">: </span>0
+<span class="hljs-attribute">x-frame-options</span><span class="hljs-punctuation">: </span>SAMEORIGIN
+<span class="hljs-attribute">alt-svc</span><span class="hljs-punctuation">: </span>h3-Q050=&quot;:443&quot;; ma=2592000,h3-29=&quot;:443&quot;; ma=2592000,h3-T051=&quot;:443&quot;; ma=2592000,h3-T050=&quot;:443&quot;; ma=2592000,h3-Q046=&quot;:443&quot;; ma=2592000,h3-Q043=&quot;:443&quot;; ma=2592000,quic=&quot;:443&quot;; ma=2592000; v=&quot;46,43&quot;

test/markup/http/response.txt

@@ -0,0 +1,11 @@
+HTTP/2 301
+location: https://www.google.com/
+content-type: text/html; charset=UTF-8
+date: Sat, 31 Oct 2020 22:47:14 GMT
+expires: Mon, 30 Nov 2020 22:47:14 GMT
+cache-control: public, max-age=2592000
+server: gws
+content-length: 220
+x-xss-protection: 0
+x-frame-options: SAMEORIGIN
+alt-svc: h3-Q050=":443"; ma=2592000,h3-29=":443"; ma=2592000,h3-T051=":443"; ma=2592000,h3-T050=":443"; ma=2592000,h3-Q046=":443"; ma=2592000,h3-Q043=":443"; ma=2592000,quic=":443"; ma=2592000; v="46,43"