Make PUBLIC/SYSTEM matching case-insensitive

Some websites may use "public" or "system" in doctypes, or completely messed up casing such as PuBlIc (unlikely, but possible). This ensures we don't care about the exact casing used. This fixes https://gitlab.com/yorickpeterse/oga/issues/199
yorickpeterse · Jan 8, 2020 · f00fa40 · f00fa40
1 parent 10e9101
commit f00fa40
Show file tree

Hide file tree

Showing 2 changed files with 28 additions and 2 deletions.
diff --git a/ext/ragel/base_lexer.rl b/ext/ragel/base_lexer.rl
@@ -289,7 +289,7 @@
     # Machine for processing doctypes. Doctype values such as the public
     # and system IDs are treated as T_STRING tokens.
     doctype := |*
-        'PUBLIC' | 'SYSTEM' => {
+        'PUBLIC'i | 'SYSTEM'i => {
             callback(id_on_doctype_type, data, encoding, ts, te);
         };
 

diff --git a/spec/oga/xml/lexer/doctype_spec.rb b/spec/oga/xml/lexer/doctype_spec.rb
@@ -108,7 +108,7 @@
 
     # Technically not valid, put in place to make sure that the Ragel rules are
     # not too greedy.
-    it 'lexes an inline doftype followed by a system ID' do
+    it 'lexes an inline doctype followed by a system ID' do
       expect(lex('<!DOCTYPE html [<!ELEMENT foo>] "foo">')).to eq([
         [:T_DOCTYPE_START, nil, 1],
         [:T_DOCTYPE_NAME, 'html', 1],
@@ -119,5 +119,31 @@
         [:T_DOCTYPE_END, nil, 1]
       ])
     end
+
+    it 'does not care about the casing when using a public doctype' do
+      expect(lex('<!DoCtYpE HtMl PuBlIc [<!ELEMENT foo>] "foo">')).to eq([
+        [:T_DOCTYPE_START, nil, 1],
+        [:T_DOCTYPE_NAME, 'HtMl', 1],
+        [:T_DOCTYPE_TYPE, 'PuBlIc', 1],
+        [:T_DOCTYPE_INLINE, '<!ELEMENT foo>', 1],
+        [:T_STRING_DQUOTE, nil, 1],
+        [:T_STRING_BODY, 'foo', 1],
+        [:T_STRING_DQUOTE, nil, 1],
+        [:T_DOCTYPE_END, nil, 1]
+      ])
+    end
+
+    it 'does not care about the casing when using a system doctype' do
+      expect(lex('<!DoCtYpE HtMl SyStEm [<!ELEMENT foo>] "foo">')).to eq([
+        [:T_DOCTYPE_START, nil, 1],
+        [:T_DOCTYPE_NAME, 'HtMl', 1],
+        [:T_DOCTYPE_TYPE, 'SyStEm', 1],
+        [:T_DOCTYPE_INLINE, '<!ELEMENT foo>', 1],
+        [:T_STRING_DQUOTE, nil, 1],
+        [:T_STRING_BODY, 'foo', 1],
+        [:T_STRING_DQUOTE, nil, 1],
+        [:T_DOCTYPE_END, nil, 1]
+      ])
+    end
   end
 end