gumbo-next

.travis.yml

@@ -9,9 +9,9 @@ os:
   - osx
 
 install:
-  - wget 'https://googletest.googlecode.com/files/gtest-1.6.0.zip'
-  - unzip gtest-1.6.0.zip
-  - ln -s gtest-1.6.0 gtest
+  - wget 'https://googletest.googlecode.com/files/gtest-1.7.0.zip'
+  - unzip gtest-1.7.0.zip
+  - ln -s gtest-1.7.0 gtest
   - sudo pip install BeautifulSoup
   - sudo pip install html5lib==0.95
   - ln -s `python -c 'import html5lib, os; print os.path.dirname(html5lib.__file__)'`/tests/testdata .

DEBUGGING.md

@@ -48,6 +48,9 @@ $ gdb .libs/lt-gumbo_test core
 
 The same goes for core dumps in other example binaries.
 
+To run only a single unit test, pass the --gtest_filter='TestName' flag to the
+lt-gumbo_test binary.
+
 Assertions
 ==========
 

Makefile.am

@@ -37,6 +37,15 @@ clean-local:
 
 endif !HAVE_SHARED_LIBGTEST
 
+src/tag_strings.h: src/tag.in
+	@sed 's/\(.*\)/"\1",/g' <$< >$@
+
+src/tag_enum.h: src/tag.in
+	@sed 's/\(.*\)/GUMBO_TAG_\U\1,/g;s/-/_/g' <$< >$@
+
+python/gumbo/gumboc_tags.py: src/tag.in
+	@sed -e '1i TagNames = [' -e 's/\(.*\)/\t"\U\1",/g' -e 's/-/_/g' -e "\$$a]" <$< >$@
+
 lib_LTLIBRARIES = libgumbo.la
 libgumbo_la_CFLAGS = -Wall
 libgumbo_la_LDFLAGS = -version-info 1:0:0 -no-undefined
@@ -55,6 +64,8 @@ libgumbo_la_SOURCES = \
 				src/string_piece.c \
 				src/string_piece.h \
 				src/tag.c \
+				src/tag_enum.h \
+				src/tag_strings.h \
 				src/token_type.h \
 				src/tokenizer.c \
 				src/tokenizer.h \

benchmarks/benchmark.cc

@@ -62,7 +62,7 @@ int main(int argc, char** argv) {
       clock_t start_time = clock();
       for (int i = 0; i < kNumReps; ++i) {
         GumboOutput* output = gumbo_parse(contents.c_str());
-        gumbo_destroy_output(&kGumboDefaultOptions, output);
+        gumbo_destroy_output(output);
       }
       clock_t end_time = clock();
       std::cout << filename << ": "

examples/clean_text.cc

@@ -66,5 +66,5 @@ int main(int argc, char** argv) {
 
   GumboOutput* output = gumbo_parse(contents.c_str());
   std::cout << cleantext(output->root) << std::endl;
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
+  gumbo_destroy_output(output);
 }

examples/find_links.cc

@@ -62,5 +62,5 @@ int main(int argc, char** argv) {
 
   GumboOutput* output = gumbo_parse(contents.c_str());
   search_for_links(output->root);
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
+  gumbo_destroy_output(output);
 }

examples/get_title.c

@@ -88,6 +88,6 @@ int main(int argc, const char** argv) {
       &kGumboDefaultOptions, input, input_length);
   const char* title = find_title(output->root);
   printf("%s\n", title);
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
+  gumbo_destroy_output(output);
   free(input);
 }

examples/positions_of_class.cc

@@ -88,5 +88,5 @@ int main(int argc, char** argv) {
   GumboOutput* output = gumbo_parse_with_options(
       &kGumboDefaultOptions, contents.data(), contents.length());
   search_for_class(output->root, contents, cls);
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
+  gumbo_destroy_output(output);
 }

examples/prettyprint.cc

@@ -210,7 +210,7 @@ static std::string prettyprint_contents(GumboNode* node, int lvl, const std::str
       contents.append(val);
 
 
-    } else if (child->type == GUMBO_NODE_ELEMENT) {
+    } else if ((child->type == GUMBO_NODE_ELEMENT) || (child->type == GUMBO_NODE_TEMPLATE)) {
 
       std::string val = prettyprint(child, lvl, indent_chars);
 
@@ -351,5 +351,5 @@ int main(int argc, char** argv) {
   GumboOutput* output = gumbo_parse_with_options(&options, contents.data(), contents.length());
   std::string indent_chars = "  ";
   std::cout << prettyprint(output->document, 0, indent_chars) << std::endl;
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
+  gumbo_destroy_output(output);
 }

examples/serialize.cc

@@ -188,7 +188,7 @@ static std::string serialize_contents(GumboNode* node) {
         contents.append(substitute_xml_entities_into_text(std::string(child->v.text.text)));
       }
 
-    } else if (child->type == GUMBO_NODE_ELEMENT) {
+    } else if (child->type == GUMBO_NODE_ELEMENT || child->type == GUMBO_NODE_TEMPLATE) {
       contents.append(serialize(child));
 
     } else if (child->type == GUMBO_NODE_WHITESPACE) {
@@ -283,5 +283,5 @@ int main(int argc, char** argv) {
 
   GumboOutput* output = gumbo_parse_with_options(&options, contents.data(), contents.length());
   std::cout << serialize(output->document) << std::endl;
-  gumbo_destroy_output(&kGumboDefaultOptions, output);
+  gumbo_destroy_output(output);
 }

python/gumbo/gumboc.py

@@ -26,6 +26,7 @@
 import contextlib
 import ctypes
 import os.path
+import gumboc_tags
 
 _name_of_lib = 'libgumbo.so'
 if sys.platform.startswith('darwin'):
@@ -246,158 +247,12 @@ def to_url(self):
 
 
 class Tag(Enum):
-  _values_ = [
-      'HTML',
-      'HEAD',
-      'TITLE',
-      'BASE',
-      'LINK',
-      'META',
-      'STYLE',
-      'SCRIPT',
-      'NOSCRIPT',
-      'TEMPLATE',
-      'BODY',
-      'ARTICLE',
-      'SECTION',
-      'NAV',
-      'ASIDE',
-      'H1',
-      'H2',
-      'H3',
-      'H4',
-      'H5',
-      'H6',
-      'HGROUP',
-      'HEADER',
-      'FOOTER',
-      'ADDRESS',
-      'P',
-      'HR',
-      'PRE',
-      'BLOCKQUOTE',
-      'OL',
-      'UL',
-      'LI',
-      'DL',
-      'DT',
-      'DD',
-      'FIGURE',
-      'FIGCAPTION',
-      'MAIN',
-      'DIV',
-      'A',
-      'EM',
-      'STRONG',
-      'SMALL',
-      'S',
-      'CITE',
-      'Q',
-      'DFN',
-      'ABBR',
-      'DATA',
-      'TIME',
-      'CODE',
-      'VAR',
-      'SAMP',
-      'KBD',
-      'SUB',
-      'SUP',
-      'I',
-      'B',
-      'U',
-      'MARK',
-      'RUBY',
-      'RT',
-      'RP',
-      'BDI',
-      'BDO',
-      'SPAN',
-      'BR',
-      'WBR',
-      'INS',
-      'DEL',
-      'IMAGE',
-      'IMG',
-      'IFRAME',
-      'EMBED',
-      'OBJECT',
-      'PARAM',
-      'VIDEO',
-      'AUDIO',
-      'SOURCE',
-      'TRACK',
-      'CANVAS',
-      'MAP',
-      'AREA',
-      'MATH',
-      'MI',
-      'MO',
-      'MN',
-      'MS',
-      'MTEXT',
-      'MGLYPH',
-      'MALIGNMARK',
-      'ANNOTATION_XML',
-      'SVG',
-      'FOREIGNOBJECT',
-      'DESC',
-      'TABLE',
-      'CAPTION',
-      'COLGROUP',
-      'COL',
-      'TBODY',
-      'THEAD',
-      'TFOOT',
-      'TR',
-      'TD',
-      'TH',
-      'FORM',
-      'FIELDSET',
-      'LEGEND',
-      'LABEL',
-      'INPUT',
-      'BUTTON',
-      'SELECT',
-      'DATALIST',
-      'OPTGROUP',
-      'OPTION',
-      'TEXTAREA',
-      'KEYGEN',
-      'OUTPUT',
-      'PROGRESS',
-      'METER',
-      'DETAILS',
-      'SUMMARY',
-      'MENU',
-      'MENUITEM',
-      'APPLET',
-      'ACRONYM',
-      'BGSOUND',
-      'DIR',
-      'FRAME',
-      'FRAMESET',
-      'NOFRAMES',
-      'ISINDEX',
-      'LISTING',
-      'XMP',
-      'NEXTID',
-      'NOEMBED',
-      'PLAINTEXT',
-      'RB',
-      'STRIKE',
-      'BASEFONT',
-      'BIG',
-      'BLINK',
-      'CENTER',
-      'FONT',
-      'MARQUEE',
-      'MULTICOL',
-      'NOBR',
-      'SPACER',
-      'TT',
-      'UNKNOWN',
-      ]
+  @staticmethod
+  def from_str(tagname):
+    text_ptr = ctypes.c_char_p(tagname.encode('utf-8'))
+    return _tag_enum(text_ptr)
+
+  _values_ = gumboc_tags.TagNames + ['UNKNOWN', 'LAST']
 
 
 class Element(ctypes.Structure):
@@ -444,7 +299,8 @@ def __repr__(self):
 
 
 class NodeType(Enum):
-  _values_ = ['DOCUMENT', 'ELEMENT', 'TEXT', 'CDATA', 'COMMENT', 'WHITESPACE']
+  _values_ = ['DOCUMENT', 'ELEMENT', 'TEXT', 'CDATA',
+              'COMMENT', 'WHITESPACE', 'TEMPLATE']
 
 
 class NodeUnion(ctypes.Union):
@@ -463,7 +319,7 @@ def _contents(self):
     # __getattr__, so we factor it out to a helper.
     if self.type == NodeType.DOCUMENT:
       return self.v.document
-    elif self.type == NodeType.ELEMENT:
+    elif self.type in (NodeType.ELEMENT, NodeType.TEMPLATE):
       return self.v.element
     else:
       return self.v.text
@@ -496,11 +352,6 @@ def __repr__(self):
 
 class Options(ctypes.Structure):
   _fields_ = [
-      # TODO(jdtang): Allow the Python API to set the allocator/deallocator
-      # function.  Right now these are treated as opaque void pointers.
-      ('allocator', ctypes.c_void_p),
-      ('deallocator', ctypes.c_void_p),
-      ('userdata', ctypes.c_void_p),
       ('tab_stop', ctypes.c_int),
       ('stop_on_first_error', ctypes.c_bool),
       ('max_errors', ctypes.c_int),
@@ -515,10 +366,10 @@ class Output(ctypes.Structure):
       ('errors', Vector),
       ]
 
-
 @contextlib.contextmanager
 def parse(text, **kwargs):
   options = Options()
+  container = kwargs.get("inner_html", Tag.LAST)
   for field_name, _ in Options._fields_:
     try:
       setattr(options, field_name, kwargs[field_name])
@@ -529,18 +380,22 @@ def parse(text, **kwargs):
   # call, it creates a temporary buffer which is destroyed when the call
   # completes, and then the original_text pointers point into invalid memory.
   text_ptr = ctypes.c_char_p(text.encode('utf-8'))
-  output = _parse_with_options(ctypes.byref(options), text_ptr, len(text))
+  output = _parse_fragment(ctypes.byref(options), text_ptr, len(text), container)
   try:
     yield output
   finally:
-    _destroy_output(ctypes.byref(options), output)
+    _destroy_output(output)
 
 _DEFAULT_OPTIONS = Options.in_dll(_dll, 'kGumboDefaultOptions')
 
 _parse_with_options = _dll.gumbo_parse_with_options
 _parse_with_options.argtypes = [_Ptr(Options), ctypes.c_char_p, ctypes.c_size_t]
 _parse_with_options.restype = _Ptr(Output)
 
+_parse_fragment = _dll.gumbo_parse_fragment
+_parse_fragment.argtypes = [_Ptr(Options), ctypes.c_char_p, ctypes.c_size_t, Tag]
+_parse_fragment.restype = _Ptr(Output)
+
 _tag_from_original_text = _dll.gumbo_tag_from_original_text
 _tag_from_original_text.argtypes = [_Ptr(StringPiece)]
 _tag_from_original_text.restype = None
@@ -550,13 +405,17 @@ def parse(text, **kwargs):
 _normalize_svg_tagname.restype = ctypes.c_char_p
 
 _destroy_output = _dll.gumbo_destroy_output
-_destroy_output.argtypes = [_Ptr(Options), _Ptr(Output)]
+_destroy_output.argtypes = [_Ptr(Output)]
 _destroy_output.restype = None
 
 _tagname = _dll.gumbo_normalized_tagname
 _tagname.argtypes = [Tag]
 _tagname.restype = ctypes.c_char_p
 
+_tag_enum = _dll.gumbo_tag_enum
+_tag_enum.argtypes = [ctypes.c_char_p]
+_tag_enum.restype = Tag
+
 __all__ = ['StringPiece', 'SourcePosition', 'AttributeNamespace', 'Attribute',
            'Vector', 'AttributeVector', 'NodeVector', 'QuirksMode', 'Document',
            'Namespace', 'Tag', 'Element', 'Text', 'NodeType', 'Node',