GitHub/HTMLKit
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
90 Commits 1 Branch 29 Tags
ccc355d80e58e0959ec0c6eed03aa46e73c23441
Commit Graph
90 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
iska ccc355d80e Replace unconsuming with a location mark in the current character in After DOCTYPE Name state to avoid consuming the character twice 
Otherwise the parse error would be emitted multiple times if it is an invalid or unexpected character
 2014-11-01 22:24:32 +01:00
iska 0ea6eb2b4e Fix Attribute Name state for NULL character 
The attribute should not be finalised here, since its value wasn't reached yet
 2014-11-01 22:22:38 +01:00
iska 477af1f4ab Change parse error message to 0x format instead of U+ 2014-11-01 22:06:01 +01:00
iska 97df7c5dd8 Fix Attribute Name state to prevent finalising a pending attribute name before it's finished 2014-11-01 22:05:22 +01:00
iska e8c74da8e2 Fix Bogus Comment state for nil-characters 2014-11-01 22:04:23 +01:00
iska 244021072c Remove erroneous parse-error emit in Comment End state 2014-11-01 22:03:34 +01:00
iska 0835e3d120 Add a "reconsume" method for current character to avoid scanning the Stream repeatedly 
Reconsume differs from the Unconsume method, since it doesn't rewind the scan location to the previous characters, but instead sets a flag to return the last consumed character on the next acess
 2014-11-01 22:02:36 +01:00
iska 622e22737b Fix emitted Token in Before Attribute Name state 2014-11-01 15:49:42 +01:00
iska 8fd6e89854 Fix scan location after reading Named Entity 2014-11-01 15:19:46 +01:00
iska 58220a464c Add missing break statement to prevent fall-through in End Tag Open state 
.. and change parse error to look like the others
 2014-11-01 15:19:20 +01:00
iska b3ee713e9b Fix DOCTYPE Token's public and system identifier initialisation issue 
On first access these must be inited before use
 2014-11-01 15:12:58 +01:00
iska a23e1a13c9 Fix surrogate pair handling and remove Input Stream's separate error-reporting class 
Upon reading a surrogate pair the scan location should be advanced by two characters instead of one
 2014-10-31 22:32:34 +01:00
iska bc8abb116f Fix UTF32Char to String conversion for invalid unicode characters 
Invalid, isolated surrogate pairs, UTF32 characters should handled specially and converted to UniChars first
 2014-10-31 21:46:54 +01:00
iska a7892a2edf Cleanup some parse error messages in Tokenizer for overall consistency 2014-10-31 18:26:34 +01:00
iska 2fa0e793e0 Replace dictionary with two arrays for Named Entity replacement 
Currently the entity name is binary-searched and its replacement-value is accessed via found index
 2014-10-31 18:25:54 +01:00
iska c04d4a35f3 Set performance baseline for tokenizing step 2014-10-31 18:07:18 +01:00
iska aaec5971b8 Add performance test for tokenizing step 2014-10-31 18:06:59 +01:00
iska bc3a7165ec Fix logic for Named Entity replacement 
This will be improved further later on
 2014-10-31 18:06:15 +01:00
iska 8a2422426e Fix several state switches for Attribute states 
Question Mark character was use instead of Quotation Mark
 2014-10-31 00:11:31 +01:00
iska d5ff28b1de Fix parsing numeric entities 
Numbers > UINT_MAX should be handled correctly now
 2014-10-31 00:10:41 +01:00
iska d45ebb611e Fix state switch in Attribute Name state 2014-10-30 21:28:08 +01:00
iska a28580b258 Fix initialisation of HTML5 Lib tests 
Checks for nil values and inits accordingly
 2014-10-30 21:18:56 +01:00
iska e2e4940100 Add implementation for finalising the current attribute of the current Tag Token 2014-10-30 21:18:15 +01:00
iska 1e4fe4ae45 Fix state switch in Tag Name state 
Otherwise it's an infinite loop
 2014-10-30 21:17:34 +01:00
iska 44fb192e22 Add helper methods for appending to attribute name/value 
Initialises lazily on first access
 2014-10-30 21:17:00 +01:00
iska e896c1fe0b Fix initial values for HTML Tokens and handle nil-parameter 2014-10-30 21:15:12 +01:00
iska dd95a639c3 Add nil checks for DOCTYPE identifiers when initing HTML5 Lib tests 2014-10-26 23:43:52 +01:00
iska 9d2664fce0 Fix regex match range index in Process Double Escaped method 2014-10-26 23:43:29 +01:00
iska f2993181a8 Change method names in HTML5 Lib test class 
Change Fixture to Dctionary
 2014-10-26 23:43:21 +01:00
iska a00c3a27d7 Fix several bugs in Named Entity Character Reference method 2014-10-26 23:21:07 +01:00
iska 8509cb1955 Remove faulty statement in Char Ref Attribute Value state 
Current char should not be unconsumed here, otherwise we land in an infinite loop
 2014-10-26 22:49:15 +01:00
iska f04abf6236 Fix numeric entity replacement bug where wrong variable and valid-range were used 2014-10-26 22:29:23 +01:00
iska d5a00713ad Fix equality methods in HTML token classes for the nil-cases 2014-10-26 22:27:58 +01:00
iska 6cfb6cd65f Fix tokenizer for Bogus Comment state 
Last character must be unconsumed: "emit a comment token whose data is the concatenation of all the characters starting from and including the character that caused the state machine to switch into the bogus comment state"
https://html.spec.whatwg.org/multipage/syntax.html#bogus-comment-state
 2014-10-26 19:03:23 +01:00
iska e7126a720f Add category to overwrite isEqual method in Parse Error tokens for testing 
All parse error are treated as equal
 2014-10-26 18:12:25 +01:00
iska e43d7c21d1 Change tokenzing to concatenate all adjacent character tokens into one 2014-10-26 18:11:21 +01:00
iska a3358dd8b0 Fix HTML5 Lib test to correctly handle ParseError token 
JSON structure is not strong with this one
 2014-10-26 18:10:32 +01:00
iska e73a8dc512 Revert "Adapt HTML5 Lib test class to break output into multiple character tokens" 
This reverts commit 79da32ed72.
 2014-10-26 18:02:47 +01:00
iska f45c9d33a6 Fix bug in named entity method 
initial character should be the next instead of current in the stream
 2014-10-26 16:37:48 +01:00
iska 971f7b3cf0 Add methods to access tokens in the Tokenizer class 2014-10-26 02:03:37 +02:00
iska 79da32ed72 Adapt HTML5 Lib test class to break output into multiple character tokens 2014-10-26 02:02:58 +02:00
iska 7d9b66ff8d Add "HTML Standarad" html file to tests resource for benchmarking 
As of the latest update (2014/10/23) the test file is 8.1MB
 2014-10-26 02:01:48 +02:00
iska 8088d239df Add common test case class 2014-10-26 01:59:52 +02:00
iska f96c69c6f7 Remove fast enumeration protocol and its method implementation from Tokenizer class 2014-10-26 01:55:12 +02:00
iska 141310f910 Add equality and hash method for HTML Token classes 2014-10-26 01:54:30 +02:00
iska 65bd4c15f1 Remove EOF token and replace it with a boolean 2014-10-26 01:53:46 +02:00
iska b7039aba1d Add html5lib tests folder to tests target supporting files 2014-10-26 01:52:05 +02:00
iska 5c28638be6 Add html5lib-tests as git submodule 2014-10-25 17:37:47 +02:00
iska 0910290ef2 Add generic HTML5LibTest class for performing HTML5lib tests 
https://github.com/html5lib/html5lib-tests
 2014-10-25 17:35:58 +02:00
iska 779fcf0a73 Improve parse error reason messages in Tokenizer class 2014-10-25 01:07:51 +02:00