aboutsummaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2011-07-15Eliminated a pack of compiler warnings (gcc-4.6.1 amd64)Jorge Arellano Cid
2011-07-14wordingcorvid
2011-07-14Update dillorc URL in splash pageJorge Arellano Cid
2011-07-14Updated about:splash page and set VERSION to 2.2.1Jorge Arellano Cid
2011-07-10MSG giving HTTP status if page was emptycorvid
Sometimes a server will send, say, 500 Internal Server Error with a Content-Length of 0, and Dillo just happily shows the empty page instead of the document that you thought you were going to get, and it's confusing.
2011-07-10MSGscorvid
2011-06-07support !important in style attributesJohannes Hofmann
2011-06-05Don't allocate gif->linebuf so soon.corvid
Occasionally, linebuf leaks have appeared in Jeremy's valgrind logs. linebuf is used in Gif_literal() and Gif_sequence(). They are called by Gif_process_code(), which is called by Gif_decode(). In Gif_process_bytes(), you have to be in state 3 to call it. linebuf is allocated inside Gif_do_img_desc(), which is called by GIF_Block(). In Gif_process_bytes(), you have to be in state 2 to call it. After the allocation, there are a couple of cases where the code can return while still in state 2, which would mean coming back to Gif_do_img_desc() later and leaking the linebuf by allocating another. Jeremy has kindly run on this patch for ten days or so without incident.
2011-05-20free content/transfer decoder in Cache_entry_freecorvid
This leak had long made occasional appearances in Jeremy's valgrind logs, and I supposed it was some complicated CCC interaction, and didn't want to put a band-aid over the problem without knowing why it's happening first. But now I finally dug into it, and I see it coming from aborting in Cache_process_queue() (not viewable, for instance), and then a_Cache_process_dbuf() never gets IOClose, which makes a lot of sense!
2011-05-02indentationcorvid
2011-05-02findtext search msgcorvid
2011-05-01fix handling of CSS length values of 0 without unitJohannes Hofmann
Reported by: "Rob S." <mr_semantics@hotmail.com> Submitted by: corvid <corvid@lavabit.com>
2011-04-04clearer MSG for suspicious image sizecorvid
I saw a MSG with size 0x[something], and of course I thought it was complaining that a size was given in hex notation.
2011-04-03flush the Textblock we last added toJohannes Hofmann
In Html_write_raw() the current textblock can change while parsing HTML. Instead of flushing the Textblock we started with, we now flush the current one. This fixes a long standing drawing bug where text was not positioned initially.
2011-03-31html.cc: Html_tag_open_meta(): check that the URL exists before redirecting.Jeremy Henty
2011-01-28**/Makefile.am: #include $(top_srcdir).Jeremy Henty
2010-12-18CssParser::parseImport(): implement media-conditional @import rules.Jeremy Henty
2010-12-18cssparser.cc: move some calls to nextToken().Jeremy Henty
2010-12-18CssParser::parseImport(): call ignoreStatement().Jeremy Henty
2010-12-15silence a couple of MSGscorvid
2010-12-14border-color: transparentcorvid
2010-12-09BUG_MSG for HTML5 doctypecorvid
2010-11-26minor bracket style normalization for "else" clausesJorge Arellano Cid
2010-11-26Support @media rules.Jeremy Henty
2010-11-24trim some spacescorvid
2010-11-24spellingcorvid
2010-11-21imported patch ignore-unknown-at-rules-1Jeremy Henty
2010-11-21imported patch ignore-unknown-at-rules-0Jeremy Henty
2010-11-15fix bug comment handling of CssParserJohannes Hofmann
The skipString() method of CssParser was eating one char too much in case of a match. This caused e.g. empty comments (/**/) to be misinterpreted. Noticed and tracked down by: Jeremy Henty <onepoint@starurchin.org>
2010-11-15mergeJorge Arellano Cid
2010-11-15Moved the table border model flag from style to DilloHtmlState (i.e. stack)Jorge Arellano Cid
2010-11-13Full CSS border-style implementationJorge Arellano Cid
The drawBorder{Top,Bottom,Left,Right} functions are similar. They use a trapezium as draw polygon, or drawTypedLine() for dots and dashes. Although the concept is simple, achieving pixel accuracy is laborious [1]. [1] http://www.dillo.org/css_compat/tests/border-style.html
2010-11-09mergeJorge Arellano Cid
2010-11-09css text-indent propertycorvid
thread: http://lists.auriga.wearlab.de/pipermail/dillo-dev/2010-November/007801.html
2010-10-24mergeJorge Arellano Cid
2010-10-24imported patch border-collapseJorge Arellano Cid
2010-10-21AREA bug msgcorvid
http://www.weather.gov/ptwc/ doesn't work because it uses real numbers rather than integers for coords. The 4.01 spec doesn't quite _rigorously_ state that real numbers are no good in that case, so far as I can see, but it's pretty clearly the intention -- and html5 _does_ spell this out.
2010-10-21imported patch border-collapse-parsingJorge Arellano Cid
2010-10-17fix another Style leak in StyleEngineJohannes Hofmann
2010-10-17remove redundant assert()Johannes Hofmann
Bounds checking is already done by SimpleVector.
2010-10-17also recompute backgroundStyle in StyleEngine::restyle()Johannes Hofmann
2010-10-17fix Style leaks in StyleEngineJohannes Hofmann
2010-10-17no need to inline CssPropertyList copy constructorJohannes Hofmann
2010-10-17mergeJohannes Hofmann
2010-10-15needn't add extra closing point to image map polygonscorvid
noticed by Johannes.
2010-10-15clearer msg when can't display something as imagecorvid
2010-10-15don't try to set lineHeight if illegal value 'auto' is specifiedJohannes Hofmann
Testcase: <div style="line-height: auto"> foo<br>foo<br>foo<br> </div> This only fixes the use of uninitialzed values in this case. The real fix will be to drop the invalid declaration in the CSS parser. This needs to be implemented with a later commit. Reported-by: Jeremy Henty <onepoint@starurchin.org>
2010-10-14set border-width to 0 if border-style is 'hidden'Johannes Hofmann
2010-10-14inline setNonCssHint()Johannes Hofmann
2010-10-14fix use of a_Utf8_ideographic()corvid