aboutsummaryrefslogtreecommitdiff
path: root/dpi
AgeCommit message (Collapse)Author
2015-06-15handle openssl derivatives in the license permissions for the https dpicorvid
FWIW, wget currently does it as follows: "If you modify this program, or any covered work, by linking or combining it with the OpenSSL project's OpenSSL library (or a modified version of that library), containing parts covered by the terms of the OpenSSL or SSLeay licenses, the Free Software Foundation grants you additional permission to convey the resulting work. Corresponding Source for a non-source form of such a combination shall include the source code for the parts of OpenSSL used as well as that of the covered work."
2014-07-30KBcorvid
2014-10-22trim the publicsuffix TLDs yet againcorvid
in January 2010, there were 42 entries. Now there are 22, and nearly all of them are rather...marginal.
2015-06-10https rm RC4 from cipher listcorvid
2015-05-27Made view-source dpi use CSS formatting (it's shorter and cleaner)Jorge Arellano Cid
BTW, is there a point in using a monospaced font? Besides it looks like code printing (which is good), a proportional-spaced font may be easier to read.
2015-05-26Fix view-source dpi to handle null characters correctlyJorge Arellano Cid
Although not allowed in text contexts, null characters should not stop/halt/fail dpi protocol, thus the patch. Test Example. Display a file with these contents: null padding^@^@ (two trailing null characters) and view source for it. Note that dillo will not _display_ the file completely correct, it will eat a char after each null, but this is not a problem in dpi nor dpip but in rendering, the cache gets it right. Adding code to correctly _display_ these anomalous pages is probably not worth the effort though.
2014-10-17everyone's finally disabling SSL3; let's do so toocorvid
2014-09-03a couple of quick https improvementscorvid
2014-02-02our simplistic TLD list grows more and more worthlesscorvid
2013-09-24our publicsuffix tld list grows even shortercorvid
2013-01-18https dpi: disable SSLv2Johannes Hofmann
SSLv2 is considered insecure. submitted by: p37sitdu@lavabit.com
2013-01-12clean up struct typedefsp37sitdu, corvid
2013-01-09Refactored FD close calls into a single new dClose() dlib function [p37sitdu]Jorge Arellano Cid
2013-01-08downloads.cc: do not test whether the file exists or prompt to overwrite.Jeremy Henty
2012-12-16Set the FTP dpi to one try per connectionJorge Arellano Cid
Bug fix: also avoid a useless retry when given just the server name.
2012-12-12Download DPI Scroll should calculate the scrollbar width properly.Jeremy Henty
2012-12-11Download DPI Scroll should resize itself before resizing its child.Jeremy Henty
2012-12-10Use titles with dpip dialogs.Jeremy Henty
2012-12-10Downloads DPI: meaningful and consistent window titles.Jeremy Henty
2012-11-28Add "#include <FL/fl_ask.H>" wherever its functions are used.Jeremy Henty
2012-11-25fix crash in datauri dpiJohannes Hofmann
noticed by: corvid <corvid@lavabit.com> and Jeremy Henty's valgrind logs
2012-10-29declarations at start of blockcorvid
2012-10-20cookies dpi tldscorvid
2012-09-28rm obsolete commentcorvid
2012-09-28datauri: fix media type when only charset was providedcorvid
2012-09-25vsource: make send_numbered_text consistent with send_html_textcorvid
2011-11-17never treat URI schemes case-sensitivelycorvid
2011-11-12some isascii()corvid
2011-11-11continuedcorvid
2011-11-11fix IPv6 addr recognitioncorvid
I was under the impression that the brackets were part of the host syntax, but a_Url_hostname() shows that they are part of the _authority_ and are stripped out when making the hostname.
2011-11-11locale-independent ASCII character case handlingcorvid
Basically, I and i are different letters in Turkic languages, and this causes problems for str(n)casecmp and toupper/tolower in these locales when dillo is dealing with ASCII.
2011-11-11more image magiccorvid
2011-11-08image magic is case-sensitivecorvid
2011-10-17downloads DPI: make the download items resize horizontally with the main window.Jeremy Henty
2011-10-05dpi/download.cc: make the indentation more consistent.Jeremy Henty
2011-09-16cookies dpi: MSG only when returning >0 cookiescorvid
2011-09-16update TLDscorvid
What an awful situation it is in general.
2011-09-15let downloads dpi use the fltk schemecorvid
In dillo, fltk putenv()s the scheme (FLTK_SCHEME), and Fl::scheme(NULL) getenv)s it. Not that it makes much of a difference to the downloads dpi...
2011-09-13use a dRealloccorvid
mentioned by Alexander on dillo-dev
2011-09-06remove trailing whitespaceJorge Arellano Cid
2011-07-31fix leak in https certificate-checking workaroundcorvid
Thanks to Jeremy's valgrind logs
2011-07-30fix clang warningRob S
2011-07-29Workaround: request to only check the root URL's certificate (https).Jorge Arellano Cid
This avoids the dialog bombing that stems from loading multiple https images/resources in a single page. A proper fix would take either to implement the https-dpi as a server (with state), or to move back https handling into dillo.
2011-07-27escape '@' in wget log output shown in downloads dpi tooltipcorvid
There are opportunities for optimizations and cleverness, but it's nice to think of workarounds as something simple and minimally intrusive that you will get to rip out again someday. I'll have to check whether I've ever complained to the FLTK people about this specific problem.
2011-07-27disable symbol/shortcut interpretation in the downloads dpi's labelscorvid
TODO: '@' in the wget output tooltip
2011-06-30Workaround for downloads dpi: It had display glitches on some machines. !?Jorge Arellano Cid
2011-06-27Replaced fl_input() with a custom dialogJorge Arellano Cid
* avoids sound bell, allows customization (e.g. widgets, keyboard)
2011-06-21Eliminated a pack of 22 compiler warnings (gcc-4.6.1 amd64)Jorge Arellano Cid
2011-05-19delete []corvid
2011-05-18don't save huge cookiescorvid
Huge cookies confuse the code that reads them back in. I suppose there's always the possibility of getting a cookies.txt from something else, or a manually-edited one, but... Incidentally, the RFC thinks user agents should allow 4096 bytes for name+value+attributes, which we must be within, say, 50 bytes of.