| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
Leaks from libfontconfig.so are excluded for now, as it doesn't seem to
be our fault. The gcc mechanism to suppress leaks doesn't seem to be
working, so we fallback to awk.
In the presence of leaks, that specific test will fail by returning
non-zero, but this will not be what we want for those tests that are
currently failing with XPASS. We will only consider the ones that are
passing as those targeted for leak detection.
|
|
|
|
Inline elements are now aligned based on the text-align value of
their containing element instead of their own text-align value.
Block level elements are no longer affected by the text-align
property.
Fixes: https://github.com/dillo-browser/dillo/issues/410
|
|
|
|
Newer versions output two numbers instead of just one, so we pick the
first number:
++ compare -metric AE b-div.html_wdir/{html.png,ref.png,diff.png}
+ diffcount='0 (0)'
+ '[' '0 (0)' = 0 ']'
+ echo FAIL
FAIL
|
|
Use the environment variable DILLO_TEST_WAIT_TIME to lower the default
time to wait since dillo window appears and when we take the screenshot
to compare the rendering output.
|
|
|
|
|
|
See: https://github.com/dillo-browser/dillo/issues/410
|
|
See: https://github.com/dillo-browser/dillo/issues/406
|
|
See: https://github.com/dillo-browser/dillo/issues/407
|
|
|
|
|
|
For now it allows building the unit test without linking problems. A
more long-term solution is to split the code into separate modules that
can be tested more easily.
|
|
|
|
|
|
|
|
When the page is first layouted, we can then determine if the vertical
scrollbar needs to be present. When the vertical scrollbar is on the
left side, the whole content needs to be shifted to the right, so we
need to allocate the top level widget again. The current fix simply
marks the top level widget for allocation.
Fixes: https://github.com/dillo-browser/dillo/issues/359
|
|
See: https://lists.mailman3.com/hyperkitty/list/dillo-dev@mailman3.com/thread/L3FFAINJJMIOZQNID5HC2DHKJIIIHEYH/
|
|
This suppresses a warning on more modern systems while preserving
the ability of older systems and CI/CD to run the test suite.
> WARNING: The convert command is deprecated in IMv7, \
> use "magick" instead of "convert" or "magick convert"
Signed-off-by: Matt Jolly <kangie@gentoo.org>
|
|
|
|
|
|
|
|
|
|
When posible, change the image size so the aspect ratio is always
preserved. This is only posible when one or both dimensions have some
room to play. When both dimensions are not fixed, a naive algorithm
tries to satisfy all {min,max}-{width,height} and aspect ratio
constraints, but it may fail. In that case, the aspect ratio is not
preserved.
Fixes the HTML tests img-aspect-ratio and img-max-bounds.
|
|
Tests several combinations of size contraints on images so we can check
if the aspect ratio is kept or not.
|
|
|
|
Prevents Dillo from hoarding the CPU due to an infinite loop in the
layouting. We also return the control to FLTK to update the screen and
process events each 100 iterations, to keep the window responsive.
It doesn't fix the root cause of the github-infinite-loop test, but it
does allow the rendering to finish with no differences with the
reference test.
|
|
|
|
The strdup function is not available in POSIX-2001, so we use our own
implementation in dlib: dStrdup.
Reviewed-by: dogma
|
|
Instead use inet_addr("127.0.0.1") which is POSIX 2001 and also more
clear.
Reviewed-by: dogma
|
|
Reviewed-by: dogma
|
|
Reviewed-by: dogma
|
|
|
|
The content="0" attribute was wrongly parsed as the URL "0". The change
makes the parser ignore a redirect with a content value that doesn't
have ";" or "url=" after the delay number.
Fixes: https://github.com/dillo-browser/dillo/issues/204
|
|
When a page has an empty title like <title></title>, don't use it to set
the tab label, but instead rely on the default tab label, which is
computed from the file name.
|
|
As we run the driver with "set -e", it will stop on the first failure.
The compare(1) command from magick exits with non-zero when the inputs
are not equal:
> The compare program returns 2 on error, 0 if the images are similar,
> or a value between 0 and 1 if they are not similar.
So, to prevent it from interrupting the cleanup steps after the
comparison, we make it always success.
|
|
In order to leave a clean build directory by distclean, all generated
files must be removed. The environment variable DILLO_TEST_LEAVE_FILES
can be set to leave the generated test results, which aids debugging
when tests fail.
|
|
This will enable tests to be easily run by downstream distributions from
the tarball release.
|
|
Partially reverts commit f5a0b6030c2001e5a0fcacf8139afa800643eb47.
Fixes the rendering test table-missing-width-in-one-column.
Fixes: https://lists.mailman3.com/hyperkitty/list/dillo-dev@mailman3.com/message/H7JEBC2HYNJ6FUPQM7ILBP7I5FLU33IZ/
|
|
It is currently failing, as the unbounded column is taking all the
space.
Reported-by: Kevin Koster <dillo@ombertech.com>
Link: https://lists.mailman3.com/hyperkitty/list/dillo-dev@mailman3.com/message/H7JEBC2HYNJ6FUPQM7ILBP7I5FLU33IZ/
Link: https://www.mail-archive.com/dillo-dev@mailman3.com/msg00020.html
|
|
Fixes: https://github.com/dillo-browser/dillo/issues/109
|
|
|
|
When the width is set to auto, a different branch is used to compute the
width via getAvailWidth(true). The returned witdh needs to be clamped to
be in the min/max-width CSS range if given, as otherwise it won't be
corrected.
Fixes: https://github.com/dillo-browser/dillo/issues/89
|
|
|
|
The max-width size of the widget is larger than the viewport, so it
should be clamped by it.
|
