aboutsummaryrefslogtreecommitdiff
path: root/dw
AgeCommit message (Collapse)Author
2016-08-04Revert commit #4653 (on hold)Jorge Arellano Cid
This patch is "on hold", as it depends on reverted code, and at least would need a non-automatic merge.
2016-07-24Remove obsolete method.Sebastian Geerken
2016-07-24Fix update problem related to CSS 'clear'.Sebastian Geerken
2016-07-17Fix Layout::viewportSizeChanged.Sebastian Geerken
2016-07-17Reverse commit 4647:8c6b1b79d329.Sebastian Geerken
2016-07-17Comment on Jorge's commit.Sebastian Geerken
2016-07-17RTFL.Sebastian Geerken
2016-07-16fix Layout::viewportSizeChanged.Jorge Arellano Cid
When changing viewport size, sometimes the resize wasn't propagated properly. e.g. pravda #2 in the results below. Here follows the test suite and its results: -------------------------------------------- Legend: d4646 = dillo version #4646 in Mercurial repository. x a = Test failed, with code a = Problems 1, 9 and 3. Custom test pages at: http://www.dillo.org/test/4648/ Results Table .-------.--------.------.-----------.-----------------------.-----.-----. | | BTG | test | pravda.ru | 4ta.html | VT | PV | | | render | t.* | #1 | #2 | Girl | footb. | horos.| | | |-------|--------|------|-----------|-----------------------|-----|-----| |d4648 | ok | ok | ok | ok | ok | ok | ok | ok | ok | |d4647 | ok | ok | ok | x g | ok | ok | ok | ok | ok | |d4646 | ok | ok | x a | x l | ok | x j | x d | x n | ok | |d4645 | ok | ok | x m | x l | ok | x j | x d | ok | ok | |d4644 | ok | ok | x a | x b | ok | x j | x d | ok | x k | |incr | ok | ok | x a | x b | x c | ok | x d | ok | x k | |d4587t2|~ x f | ok | ok | ok | ok | ok | ok | x i | | |d4587+ | ok | ok | ok | ok | ok | ok | ok | x i | x k | |d4587 | ok | x e | ok | ok | ok | ok | ok | x i | x k | |d4584 | ok | x h | ok | ok | ok | ok | ok | x i | x k | -----------------------------------------------------------------------| |ok7 | ok | ok | ok | x g | ok | ok | ok | | | |tmp8 | ok | x e | | | | | | | | |tmp9 | ok | ok | | | | | | | | |tmp9.1 | ok | ok | ok | x g | ok | ok | ok | | | '-----------------------------------'-----------------------------'-----' ___________________________________________________________ / a | b | c | d | e | f | g | h | i | j | k \ incr = d4594 | 1 9 3 | 1 9 4 | 9 | 1 5 | 6 | 7 | 4 | 12 | 13 | 8 9 | 10 | 47 = prune patch '------------------------------------------------------'----' 45 = idle patch | l: 13 | m: 4 2| n: 5 9 | 46 = words patch 48 = viewportchg Problems: 1: overwrites text with images. 2: crops images (aspect ratio). 3: different layout. 4: overwrites text with float. 5: overwrites text with text. 6: CPU HOG with: t2.n, t2.n.sm 7: Missing page sections (sometimes works for the same page). 8: Small images cut in two sometimes ("ESP", "CRO", ...) 9: Wrong aspect ratio for images (text line height) 10: Page drawn with offset (but browser thinks other section is on screen). 11: CPU HOG. 12: CPU HOG with: t1, t1n, t2.n, t2.n.sm, t3. 13: overwrites text and images with text and images. Goods: *: works for the whole t.* suite. Testing (all with background images disabled): BTG: 1.- Bck/Fwd several times (first at normal then maximized). 2.- resize from max to normal several times. pravda #1: 1. Load [1] at normal size, wait most images to load, check images and check them not to overlap text, go Bck/Fwd, check again. 2. Maximize window, check images don't overlap text, check the page layout to have three main columns (the upper part). Go Bck/Fwd, check again. pravda #2: 1. Load [1] at normal size, wait most images to load, then load [2], press page down, check the "print version" float doesn't overlap text, or images (they may stack), check the two images inside it have the same size and ratio, maximize, return to normal, check the float again. 4ta: 1. Run dillo, disable images, load 4ta.html, click Girl img, go Bck/Fwd, check img, exit dillo. 2. Run dillo, disable images, load 4ta.html, click football img, go Bck/Fwd, check img, exit dillo. 3. Run dillo, disable images, load 4ta.html, click tauro img, check horoscope text, go Bck/Fwd, check h. text, click aries img, check h. text, go Bck/Fwd, check h. text, exit dillo. 4. Run dillo, disable images, load 4ta.html, click football img, go Bck/Fwd, click dep.png img, check images, go Bck/Fwd, check images. VT: 1. Load [3] at normal size, wait for most images to load, go Bck/Fwd, check images and text not to override each other. PV: 1. Load [5] at normal size, wait for most images to load, search for "re-re" with Find Text, press page down (you'll see an image), go Bck/Fwd, check the image is there. If not, you can place the cursor where the image was, if you get a hand cursor, Dillo thinks it is there. Wait for the tooltip and drag it a bit downwards to the right, Dillo will repaint from the image data. [1] http://tinyurl.com/j2yavxk [2] http://tinyurl.com/gqso9eu [3] http://tinyurl.com/msyl7x [4] http://tinyurl.com/grd6yg4 [5] http://tinyurl.com/huvf6pn
2016-07-16Change OOFFloatsMgr::markSizeChange()Jorge Arellano Cid
We change to a graph coverage model, with pruning.
2016-07-16Optimize Textblock::updateReference.Jorge Arellano Cid
This helps to avoid unnecessary resize calls.
2016-07-16Fix Layout::resizeIdle()Jorge Arellano Cid
This was three year old bug, that went undiscovered (hg#2863). The patch is simpler than it looks, it's just a while cycle enclosing the body, indentation makes it look large. How to reproduce: 1. Load [5] at normal size, wait for most images to load, search for "re-re" with Find Text, press page down (you'll see an image), go Bck/Fwd, check the image is there. If not, you can place the cursor where the image was, if you get a hand cursor, Dillo thinks it is there. Wait for the tooltip and drag it a bit downwards to the right, Dillo will repaint from the image data. [5] http://tinyurl.com/huvf6pn
2016-07-13cppcheck: fix format stringJohannes Hofmann
2016-07-12cppcheck: AFAICS a false positive (should be reported to them)Jorge Arellano Cid
A simple code order changes, which is not relevant in this case, but pleases cppcheck: - for (a,b; ; ) + for (b,a; ; )
2016-07-10RTFL.Sebastian Geerken
2016-07-09Cleanup OOFFloatsMgr::collidesH().Sebastian Geerken
2016-06-26Fix possible memory leak (SizeParams::forChild => childParams).Sebastian Geerken
2016-06-26Fix memory leak (OOFFloatsMgr::Float).Sebastian Geerken
2016-06-26Fix another valgrind complain.Sebastian Geerken
2016-06-26Fix valgrind complain.Sebastian Geerken
2016-06-26Simplify again the signature of OOFAwareWidget:getGeneratorWidth(), with a ↵Sebastian Geerken
slightly more complex (but more correct) implementatin for Textblock.
2016-06-24typoJorge Arellano Cid
2016-06-20Incremental resizing for OOFFloatsMgr.Sebastian Geerken
2016-06-18Fix <button> size (treat as inline).Sebastian Geerken
2016-06-18Merge, undo Jorges change.Sebastian Geerken
2016-06-18Documentation and comments on Jorge's patch.Sebastian Geerken
2016-06-17Fix for updateReference() (missing content, overlapped text sometimes)Jorge Arellano Cid
This is a race condition. When loading [1], tables under the form date camp, may dissapear and some overlapping text be shown below. How to reproduce: load the site (network effect may be enough), or go back and forward a few times, or maximize then back and forward. [1] http://www.btgpactual.mdgms.com/www/chile/resume.html FWIW this is the only regression I've found so far.
2016-06-17Fix for CPU-hog bugs with deeply nested elements.Jorge Arellano Cid
This patch fixes a cpu-hog regression of GROWS2 with respect to dillo-3.0.5. The problem arises with some deeply nested elements. e.g.: (for i in $(seq 1 20); do echo '<div style="float:left"><div></div><div style="display:table"><span></div></div>'; done) > t0.html (for i in $(seq 1 20); do echo '<div style="float:left"><div></div><div style="display:table"></div>'; done) > t1.html For both, taking away the SPAN or float is much faster. Note: Some "unrelated" sited (i.e. not clearly hogging the CPU) have been measured to render near two times faster with the patch.
2016-06-17Optimize Textblock::updateReference().Jorge Arellano Cid
2016-06-17RTFL.Sebastian Geerken
2016-06-14Whitespace.Sebastian Geerken
2016-06-04Revert an accidental commit of testing code (2 patches in the queue).Jorge Arellano Cid
2016-06-04imported patch pos-adjustmentsJorge Arellano Cid
2016-06-04imported patch renamesJorge Arellano Cid
2016-05-29Comment.Sebastian Geerken
2016-05-29Fix position for sizeRequest in Textblock widget (simplification).Sebastian Geerken
2016-05-29Fix position for sizeRequest in Textblock widget (comments).Sebastian Geerken
2016-05-29Fix position for sizeRequest in Textblock widget (corrections).Sebastian Geerken
2016-05-29Comments, cleanup.Sebastian Geerken
2016-05-29Comments.Sebastian Geerken
2016-05-29Fix position for sizeRequest in Textblock widget (draft).Sebastian Geerken
2016-05-29Fix SRDOP related bug in Widget::sizeRequest.Sebastian Geerken
2016-05-27RTFL messages.Sebastian Geerken
2016-05-27TyposJorge Arellano Cid
2016-05-27Textblock::getGeneratorWidth() may use caller information to adjust return ↵Sebastian Geerken
value. This fixes too small return values of getGeneratorRest().
2016-05-26Fix RTFL messages.Sebastian Geerken
2016-05-19Rewrote OOFAwareWidgetIterator next() and prev() for safety.Jorge Arellano Cid
Now: getPart(): can't receive out-of-range values for "index" and "sectionIndex". setValues(): below-range indexes go to START and above-range to END. prev(), next(): have a single value-checked entry-point to getPart(). PD: it's easier to read the source of these functions than the diff.
2016-05-17Fixed a CPU hog bug when rendering slashdot.Jorge Arellano Cid
https://linux.slashdot.org/comments.pl?sid=5852295&cid=48188823 It was hard to find, sometimes backtraces gave near 85 or more nested calls, full of different functions. There was no CPU lock because it finally solved the page (after more than two minutes in my machine). Finding the bug: > 99,9% of the effort. Actual fix : ~ 0.01% of the effort.
2016-05-16Fix segfault in OOFAwareWidget::OOFAwareWidgetIterator.Jorge Arellano Cid, Sebastian Geerken
2016-05-10Use border values only when ther is a border (float).Sebastian Geerken
2016-05-08RTFL.Sebastian Geerken