Age | Commit message (Collapse) | Author |
|
Branches can be dirty when a library sync'ed with gclient sync is
removed from DEPS and from .gitignore. The library will be removed
by "gclient sync --force".
Change-Id: I19fa4a3f13989edb41cd671fd412cc67e67819eb
Reviewed-on: https://pdfium-review.googlesource.com/42211
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
This overwrites deps that are dirty, fixing an issue with deps that are
removed in the "after" branch but not in the "before" branch.
Change-Id: I1c905dc46522455ad160a6fd57fbab979c42162e
Reviewed-on: https://pdfium-review.googlesource.com/42191
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
This reverts commit aaff8f8b902ac3058b44b0f11a236df606082383.
Reason for revert: does not fix the problem, I'll fix in another way
Original change's description:
> Discard deps that are not sync'ed anymore in safetynet_compare.py.
>
> The job is currently blocked by the tools/gyp directory, which is being
> downloaded in the "before" branch, but is not ignored anymore in the
> "after" branch, making it an uncommited change.
>
> Change-Id: I9aca33e2046fbbe2626431febcbbdbebe1937f3b
> Reviewed-on: https://pdfium-review.googlesource.com/42110
> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
> Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
TBR=thestig@chromium.org,hnakashima@chromium.org,rharrison@chromium.org
Change-Id: I840004d0d9c8d83340fdd0b4d701859cc628f9e6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://pdfium-review.googlesource.com/42190
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
The job is currently blocked by the tools/gyp directory, which is being
downloaded in the "before" branch, but is not ignored anymore in the
"after" branch, making it an uncommited change.
Change-Id: I9aca33e2046fbbe2626431febcbbdbebe1937f3b
Reviewed-on: https://pdfium-review.googlesource.com/42110
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
$SEND_EVENTS should not have been put in quotes.
Change-Id: Ia6bf7d1ac463c320c0afccf819becbb0cebf13ab
Reviewed-on: https://pdfium-review.googlesource.com/41330
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
Change-Id: I91c55632627bb9cc0d6b61ae1004db1343d6bff2
Reviewed-on: https://pdfium-review.googlesource.com/39910
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
This CL adds a proxy, FXSYS_time, for the time syscall, so that a
testing mechanism can be implemented. Specically there is now a flag
for pdfium_test, --time=, that allows setting the time since the epoch
that will be returned. This plumbed all the way down into the proxy
and allows for stable results for tests that depend on getting the
current time.
There are other places in the code base that will need to be patched
like this, that will be dealt with in follow on CLs.
BUG=pdfium:1104
Change-Id: I2de185f8d47abe46704dd579c13a54948b7f81e0
Reviewed-on: https://pdfium-review.googlesource.com/39750
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
|
|
BUG=pdfium:1024
Change-Id: If7e00eb4f6ebf6499be20e88bfc9b2ac35f2d797
Reviewed-on: https://pdfium-review.googlesource.com/39390
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
|
|
Change-Id: Idfc1c2282cc9318d1135f360ff03d99e27105127
Reviewed-on: https://pdfium-review.googlesource.com/35910
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
- Changes the URL to retrieve baselines which returns the
baselines based on the issue number (if available)
- Compare against a single baseline (instead of multiple)
- Writes the results of comparing to the baselines to a
separate output file that can then be parsed by the
recipe code.
See complementary CLs:
https://skia-review.googlesource.com/c/buildbot/+/131923
https://chromium-review.googlesource.com/c/chromium/tools/build/+/1085853
Change-Id: I90cdd77d0d26e63a5e686f4862fc68dc9c4429d3
Reviewed-on: https://pdfium-review.googlesource.com/33772
Commit-Queue: Stephan Altmueller <stephana@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
When generating PDFs using the fixup script and a .in, specify them to
to use Ahem.ttf as the text font. Ahem is a font designed to reduce
flakiness in pixel tests due to font rendering differences between
platforms. Specifically the glyphs in the font are standard size black
boxes, so though the specific text content cannot be easily compared,
things like layout and high level changes to the content can still
be tested. Testing things like specific text changes should be done
via an embedder test, where the strings in the elements can be
extracted and inspected.
The font itself is is CID type 2 font, which is relatively complex to
correctly embed in a PDF due to subsetting and other features. Instead
of embedding it in the generated PDFs, which was originally attempted,
it is being supplied to pdfium_test via the --font-dir flag. This flag
overrides where the binary looks for system fonts. This works
correctly on Mac and Linux, but not on Windows, which is why that
platform remains suppressed.
BUG=pdfium:1008,pdfium:1020
Change-Id: I00811536de98f736fc599d96b397194ccf8db0cd
Reviewed-on: https://pdfium-review.googlesource.com/27790
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
Change-Id: Iee691870bd890da8e5c8ce9f9f74d15bc213d4a8
Reviewed-on: https://pdfium-review.googlesource.com/32871
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
This adds tools/code_coverage from Chromium to DEPS and converts our
existing coverage_report.py to use it instead of gcov & lcov. This
generates a different format of HTML report, but the content appears
to be the same. Some of the coverage numbers changed a bit, due to
differences in how llvm-cov and gcov calculate executable lines, but
drilling down into the reports I think llvm-cov is more accurate
overall and there are no major discrepancies.
Large portions of the existing script are left as is and just the
report generation has been changed. I plan in follow up CLs to remove
the duplication of functionality in the PDFium scripts and modularlize
the upstream code better.
BUG=pdfium:1069
Change-Id: I009bfb8aac8f1a878e01ff70923e19bbb4774a9c
Reviewed-on: https://pdfium-review.googlesource.com/32894
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: dsinclair <dsinclair@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
This causes a crash if running verbose mode
Change-Id: Ib5fe3dada8c81874aae110e21252edad05668a36
Reviewed-on: https://pdfium-review.googlesource.com/32737
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
|
|
Also format the trailer in a cleaner multi-line format.
Change-Id: If145834b56f8678f97247b346cc0cfd4d367c501
Reviewed-on: https://pdfium-review.googlesource.com/32571
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Tom Sepez <tsepez@chromium.org>
|
|
This is useful for comparing images without slowing down performance
with a profiler.
Change-Id: Ia0c41bf8ef32d82bbc24f47f44586e22d991aa51
Reviewed-on: https://pdfium-review.googlesource.com/32352
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
Do not open the html in a browser automatically.
Change-Id: I7394b86f4a46a701dedb4fa0887515211561aee7
Reviewed-on: https://pdfium-review.googlesource.com/32315
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
The source type value Gold results is just a logical grouping
label. Having a different value for each type will allow the Gold web
UI to have a set of labels at the top that allows you to easily switch
between specific test tool results.
Currently the test scripts are using the test directories 'corpus',
'javascript', and 'pixel', which seem like reasonable labels.
BUG=pdfium:1073
Change-Id: I893a1afeb3b87ffb84948c18ae3052b2dd6216ea
Reviewed-on: https://pdfium-review.googlesource.com/32350
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
|
|
Change-Id: Ief6853cbfdb5dca0c81d76978075a2fd04989f41
Reviewed-on: https://pdfium-review.googlesource.com/8830
Reviewed-by: dsinclair <dsinclair@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
- If a .pdf had more than one page and any of them except the first
had no expectation, that page and the ones after it were not
diff'ed.
- If a .pdf did not have expectations for any or all pages, --regen
did not generate those expectations.
Change-Id: Ie6dacf07e44feb4a14a6a92eb20c9aa19858a8f9
Reviewed-on: https://pdfium-review.googlesource.com/25770
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
Change-Id: I10b1844052431fdb25d028eac39f45f481f32ba1
Reviewed-on: https://pdfium-review.googlesource.com/31351
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
|
|
Allow them to run a bit more consistently.
Change-Id: I6b0f528a73f2c872813f4644aeae886ec04879f4
Reviewed-on: https://pdfium-review.googlesource.com/29930
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
|
|
Fixup API check filename after fpdf_view_c_api_test was moved.
TBR: hnakashima@chromium.org
Change-Id: I49f1015616cb8c8a5d6a9f0646f1e19e8ed94779
Reviewed-on: https://pdfium-review.googlesource.com/29470
Reviewed-by: dsinclair <dsinclair@chromium.org>
Commit-Queue: dsinclair <dsinclair@chromium.org>
|
|
This CL makes the fpdfsdk/fpdf* files to be consistently prefixed with
fpdf_ instead of randomly dropping the _.
Change-Id: I23e3c8a0831b56bcd17c788d9fe874b2ab8b24fc
Reviewed-on: https://pdfium-review.googlesource.com/29390
Commit-Queue: dsinclair <dsinclair@chromium.org>
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
|
|
This removes the need for them to have their own runner. Due to other
upstream changes some of the expectation files needed to be
updated. The suppressions have been updated to make sure the tests do
not run for non-XFA builds.
Having these tests run on the bots surfaced an MSAN issue in the XFA
code, which is fixed in this CL.
BUG=pdfium:1020,pdfium:1008
Change-Id: I44e2fa1cba2d07f30f56f5792749acbe1ebf4b62
Reviewed-on: https://pdfium-review.googlesource.com/27710
Reviewed-by: dsinclair <dsinclair@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
|
|
This adds a number of {{}} style tags to the .in format that allow the
fix-up script to inject XFA related boilerplate into the PDF, instead of
having the .in file replicate this.
A simple example XFA .in file, called xfa_example.in has been added as a
template for future files to follow.
Finally a run_xfa_tests.py has been added for executing all of the tests
in testing/resources/xfa, which from what I can tell are not executed any
other way. All of the existing PDFs in that directory are currently
suppressed.
BUG=pdfium:1008
Change-Id: Ie055b6640969ce8291b4c96b401ebf6887dfa0c0
Reviewed-on: https://pdfium-review.googlesource.com/27631
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
BUG=pdfium:1008
Change-Id: I5136d57bd401d44b56ac19e5cfb52702afa32200
Reviewed-on: https://pdfium-review.googlesource.com/26651
Commit-Queue: Lei Zhang <thestig@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
Change-Id: I01956bff9e7250c9df84afe6b6de96ab4e9b4851
Reviewed-on: https://pdfium-review.googlesource.com/26650
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
Short timeout for now as we don't yet rely on this json.
Change-Id: I2dcbf714c04f39b120f2ad380dbd34e2569b8b78
Reviewed-on: https://pdfium-review.googlesource.com/24010
Reviewed-by: dsinclair <dsinclair@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
TBR=hnakashima@chromium.org
Change-Id: I50877f5a2ff3fb6f48630019d2481aa451945118
Reviewed-on: https://pdfium-review.googlesource.com/22774
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
|
|
Currently the warning does not mention that fpdfview_c_api_test.c needs
to be updated, and someone unfamiliar with the code needs to ask of dig
into the warning script to find what went wrong.
Change-Id: I56848438959ec64b0eab978d2e3638d4d4dc8e49
Reviewed-on: https://pdfium-review.googlesource.com/22430
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
Add a presubmit check to prevent future missing entries. Also fix an
erroneous header entry.
Change-Id: I8aeafd820de984f5af90b3e4ea428f582e82f254
Reviewed-on: https://pdfium-review.googlesource.com/21571
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
|
|
Change-Id: I73b6a477c1a66beec2b73072f23faab7db7d3342
Reviewed-on: https://pdfium-review.googlesource.com/18513
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
Change-Id: Id4798fe9a4d297678a76d0511cde7fecbf130e3e
Reviewed-on: https://pdfium-review.googlesource.com/16613
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
|
|
After a change in rendering, corpus and pixel test expected pngs
need to be regenerated manually. This option reduces the toil by
automating renaming and moving the new expected files over the old
ones.
Change-Id: I9d490369ccf946d4d4567e440b7b5252469eec46
Reviewed-on: https://pdfium-review.googlesource.com/16451
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Nicolás Peña Moreno <npm@chromium.org>
|
|
Storing the machine-readable data is a good idea in general, in this
case allowing us to create graphs with the historical runs.
The json now also contains:
- a version number in case there are future changes in the format
- the date and time when the run started
- which profiler was used
- which commits were being compared
- whether a comparison was run or not, that is, if the commit hashes
for before and after differ
Change-Id: I2913beaef30b90669ac4ffe60a656d4b69030588
Reviewed-on: https://pdfium-review.googlesource.com/13370
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Nicolás Peña <npm@chromium.org>
|
|
Given a point, if there is an annotation at the point, give if focus if
it is not focus already. If there is no annotation at the point, then
remove the focus from the focused annotation.
BUG=chromium:754594
Change-Id: Iec3070472bbbfbad9d86e517f25da560f82efd4e
Reviewed-on: https://pdfium-review.googlesource.com/12530
Commit-Queue: Lei Zhang <thestig@chromium.org>
Reviewed-by: dsinclair <dsinclair@chromium.org>
|
|
- Add --no-color option.
- Fix exit code ambiguity (2 means bad usage)
Change-Id: I04be9c3f21b5a71857b7847b3cc32bf595084994
Reviewed-on: https://pdfium-review.googlesource.com/11490
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Nicolás Peña <npm@chromium.org>
|
|
Also trying to get unicode filenames right again.
Change-Id: I501c94921b92b8a8cd6a10441aff1595fc6d878e
Reviewed-on: https://pdfium-review.googlesource.com/10630
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Nicolás Peña <npm@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
|
|
This applies the proper configuration rules the end HTML generation.
Bug:
Change-Id: Ica159131e574fd805a1363db5b26a6581ecc0ac6
Reviewed-on: https://pdfium-review.googlesource.com/10810
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
|
|
safetynet.job.py verifies if there were performance changes since
its last run. Its state is comprised only of the last checkpoint,
and is kept in a directory passed by argument. Results of runs are
written to this directory as well.
Change-Id: I94e0e176b10fd1d2b18d84f82427f63be107ae17
Reviewed-on: https://pdfium-review.googlesource.com/10370
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
|
|
The previous fix (encoding on print) breaks when running compare.py
directly. The real problem was in writing the json.
Change-Id: Ie4cf926b5c6153731c1ae187ec6a7cffa74ce67b
Reviewed-on: https://pdfium-review.googlesource.com/10653
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
Change-Id: I5475b37aad69361e2e7fae4ce69faf011e3b5c40
Reviewed-on: https://pdfium-review.googlesource.com/10030
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
json.loads() reads non-ascii filenames to a non-unicode string and an
encode step is required before a format.
Change-Id: I3145c7aa41e947f78898e73484c3cb1655418689
Reviewed-on: https://pdfium-review.googlesource.com/10410
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
Change-Id: Icb1d85167a234c9862a4011dea5dfaa09299c71c
Reviewed-on: https://pdfium-review.googlesource.com/10431
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
Run from the pdfium root:
$ testing/tools/safetynet_compare.py testing
This compares the current branch with and without local changes.
$ testing/tools/safetynet_compare.py testing/corpus --branch-before x
This compares the current branch + local changes against branch x.
It runs only the corpus tests.
$ testing/tools/safetynet_compare.py testing --branch-before x
--branch-after y --build-dir=~/output_compare
This compares branch x and branch y. x and y can be revision hashes.
The callgrind.out files of cases with significant changes will be
created in ~/output_compare.
$ testing/tools/safetynet_compare.py -h
Print all options.
Change-Id: I43aaf5fe890745db611fb3bc00a656ef799fdfef
Reviewed-on: https://pdfium-review.googlesource.com/7390
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
Change-Id: I35423437015ad9057ed862ab630c9a2f403a8e1f
Reviewed-on: https://pdfium-review.googlesource.com/8710
Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
|
|
This is a script that measures the performance of pdfium when loading
and rendering a single pdf file. Supported profilers for now are
callgrind (default) and perfstat.
It will be used by the compare script, which will compare measurements
this script takes.
Change-Id: I9c78ec879a7f4839e60f217d5dd8ee35b1e9a051
Reviewed-on: https://pdfium-review.googlesource.com/7354
Reviewed-by: Lei Zhang <thestig@chromium.org>
Reviewed-by: Nicolás Peña <npm@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|
|
When Chrome renders PDF pages it uses the progressive renderer. The one
shot renderer is used to render print preview pages. This CL changes
pdfium_test to use the progressive renderer by default to better match
Chrome. This also adds a --render-oneshot flag which will use the
non-progressive renderer.
Bug: chromium:736695
Change-Id: If79333ae10a0ed1c9bce1290caad8d531e6709aa
Reviewed-on: https://pdfium-review.googlesource.com/8130
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Commit-Queue: dsinclair <dsinclair@chromium.org>
|
|
Change-Id: I957199b63f6049a586b8de2688501f5ae3449964
Reviewed-on: https://pdfium-review.googlesource.com/7038
Reviewed-by: Nicolás Peña <npm@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
|