Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
183763025:
vm_run_skia_try/write_json_summary: run skpdiff within imagediffdb, not on its own (Closed)
Descriptionvm_run_skia_try/write_json_summary: run skpdiff within imagediffdb, not on its own
This is a prerequisite for making write_json_summary use imagepairset, and thus
work with the same frontend UI as rebaseline_server.
BUG=skia:2230
NOTRY=True # no C++ files in the CL
Committed: https://skia.googlesource.com/buildbot/+/0a366a318e6e8a4601968331e17aad4e66fbe8c5
Patch Set 1 #
Total comments: 12
Patch Set 2 : ravi's comments #
Messages
Total messages: 16 (0 generated)
Ready for review at patchset 1. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/output.csv (left): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/output.csv:2: file_.png, 0.1 We don't need this file anymore, now that imagediffdb_mock.py provides the perceptual difference numbers. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/summary.json (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/summary.json:9: "perceptualSimilarity": 5, Now that perceptualSimilarity is being provided by imagediffdb_mock, it has a consistent value like the other fields. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/vm_run_skia_try.sh (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/vm_run_skia_try.sh:206: --nopatch_images_base_url=file:/$OUTPUT_DIR_NOPATCH \ I *think* this will work, but I am not 100% sure, because the unittest uses a mock ImageDiffDB which doesn't actually try to compute any diffs. Ravi, can you please try it out and make sure? It would probably be better for the unittest to either use a real ImageDiffDB pointing at images on local disk, or for the mock ImageDiffDB to make lots of assertions about the arguments that get passed to it during the test. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py:31: nopatch_images_base_url: (str) URL of directory containing all nopatch This changes because ImageDiffDB expects a URL from which it can download source images. A file:// URL is OK.
FYI, do not need NOTRY=true in buildbot CLs because we do not have trybots setup for the buildbot repository. LGTM https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/imagediffdb_mock.py (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/imagediffdb_mock.py:43: expected_image_url, expected_image_locator, Nit: Combine line 43 into line 42. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/output.csv (left): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/output.csv:2: file_.png, 0.1 On 2014/03/03 21:51:05, epoger wrote: > We don't need this file anymore, now that imagediffdb_mock.py provides the > perceptual difference numbers. Acknowledged. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/summary.json (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/summary.json:9: "perceptualSimilarity": 5, On 2014/03/03 21:51:05, epoger wrote: > Now that perceptualSimilarity is being provided by imagediffdb_mock, it has a > consistent value like the other fields. Acknowledged. (Rietveld needs an 'Ack' button, I should add it). https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/vm_run_skia_try.sh (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/vm_run_skia_try.sh:206: --nopatch_images_base_url=file:/$OUTPUT_DIR_NOPATCH \ On 2014/03/03 21:51:05, epoger wrote: > I *think* this will work, but I am not 100% sure, because the unittest uses a > mock ImageDiffDB which doesn't actually try to compute any diffs. > > Ravi, can you please try it out and make sure? I am ok with submitting this and then I will run end-to-end tests on it. > > It would probably be better for the unittest to either use a real ImageDiffDB > pointing at images on local disk, or for the mock ImageDiffDB to make lots of > assertions about the arguments that get passed to it during the test. The unittest should probably not use a real ImageDiffDB but your suggestion about lots of assertions in the mock one makes sense to me. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py:31: nopatch_images_base_url: (str) URL of directory containing all nopatch On 2014/03/03 21:51:05, epoger wrote: > This changes because ImageDiffDB expects a URL from which it can download source > images. A file:// URL is OK. Acknowledged.
https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/imagediffdb_mock.py (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/imagediffdb_mock.py:43: expected_image_url, expected_image_locator, On 2014/03/04 13:14:19, rmistry wrote: > Nit: Combine line 43 into line 42. Done. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/summary.json (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/test_data/summary.json:9: "perceptualSimilarity": 5, On 2014/03/04 13:14:19, rmistry wrote: > On 2014/03/03 21:51:05, epoger wrote: > > Now that perceptualSimilarity is being provided by imagediffdb_mock, it has a > > consistent value like the other fields. > > Acknowledged. (Rietveld needs an 'Ack' button, I should add it). Ack. Ack-tually, I guess it would also work for us to use the abbreviation "ack" along with SGTM, etc. Just clicking a button is faster, but a short word ain't too bad. https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... File compute_engine_scripts/telemetry/telemetry_slave_scripts/vm_run_skia_try.sh (right): https://codereview.chromium.org/183763025/diff/1/compute_engine_scripts/telem... compute_engine_scripts/telemetry/telemetry_slave_scripts/vm_run_skia_try.sh:206: --nopatch_images_base_url=file:/$OUTPUT_DIR_NOPATCH \ On 2014/03/04 13:14:19, rmistry wrote: > On 2014/03/03 21:51:05, epoger wrote: > > Ravi, can you please try it out and make sure? > > I am ok with submitting this and then I will run end-to-end tests on it. Thanks. > The unittest should probably not use a real ImageDiffDB but your suggestion > about lots of assertions in the mock one makes sense to me. Added a TODO in imagediffdb_mock.py
The CQ bit was checked by epoger@google.com
CQ is trying da patch. Follow status at https://skia-tree-status.appspot.com/cq/epoger@google.com/183763025/20001
Message was sent while issue was closed.
Change committed as 0a366a318e6e8a4601968331e17aad4e66fbe8c5
Message was sent while issue was closed.
Unittests passed in http://108.170.219.160:10115/builders/Housekeeper-PerCommit/builds/386
Message was sent while issue was closed.
On 2014/03/04 14:54:24, epoger wrote: > Unittests passed in > http://108.170.219.160:10115/builders/Housekeeper-PerCommit/builds/386 I kicked off a skia tryserver run in cluster telemetry.
Message was sent while issue was closed.
On 2014/03/04 14:55:10, rmistry wrote: > On 2014/03/04 14:54:24, epoger wrote: > > Unittests passed in > > http://108.170.219.160:10115/builders/Housekeeper-PerCommit/builds/386 > > I kicked off a skia tryserver run in cluster telemetry. Failed with: == Comparing pictures and saving differences in JSON output file == /home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py:62: SyntaxWarning: assertion is always true, perhaps remove parentheses? assert (files_to_checksums_nopatch.keys() == /home/default/skia-repo/trunk/gm/gm_json.py:15: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute import import io /home/default/skia-repo/trunk/gm/gm_json.py:16: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute import import json /home/default/skia-repo/trunk/gm/gm_json.py:17: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute import import os /home/default/skia-repo/trunk/gm/gm_json.py:18: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute import import posixpath /home/default/skia-repo/trunk/gm/gm_json.py:19: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute import import re /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:12: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import contextlib /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:13: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import csv /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:14: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import logging /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:15: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import os /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:16: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import re /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:17: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import shutil /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:18: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import sys /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:19: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import tempfile /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:20: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import urllib /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:22: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import from PIL import Image, ImageChops /home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py:32: RuntimeWarning: Parent module '/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb' not found while handling absolute import import find_run_binary ERROR:root:got exception while creating new DiffRecord Traceback (most recent call last): File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", line 257, in add_image_pair actual_image_locator=actual_image_locator) File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", line 97, in __init__ expected_image_url) File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", line 361, in _download_and_open_image with contextlib.closing(urllib.urlopen(url)) as url_handle: File "/usr/lib/python2.7/urllib.py", line 86, in urlopen return opener.open(url) File "/usr/lib/python2.7/urllib.py", line 207, in open return getattr(self, name)(url) File "/usr/lib/python2.7/urllib.py", line 460, in open_file return self.open_ftp(url) File "/usr/lib/python2.7/urllib.py", line 519, in open_ftp host = socket.gethostbyname(host) IOError: [Errno socket error] [Errno -2] Name or service not known Traceback (most recent call last): File "/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py", line 198, in <module> options.gm_json_path, options.imagediffdb_path) File "/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py", line 100, in WriteJsonSummary actual_image_locator=checksum_withpatch) File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", line 270, in get_diff_record return self._diff_dict[key] KeyError: ('1582992237666751297', '15063702165824040060') == Copy everything to Google Storage == cat: /tmp/summary-rmistry-1393944525.78/slave1.json: No such file or directory CommandException: No URIs matched: /tmp/diffs/* CommandException: No URIs matched: /tmp/whitediffs/* CommandException: No URIs matched: /tmp/summary-rmistry-1393944525.78/* HEAD is now at f23910b New baselines for perspective switch Copying file:///tmp/skia-try.rmistry-1393944525.78.log [Content-Type=application/octet-stream]...
Message was sent while issue was closed.
On 2014/03/04 15:05:08, rmistry wrote:
> > I kicked off a skia tryserver run in cluster telemetry.
>
> Failed with:
Yikes, all sorts of problems here. I'll respond to some of them, inline.
> == Comparing pictures and saving differences in JSON output file ==
>
/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py:62:
> SyntaxWarning: assertion is always true, perhaps remove parentheses?
> assert (files_to_checksums_nopatch.keys() ==
I don't understand this one, but it seems harmless enough.
> /home/default/skia-repo/trunk/gm/gm_json.py:15: RuntimeWarning: Parent module
> '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute
I see these "Parent module 'xxx' not found while handling absolute" warnings
every time I run
compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary_test.py
, both before and after I made the change. Do you see those also?
I think they're worth fixing, but I don't think they actually cause failure...
> ERROR:root:got exception while creating new DiffRecord
> Traceback (most recent call last):
> File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py",
line
> 257, in add_image_pair
> actual_image_locator=actual_image_locator)
> File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py",
line
> 97, in __init__
> expected_image_url)
> File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py",
line
> 361, in _download_and_open_image
> with contextlib.closing(urllib.urlopen(url)) as url_handle:
> File "/usr/lib/python2.7/urllib.py", line 86, in urlopen
> return opener.open(url)
> File "/usr/lib/python2.7/urllib.py", line 207, in open
> return getattr(self, name)(url)
> File "/usr/lib/python2.7/urllib.py", line 460, in open_file
> return self.open_ftp(url)
> File "/usr/lib/python2.7/urllib.py", line 519, in open_ftp
> host = socket.gethostbyname(host)
> IOError: [Errno socket error] [Errno -2] Name or service not known
I think this error is the real problem here. Unfortunately, it does not report
the URL it tried to open. I will create a CL adding that info to the log.
> Traceback (most recent call last):
> File
>
"/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py",
> line 198, in <module>
> options.gm_json_path, options.imagediffdb_path)
> File
>
"/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py",
> line 100, in WriteJsonSummary
> actual_image_locator=checksum_withpatch)
> File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py",
line
> 270, in get_diff_record
> return self._diff_dict[key]
> KeyError: ('1582992237666751297', '15063702165824040060')
get_diff_record() presumably fails because the previous operation (download
image file) failed.
Message was sent while issue was closed.
On 2014/03/04 16:10:48, epoger wrote: > On 2014/03/04 15:05:08, rmistry wrote: > > > I kicked off a skia tryserver run in cluster telemetry. > > > > Failed with: > > Yikes, all sorts of problems here. I'll respond to some of them, inline. > > > == Comparing pictures and saving differences in JSON output file == > > > /home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py:62: > > SyntaxWarning: assertion is always true, perhaps remove parentheses? > > assert (files_to_checksums_nopatch.keys() == > > I don't understand this one, but it seems harmless enough. > > > /home/default/skia-repo/trunk/gm/gm_json.py:15: RuntimeWarning: Parent module > > '/home/default/skia-repo/trunk/gm/gm_json' not found while handling absolute > > I see these "Parent module 'xxx' not found while handling absolute" warnings > every time I run > compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary_test.py > , both before and after I made the change. Do you see those also? > > I think they're worth fixing, but I don't think they actually cause failure... > > > ERROR:root:got exception while creating new DiffRecord > > Traceback (most recent call last): > > File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", > line > > 257, in add_image_pair > > actual_image_locator=actual_image_locator) > > File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", > line > > 97, in __init__ > > expected_image_url) > > File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", > line > > 361, in _download_and_open_image > > with contextlib.closing(urllib.urlopen(url)) as url_handle: > > File "/usr/lib/python2.7/urllib.py", line 86, in urlopen > > return opener.open(url) > > File "/usr/lib/python2.7/urllib.py", line 207, in open > > return getattr(self, name)(url) > > File "/usr/lib/python2.7/urllib.py", line 460, in open_file > > return self.open_ftp(url) > > File "/usr/lib/python2.7/urllib.py", line 519, in open_ftp > > host = socket.gethostbyname(host) > > IOError: [Errno socket error] [Errno -2] Name or service not known > > I think this error is the real problem here. Unfortunately, it does not report > the URL it tried to open. I will create a CL adding that info to the log. That sounds great, thanks! Also, I can show you how to test it end-to-end, it is simple: * Go to http://skia-tree-status.appspot.com/skia-telemetry/skia_try * Paste '--clone 1 --config 8888' in Render Pictures Args (non tiling runs are quicker) * Withpatch compile with Mesa use 'True' * Enter 'No patch no tiling run with mesa' in description. * Click on 'Queue SkiaTry Task'. It will send you an email once the run is finished, the email will contain the output HTML and a link to the logs of the first slave (which is where I got this error snippet from). > > > Traceback (most recent call last): > > File > > > "/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py", > > line 198, in <module> > > options.gm_json_path, options.imagediffdb_path) > > File > > > "/home/default/skia-repo/buildbot/compute_engine_scripts/telemetry/telemetry_slave_scripts/write_json_summary.py", > > line 100, in WriteJsonSummary > > actual_image_locator=checksum_withpatch) > > File "/home/default/skia-repo/trunk/gm/rebaseline_server/imagediffdb.py", > line > > 270, in get_diff_record > > return self._diff_dict[key] > > KeyError: ('1582992237666751297', '15063702165824040060') > > get_diff_record() presumably fails because the previous operation (download > image file) failed.
Message was sent while issue was closed.
On 2014/03/04 16:22:10, rmistry wrote: > > I think this error is the real problem here. Unfortunately, it does not > report > > the URL it tried to open. I will create a CL adding that info to the log. > > That sounds great, thanks! > > Also, I can show you how to test it end-to-end, it is simple: The error log I saw when I reran it is at https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/lo... The important line is: ERROR:root:unable to download expected_image_url file://tmp/nopatch-pictures-epoger-1393951947.19/http___www_google_co_in_.png to file /tmp/images/1582992237666751297.png That file:// URL needs an extra slash, like file:///tmp/nopatch... I'll create a new CL to fix it.
Message was sent while issue was closed.
My most recent skia_try seems like it ran without error: https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/lo... But the difference images don't show up in the HTML report: https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... Ravi, PTAL and let me know how I can help.
Message was sent while issue was closed.
On 2014/03/04 20:25:15, epoger wrote: > My most recent skia_try seems like it ran without error: > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/lo... > > But the difference images don't show up in the HTML report: > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... > > Ravi, PTAL and let me know how I can help. Looks good to me. Click on 'To view all 2367 failing files click here.' And then click on any 'Failed Files' it will take you to a page with the images. Ah, I just noticed something that is not working, compare: https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... with https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... Looks like https://skia.googlesource.com/buildbot/+/master/compute_engine_scripts/teleme... (line 107) is converting the perceptual similarity into perceptual diff (which it no longer needs to).
Message was sent while issue was closed.
On 2014/03/04 21:11:18, rmistry wrote: > On 2014/03/04 20:25:15, epoger wrote: > > My most recent skia_try seems like it ran without error: > > > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/lo... > > > > But the difference images don't show up in the HTML report: > > > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... > > > > Ravi, PTAL and let me know how I can help. > > Looks good to me. > Click on 'To view all 2367 failing files click here.' > And then click on any 'Failed Files' it will take you to a page with the images. What I mean is: If I go to https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... and then click on "To view all 2367 failing files click here." (which takes me to http://tinyurl.com/qbtcd8p ) and then click on the first link ( https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... ) I see the NoPatch image and the WithPatch image, but no image shows up in the "Differing pixels in white" or "Difference per pixel" column.
Message was sent while issue was closed.
On 2014/03/04 21:18:01, epoger wrote: > On 2014/03/04 21:11:18, rmistry wrote: > > On 2014/03/04 20:25:15, epoger wrote: > > > My most recent skia_try seems like it ran without error: > > > > > > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/lo... > > > > > > But the difference images don't show up in the HTML report: > > > > > > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... > > > > > > Ravi, PTAL and let me know how I can help. > > > > Looks good to me. > > Click on 'To view all 2367 failing files click here.' > > And then click on any 'Failed Files' it will take you to a page with the > images. > > What I mean is: > > If I go to > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... > > and then click on "To view all 2367 failing files click here." (which takes me > to http://tinyurl.com/qbtcd8p ) > > and then click on the first link ( > https://storage.cloud.google.com/chromium-skia-gm/telemetry/skia-tryserver/ht... > ) > > I see the NoPatch image and the WithPatch image, but no image shows up in the > "Differing pixels in white" or "Difference per pixel" column. Ah I see, yes I will have to investigate that. Will post findings here. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
