1338395 - Update pdf.js to version 1.7.290

Status: RESOLVED FIXED

(Firefox :: PDF Viewer, task)

Description

[Ryan VanderMeulen [:RyanVM]]

+++ This bug was initially created as a clone of Bug #1336418 +++

Changes since last update:
#8023 Moves preprocessor stuff to the gulpfile.
#8025 Adjust the `brace-style` ESLint rule to disallow single lines (and also enable `no-iterator`)
#8027 [Firefox addon] Remove the unused `MOZ_CENTRAL` constant
#8002 [api-minor] Fix #7798: Refactor scratch canvas usage.
#8030 Interactive forms: rewrite AcroForms example
#8028 Prevent browser console errors during testing
#8035 [api-minor] Add a `getDocument` parameter that allows disabling of the `NativeImageDecoder` (e.g. for use with Node.js)
#8036 [api-minor] Fixes behaviour of DOMCanvasFactory to return {canvas, context}.
#8038 Moves sed processing into the preprocessing tasks and merging tasks.
#8039 Moving interactive examples to jsfiddle.net
#8040 Update localization files
#8041 Interactive forms: set the `buttonValue` for radio buttons that do not have a `fieldValue`
#8043 Allow automatic print rotation via the enablePrintAutoRotate preference
#8046 Replacing custom bundling with webpack 2

Comment 1

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/e9d5d057344d
Update pdf.js to version 1.7.274. r=bdahl

Comment 2

[Ryan VanderMeulen [:RyanVM]]

I attempted to land this and ended up backing out due to changes in where the localization files were placed in the tree. It wasn't clear to me that it was an intentional change.
https://treeherder.mozilla.org/logviewer.html#?job_id=76170941&repo=mozilla-inbound&lineNumber=1976

https://hg.mozilla.org/integration/mozilla-inbound/rev/1b290796502d

Comment 3

[Francesco Lodolo [:flod]]

(In reply to Ryan VanderMeulen [:RyanVM] from comment #2)
> I attempted to land this and ended up backing out due to changes in where
> the localization files were placed in the tree. It wasn't clear to me that
> it was an intentional change.

Just noticed it too when looking at the first landing. 

This would require us to move files around on merge day for all languages, otherwise we would lose all existing translations. And I would still need to coordinate with one of our tools, just to make sure they don't lose them anyway.

Comment 4

[Ryan VanderMeulen [:RyanVM]]

Yeah, I think the change was accidental.

Comment 5

[Ryan VanderMeulen [:RyanVM]]

Changes since last update:
#8053 [Chromium addon] Prevent errors that break the addon, caused by the `DEFAULT_URL` constant being replaced by a `defaultUrl` viewer configuration parameter (PR 8046 follow-up)
#8054 Write the l10n files to the correct destination for `gulp mozcentral` builds (PR 8023 follow-up)

Comment 6

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/f884a597f402
Update pdf.js to version 1.7.278. r=bdahl

Comment 7

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Backed out for timing out in browser_pdfjs_main.js, at least on Linux with e10s:

https://hg.mozilla.org/integration/mozilla-inbound/rev/8b706863e9726f58f1fb34e2e16d5cf93d6e95e5

Push with failures: https://treeherder.mozilla.org/#/jobs?repo=mozilla-inbound&revision=9cf238ae7ef7ee3cdd5f83778ec729bf6fbf250c&filter-resultStatus=testfailed&filter-resultStatus=busted&filter-resultStatus=exception&filter-resultStatus=retry&filter-resultStatus=usercancel&filter-resultStatus=runnable
Failure log: https://treeherder.mozilla.org/logviewer.html#?job_id=76415653&repo=mozilla-inbound

[task 2017-02-10T18:47:45.335226Z] 18:47:45     INFO - TEST-START | browser/extensions/pdfjs/test/browser_pdfjs_main.js
[task 2017-02-10T18:48:30.405161Z] 18:48:30     INFO - TEST-INFO | started process screentopng
[task 2017-02-10T18:48:32.064282Z] 18:48:32     INFO - TEST-INFO | screentopng: exit 0
[task 2017-02-10T18:48:32.064711Z] 18:48:32     INFO - Buffered messages logged at 18:47:45
[task 2017-02-10T18:48:32.064923Z] 18:48:32     INFO - Entering test bound test
[task 2017-02-10T18:48:32.066989Z] 18:48:32     INFO - TEST-PASS | browser/extensions/pdfjs/test/browser_pdfjs_main.js | pdf handler defaults to always-ask is false - 
[task 2017-02-10T18:48:32.068483Z] 18:48:32     INFO - TEST-PASS | browser/extensions/pdfjs/test/browser_pdfjs_main.js | pdf handler defaults to internal - 
[task 2017-02-10T18:48:32.073190Z] 18:48:32     INFO - Pref action: 3
[task 2017-02-10T18:48:32.075310Z] 18:48:32     INFO - Buffered messages finished
[task 2017-02-10T18:48:32.079239Z] 18:48:32     INFO - TEST-UNEXPECTED-FAIL | browser/extensions/pdfjs/test/browser_pdfjs_main.js | Test timed out - 
[task 2017-02-10T18:48:32.082574Z] 18:48:32     INFO - MEMORY STAT vsizeMaxContiguous not supported in this build configuration.
[task 2017-02-10T18:48:32.084515Z] 18:48:32     INFO - MEMORY STAT | vsize 1679MB | residentFast 212MB | heapAllocated 87MB
[task 2017-02-10T18:48:32.090552Z] 18:48:32     INFO - TEST-OK | browser/extensions/pdfjs/test/browser_pdfjs_main.js | took 45141ms
[task 2017-02-10T18:48:32.092582Z] 18:48:32     INFO - Not taking screenshot here: see the one that was previously logged
[task 2017-02-10T18:48:32.094803Z] 18:48:32     INFO - TEST-UNEXPECTED-FAIL | browser/extensions/pdfjs/test/browser_pdfjs_main.js | Found a tab after previous test timed out: about:blank - 
[task 2017-02-10T18:48:32.101563Z] 18:48:32     INFO - checking window state

And more, please check the log.

Comment 8

[Ryan VanderMeulen [:RyanVM]]

I'm working on bisecting this now, but I strongly suspect that the root cause here is that pdfjschildbootstrap.js is somehow no longer being packaged, which of course breaks e10s mode. Yury, I'm thinking this is fallout from the gulpfile preprocessor changes?

Comment 9

[Ryan VanderMeulen [:RyanVM]]

As expected, this goes back to rev 2049cf0.

Comment 10

[Yury Delendik (:yury)]

Fixed by https://github.com/mozilla/pdf.js/pull/8067

Comment 11

[Ryan VanderMeulen [:RyanVM]]

Changes since last update:
#8058 Fixes pdf.combined.js for webpack
#8056 Use `ChildNode.remove` instead of `ChildNode.ParentNode.removeChild` in a couple of places (bug 1334831, issue 8008)
#8060 Font ascent descent calculation fix
#8062 Improve the README by removing outdated information
#8063 Consume the current character when encountering illegal characters in `Lexer.getObject`, in order to prevent infinite loops during reading of streams (issue 8061)
#8067 Include the `pdfjschildbootstrap.js` file in the output for `gulp mozcentral` builds (PR 8023 follow-up)

Comment 12

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/e28772b44058
Update pdf.js to version 1.7.290. r=bdahl

Comment 13

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/e28772b44058