mozilla :: #pdfjs

18 Apr 2017
18:00yurySnuffleupagus: hi, where we stopped with annotations?
18:01Snuffleupagusyury: Hi, I don't know what the status is, I've not done much work on that part of the code (mostly just helped review some stuff).
18:02SnuffleupagusOn a different note, it seems that pdfjs-dist have stopped updating.
18:03Snuffleupagus(However, the gh-pages branch still seem to be updating as it should.)
18:04yurySnuffleupagus: we need to trying migrating minified from make.js
18:05yuryI looked at the task manager at one point and saw two node.js processes spiking during dist
18:06yuryI'm thinking make.js' exec has to do something with it
18:08SnuffleupagusThat's certainly a possibility.
18:09yuryor the way we interacting between them
18:10SnuffleupagusIt seems that there's also a Babel minifier, https://github.com/babel/babili, that we could perhaps use instead of Uglify.
18:12* yury does not think minifying is our job
18:12yuryprobably 'dist' and not 'minifiedpost'
18:13yurydist has lots of exec()s
18:15yurywe need to stabilize what we have atm
18:54* Snuffleupagus sighs and submits PR #8311
19:01yurySnuffleupagus: let's try https://github.com/mozilla/pdf.js/pull/8312 and see if it will fix timeouts
19:03yurybtw, webpack can perform minification automatically using uglify -- we just need to refactor it in the future
19:04Snuffleupagusyury: Nice, I'll take a look
19:12yuryI wonder if robwu_nl can help to migrate signchromium for us (I don't know how to test it)
19:19github_pdfjs[pdf.js] yurydelendik pushed 2 new commits to master: https://github.com/mozilla/pdf.js/compare/5ad3611cc436...bc20181a1e10
19:19github_pdfjspdf.js/master cf87c9b Jonas Jenwald: [Firefox addon] Upstream changes from: Bug 1356569 - Remove optional trailing parameters (issue 8310)...
19:19github_pdfjspdf.js/master bc20181 Yury Delendik: Merge pull request #8311 from Snuffleupagus/bug-1356569...
19:19soakbotPull 8311: [Firefox addon] Upstream changes from: Bug 1356569 - Remove optional trailing parameters (issue 8310). https://github.com/mozilla/pdf.js/pull/8311
19:30yurybdahl: I think make.js is broken on botio
19:31yuryit is wasting CPU, on_push.js shows 50% usage
19:31github_pdfjs[pdf.js] Snuffleupagus pushed 2 new commits to master: https://github.com/mozilla/pdf.js/compare/bc20181a1e10...19321937da8c
19:31github_pdfjspdf.js/master 1d30cea Yury Delendik: Moves dist and minifiedpost tragets from make.js.
19:31github_pdfjspdf.js/master 1932193 Jonas Jenwald: Merge pull request #8312 from yurydelendik/mv-dist-gulpfile...
19:31soakbotPull 8312: Moves dist and minifiedpost tragets from make.js.. https://github.com/mozilla/pdf.js/pull/8312
19:38github_pdfjs[pdf.js] pdfjsbot force-pushed gh-pages from ec14581 to 6fea1c0: https://github.com/mozilla/pdf.js/commits/gh-pages
19:38github_pdfjspdf.js/gh-pages 6fea1c0 pdfjsbot: gh-pages site created via gulpfile.js script...
19:50* yury replaced child.exec to child.execSync
19:57github_pdfjs[pdf.js] pdfjsbot force-pushed gh-pages from 6fea1c0 to dd54a3b: https://github.com/mozilla/pdf.js/commits/gh-pages
19:57github_pdfjspdf.js/gh-pages dd54a3b pdfjsbot: gh-pages site created via gulpfile.js script...
19:58yuryit's about 15min for gulp web
20:18yurybdahl: I wonder how something that takes 35s locally runs 15min on vm
20:21bdahlyury: different node versions?
20:21yuryand 20 mins for gulp dist
20:22yuryironically, exactly the same
20:22yury:)
20:23bdahlmake.js versions?
20:23yurybdahl: also I found out that node.js/gulp use only one core
20:23yurybdahl: now we run gulp web/dist without make.js
20:23yuryfew tasks left to migrate
20:26bdahlyury: same gulp versions?
20:27yuryyep
20:27yuryCLI on my computer older
20:27yuryit's fishy
20:28yurybdahl: is access to the drive suppose to be slow?
20:29bdahlnot that i know of
20:31yurythe gulp web/dist uses all CPU it can, so it's bound to it
20:31yurywebpack+babel
20:31yurymemory looks fine
20:32bdahlyury: are you locally testing on windows?
20:32yuryno looking at botio
20:33yuryif it says 8gb, does it mean it all "physical" memory?
20:42bdahlyury: where does it say 8gb?
20:42bdahlyury: also have you tried running the commands not through botio
20:42yurytask manager
20:43yuryon windows, no?
20:43yuryno
20:45bdahlyury: as for memory, that's how much the vm has, who knows how much underlying real machine has
20:52yurybdahl: as for linux botio, firefox refusing to start testing (even unit testing)
20:54bdahlyury: any idea if it's an ff change or from es6?
20:54yurymaybe, but there is not proof
21:00yuryso we are at Firefox 54.0a2 at botio-linuc
21:04yurybdahl: tried on macos
21:05yuryno problems there
21:05yurycan there be a problem with linux version?
21:07yurydid we update it receintly?
21:18bdahlyury: i haven't touch anything for awhile
21:18yuryokay
21:18yuryall is strange
22:08robwu_nlyury: What do you mean by migrating signchromium?
22:13yuryrobwu_nl: https://github.com/mozilla/pdf.js/blob/master/make.js#L165
22:13yurywe are trying to move to gulp
22:14yuryit's probably simple, but needs to be tested
22:31robwu_nlyury: I don't use that FYI. To publish the extension, I publish an unsigned zip file to the Chrome Web Store
22:32yuryso removing that?
22:32robwu_nlAnd considering that crx files cannot easily be installed in Chrome (at least not easier than zip files), I'd say that it can safely be removed
22:33yuryokay, thanks
22:33yuryless dead code
19 Apr 2017
No messages
   
Last message: 11 days and 8 hours ago