avoid sync messages on document load #2259

the8472 · 2015-08-26T02:13:50Z

This removes the sync greasemonkey:scripts-for-url message from the framescript.

builds on work from PR #2243 and #2255

old: strings sent via e10s message manager are duplicated and the sandbox holds onto them as source code, this causes unnecessary memory overhead. new: load scripts from file:// URIs and use lazy getters for scriptSource

fixes greasemonkey#2103

janekptacijarabaci · 2015-08-26T05:57:25Z

The warning:

Services.ppmm

(Services.mm, Services.cpmm)

Not compatible for Firefox 38.0.5-
https://hg.mozilla.org/mozilla-central/diff/1b3657fee5a8/toolkit/modules/Services.jsm

arantius · 2015-08-26T13:23:07Z

Thanks for all your effort @the8472
Please read http://greasemonkey.github.io/style.html and follow it in your pull requests . (I see lots of missing spaces and braces, and the general style is to prefix function arguments' names with an "a".)
I'd find it much easier to review/understand pull requests if they were independent; this seems to be a mix of some other ones, plus further changes.

What's the overall approach here? Effectively the whole config of all scripts and their include/exclude data is stored in every child all the time? And the fileXhr trick means that we don't have to duplicate the script itself?

the8472 · 2015-08-26T14:45:48Z

The Warning

ok, will fix that

style

Is there some auto-formatter I could apply? That's more a job for a machine than a human.

pull requests review

Once you looked at the other two you only need to review f21e40a

Effectively the whole config of all scripts and their include/exclude data is stored in every child all the time? And the fileXhr trick means that we don't have to duplicate the script itself?

Yes. In each child process, not in each frame.

Additionally, the old code used the sync messages to check for script refreshes, this now piggybacks on http observers which carry similar information to the sync messages anyway. We can extract all the necessary information from requests with TYPE_DOCUMENT/TYPE_SUBDOCUMENT in their loadInfo.

janekptacijarabaci · 2015-09-02T16:15:29Z

The warning:

initialProcessData

Not compatible for Firefox 42- (maybe 41- , https://bugzilla.mozilla.org/show_bug.cgi?id=1162838 , but: 41.0b6 - in this version does not work).

janekptacijarabaci · 2015-09-13T18:27:10Z

modules/ipcscript.js

  this.localized = aScript.localized;
  this.matches = aScript.matches.map(function(m) { return m.pattern; });
+  this.userMatches = aScript.matches.map(function(m) { return m.pattern; });


I guess it should be: matches => userMatches

yes, thanks

the8472 · 2015-09-13T21:37:52Z

@janekptacijarabaci thanks for testing, I think those issues should be fixed now.

janekptacijarabaci · 2015-09-14T05:47:47Z

You're welcome.

However, I see another problem...

After editing the script in Scratchpad (e.g.) and saving - properties of the object (GM_info) is not valid (is not changed):
https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-9f06c50d889eb78e6776c4b4fc7314cbL66

the8472 · 2015-09-14T20:19:15Z

@janekptacijarabaci when I save + refresh the page GM_info.script does update accordingly for me

demo: https://i.pantsu.cat/lemhtw.webm

It's pretty much the same old logic, just that it infers a page load from a HTTP observer instead of relying on sync messages from the frame script: https://github.com/the8472/greasemonkey/blob/f21e40a74147e409231e39b08f9387e55cc0cb07/modules/refererSetter.js#L33-L53

janekptacijarabaci · 2015-09-14T21:20:59Z

I'm sorry, but I do not work for me.
Firefox Nightly 43.0a1 (2015-09-14)
Firefox 40.0.3

Steps to reproduce:

I trying make a new profile of Firefox (but in the default profile too)
Installing Greasemonkey (your version)
Installation of this script: https://gist.github.com/janekptacijarabaci/e4a939351e5792d3ba90
Open Browser Console
Open page github.com in browser - only one page is open (one tab)
5a. In Browser Console: e.g. GM_info.script.version: 1.0
Open Scratchpad, the script editing (@Version 1.0 => @Version 1.1), saving
Refresh github.com in browser
7a. In Browser Console: GM_info.script.version: 1.0

I don't have time to find the bug now...

janekptacijarabaci · 2015-09-15T07:17:17Z

when = "any": this value does not exist...

Fix this bug (for example):

https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-9be6aefe99ccf1e149776242db5940a6R48

From:

GM_util.getService() 
    .scriptRefresh("any", channel.URI.spec, windowId, browser);

To:

GM_util.getService() 
    .scriptRefresh("document-start", channel.URI.spec, windowId, browser);
GM_util.getService() 
    .scriptRefresh("document-end", channel.URI.spec, windowId, browser);

...or:

https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-9be6aefe99ccf1e149776242db5940a6R49

From:

.scriptRefresh("any", channel.URI.spec, windowId, browser);

To:

.scriptRefresh(channel.URI.spec, windowId, browser);

https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-e07419236596eaa414a5954c7c90c385R193

From:

service.prototype.scriptRefresh = function(when, url, windowId, browser) {

To:

service.prototype.scriptRefresh = function(url, windowId, browser) {

https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-e07419236596eaa414a5954c7c90c385R199

From:

this.config.updateModifiedScripts(when, url, windowId, browser);

To:

this.config.updateModifiedScripts("document-start", url, windowId, browser);
this.config.updateModifiedScripts("document-end", url, windowId, browser);

- beam down IPCScripts in advance from parent when a script changes - do script filtering/url matching in content process - piggyback script change detection on http observers instead by checking document/sub-document loads

the8472 · 2015-09-15T22:38:30Z

ok, incorporated your changes and rebased on master

arantius · 2015-09-23T21:41:09Z

I've started merging this down. It's a huge change, I want to give it some time to bake before I commit it to master.

master...arantius:broadcast-script-config

janekptacijarabaci · 2015-09-24T00:51:26Z

@arantius:
In case you missed this file "modules/refererSetter.js" (https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-e07419236596eaa414a5954c7c90c385R104):
#2259 (comment)
#2259 (comment)

@the8472, @arantius:
...and fix typo:
https://github.com/greasemonkey/greasemonkey/pull/2259/files#diff-eed5161116c8f3737a46e5dd4ac7fe8fR196
master...arantius:broadcast-script-config#diff-eed5161116c8f3737a46e5dd4ac7fe8fR188
fileName => e.fileName

arantius · 2015-09-24T02:36:09Z

Point one: no, intentional, merge conflict with c274845 .

Point two: good catch!

arantius · 2015-09-24T13:30:27Z

See #2289 -- doesn't work with e10s off. Looking into why now.

arantius · 2015-09-24T13:37:59Z

Also not working in FF 44 with e10s off, so e10s off appears to be the real issue.

the8472 · 2015-09-24T13:44:07Z

Odd, I did test with e10s off, maybe I skipped that during the recent rebase.

Anyway, how do you want fixes? New commits on this PR? Rebase? Or against your own branch?

arantius · 2015-09-24T13:58:09Z

Gimme 5, I might be near a "fix". Something at appears to work better, assuming I understand the operating model.

Refs greasemonkey#2259

arantius · 2015-09-24T14:08:02Z

I think 73736ed is the right fix for #2289 and some related issues.

the8472 · 2015-09-24T14:21:29Z

Hum, was that some merge conflict? The broadcast shouldn't have been in an else block. It doesn't exist in my branch

arantius · 2015-09-24T14:27:22Z

Indeed, must have been my fault somehow. Whoops, there it is.

the8472 · 2015-09-24T15:16:29Z

In case you missed this file "modules/refererSetter.js"
Point one: no, intentional, merge conflict with c274845 .

What about the script refresh part?

arantius · 2015-09-24T15:47:31Z

I didn't know it existed because it was in a file called refererSetter.js. Add it back in a place that makes a bit more sense please?

janekptacijarabaci · 2015-09-25T01:49:36Z

The reminder (a typo):
#2259 (comment)

Refs greasemonkey#2259

janekptacijarabaci · 2015-09-27T18:56:09Z

The suggestion:
arantius#3

arantius · 2015-09-30T13:44:44Z

Merged this into master.

the8472 added 2 commits August 21, 2015 02:47

avoid holding duplicate script source strings in memory

70a16c4

old: strings sent via e10s message manager are duplicated and the sandbox holds onto them as source code, this causes unnecessary memory overhead. new: load scripts from file:// URIs and use lazy getters for scriptSource

Support GM_xmlhttpRequest referer override in e10s

e4a38a3

fixes greasemonkey#2103

arantius added this to the 3.5 milestone Aug 26, 2015

janekptacijarabaci reviewed Sep 13, 2015
View reviewed changes

fix xray access denied error

06349bf

the8472 added 2 commits September 15, 2015 23:56

Merge branches 'e10s-referer' and 'dedup_string' into async-messages2

0bfa0ef

reduce sync messages from frame scripts

70fab3e

- beam down IPCScripts in advance from parent when a script changes - do script filtering/url matching in content process - piggyback script change detection on http observers instead by checking document/sub-document loads

the8472 force-pushed the async-messages2 branch from 4f0dd3f to 70fab3e Compare September 15, 2015 22:37

arantius mentioned this pull request Sep 24, 2015

No script execution at all #2289

Closed

arantius added a commit to arantius/greasemonkey that referenced this pull request Sep 24, 2015

Service broadcastScriptUpdates(): set initial AND broadcast.

73736ed

Refs greasemonkey#2259

arantius mentioned this pull request Sep 24, 2015

avoid duplication of source strings #2243

Closed

the8472 mentioned this pull request Sep 24, 2015

piggyback script refresh checking on http requests for page/frame loads arantius/greasemonkey#2

Merged

arantius added a commit to arantius/greasemonkey that referenced this pull request Sep 25, 2015

Fix filename reference typo.

1c58833

Refs greasemonkey#2259

arantius closed this Sep 30, 2015

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

avoid sync messages on document load #2259

avoid sync messages on document load #2259

the8472 commented Aug 26, 2015

janekptacijarabaci commented Aug 26, 2015

arantius commented Aug 26, 2015

the8472 commented Aug 26, 2015

janekptacijarabaci commented Sep 2, 2015

janekptacijarabaci Sep 13, 2015

the8472 Sep 13, 2015

the8472 commented Sep 13, 2015

janekptacijarabaci commented Sep 14, 2015

the8472 commented Sep 14, 2015

janekptacijarabaci commented Sep 14, 2015

janekptacijarabaci commented Sep 15, 2015

the8472 commented Sep 15, 2015

arantius commented Sep 23, 2015

janekptacijarabaci commented Sep 24, 2015

arantius commented Sep 24, 2015

arantius commented Sep 24, 2015

arantius commented Sep 24, 2015

the8472 commented Sep 24, 2015

arantius commented Sep 24, 2015

arantius commented Sep 24, 2015

the8472 commented Sep 24, 2015

arantius commented Sep 24, 2015

the8472 commented Sep 24, 2015

arantius commented Sep 24, 2015

janekptacijarabaci commented Sep 25, 2015

janekptacijarabaci commented Sep 27, 2015

arantius commented Sep 30, 2015

avoid sync messages on document load #2259

avoid sync messages on document load #2259

Conversation

the8472 commented Aug 26, 2015

janekptacijarabaci commented Aug 26, 2015

arantius commented Aug 26, 2015

the8472 commented Aug 26, 2015

janekptacijarabaci commented Sep 2, 2015

janekptacijarabaci Sep 13, 2015

Choose a reason for hiding this comment

the8472 Sep 13, 2015

Choose a reason for hiding this comment

the8472 commented Sep 13, 2015

janekptacijarabaci commented Sep 14, 2015

the8472 commented Sep 14, 2015

janekptacijarabaci commented Sep 14, 2015

janekptacijarabaci commented Sep 15, 2015

when = "any": this value does not exist...

the8472 commented Sep 15, 2015

arantius commented Sep 23, 2015

janekptacijarabaci commented Sep 24, 2015

arantius commented Sep 24, 2015

arantius commented Sep 24, 2015

arantius commented Sep 24, 2015

the8472 commented Sep 24, 2015

arantius commented Sep 24, 2015

arantius commented Sep 24, 2015

the8472 commented Sep 24, 2015

arantius commented Sep 24, 2015

the8472 commented Sep 24, 2015

arantius commented Sep 24, 2015

janekptacijarabaci commented Sep 25, 2015

janekptacijarabaci commented Sep 27, 2015

arantius commented Sep 30, 2015