CLOSED TREE
Backed out changeset 2d42350d209a (bug 1203330)
Backed out changeset 3a12c51c3eca (bug 1203330)
Backed out changeset 31fac390e15d (bug 1203330)
And remove redundant `Promise.resolve()` because it is the default
for async functions.
setIcon is not supported on Android, so there was no need to change
mobile/android/components/extensions/ext-pageAction.js.
MozReview-Commit-ID: 94ebaJFxLAi
--HG--
extra : rebase_source : 20466181501b264ba33fc8ab61fdf2bed20f9eef
- Add `envType` to BaseContext.
- Pass an explicit envType to all `registerSchemaAPI` invocations.
- The factories passed to `registerSchemaAPI` will be split up later, so
that content scripts (`content_child`) and addon pages can share
common implementations.
- The factories that implement the addon API will also be split up,
to separate code running in the main process (`addon_parent`) from
code running in a child process (`addon_child`).
- Remove the use of a hardcoded list of `namespaces` from ProxyContext.
Now `envType` is used to specify whether an API should be activated.
MozReview-Commit-ID: Jiff8HIwG92
--HG--
extra : rebase_source : 946a3c0009a4e3223c2d10044b3099a94c845394
ExtensionContext in Extension.jsm has |extension| as an instance member,
so use it instead of passing |extension| to registerSchemaAPI's
callback.
And to make sure that this pattern also works in content processes, move
the |extension| member to BaseContext.
MozReview-Commit-ID: BgsGGCPQxJR
--HG--
extra : rebase_source : 7aa9fb7a53e057e8d3d8c477bd6821f8344c571a
ExtensionContext in Extension.jsm has |extension| as an instance member,
so use it instead of passing |extension| to registerSchemaAPI's
callback.
And to make sure that this pattern also works in content processes, move
the |extension| member to BaseContext.
MozReview-Commit-ID: BgsGGCPQxJR
--HG--
extra : rebase_source : 9a85d95f0c147198938b86c57127f6e7403fd53c
This version addresses some popup sizing bugs, and also a few other issues I
ran into when debugging Blake's problems:
* The standalone popup needs a max width of 800px for Chrome compatibility,
which is wider than our default max width.
* I added a flex attribute to our browser so that it fills the entire space
of the slide-in panel. This is only necessary for browsers with content
that is shorter than the height of the panel when it gets its desired
width, but becomes longer when it doesn't, so it didn't show up in my
initial tests.
* I also added an extra pixel to the width calculations, since I noticed that
a lot of single lines of text were unexpectedly wrapping without it. I'll
look into this more in a follow-up bug.
I also added some comments, and renamed a couple of variables, where things
seemed unclear.
The test changes are mostly just updates to older browser action tests to use
newer helpers, rather than ad-hoc events, to open/close/click the widgets. A
few tests also needed updates to explicitly close the panel when they were
done with it.
--HG--
extra : commitid : BhHv1aZSrBL
extra : rebase_source : 6fe0069ebd2f94b0ffd02ad757a799fbdff9226e
extra : source : 628af985c7eba7e4665a1b62f9a60f6a6fd39822
This brings our behavior closer in line with Chrome's.
--HG--
extra : commitid : DaDofvmQEZW
extra : rebase_source : cbeda3889362511c5002f2710c063a2c83047d3c
extra : source : 7ad8b56958c54787313ecbe60eb1e8fbafb2f376
This version addresses some popup sizing bugs, and also a few other issues I
ran into when debugging Blake's problems:
* The standalone popup needs a max width of 800px for Chrome compatibility,
which is wider than our default max width.
* I added a flex attribute to our browser so that it fills the entire space
of the slide-in panel. This is only necessary for browsers with content
that is shorter than the height of the panel when it gets its desired
width, but becomes longer when it doesn't, so it didn't show up in my
initial tests.
* I also added an extra pixel to the width calculations, since I noticed that
a lot of single lines of text were unexpectedly wrapping without it. I'll
look into this more in a follow-up bug.
I also added some comments, and renamed a couple of variables, where things
seemed unclear.
The test changes are mostly just updates to older browser action tests to use
newer helpers, rather than ad-hoc events, to open/close/click the widgets. A
few tests also needed updates to explicitly close the panel when they were
done with it.
--HG--
extra : commitid : 9OyRnaMTqn8
extra : rebase_source : efd28a49ca5961e2cf545420a81bd19b796d8957
This brings our behavior closer in line with Chrome's.
--HG--
extra : commitid : 7X5knx0CtGm
extra : rebase_source : b301e9c568df0edb313b10cf06e2f7a9c23a6e78
The base .eslintrc is essentially a merge of the root Toolkit .eslintrc and
the devtools .eslintrc, with some minor changes to match our prevalent style.
For the most enforces the coding styles that we've been using most
consistently. There are a couple of significant differences, though:
* The rule for opening brace alignment can only be applied globally, and
doesn't make exceptions for top-level functions. I chose to turn it on, and
change the brace style of existing top-level functions that violated it,
since the rule seemed worth using, and that's the direction most Toolkit JS
code has been headed anyway.
* The rule for switch/case statements requires an added indentation level for
case statements. Most of our switch statements did not use an extra level
of indentation, and I initially wrote the rule to enforce that style, until
I came across case statements that used blocks, and required the extra
indentation level for sanity.
--HG--
rename : devtools/.eslintrc => toolkit/components/extensions/.eslintrc
extra : commitid : 4aDZOi5MNPc
extra : rebase_source : f34ba43aeb8046a0c971cb76b76410497bd12b42
extra : source : 2a7eb4ef7b375e5fd52d843ca206a071367a6475
