HomeConfig.java saved a list of events to be sent later in a batch. This
patch makes it save a pair of strings instead, and the strings are later
used to make calls to GeckoAppShell.
The patch also makes two small optimizations. It makes the queue an
ArrayList instead of a LinkedList to save memory. It also makes copying
the queue a swap instead of a true copy.
On a CLOSED TREE -- Android and automation only.
MozReview-Commit-ID: AU8bt4CDC1V
--HG--
extra : amend_source : ff83d4b0513102abf095949a79bf7f7616bbfa14
extra : histedit_source : 576ce07d4f7d2ea9d1ae1997736cfe59d751be6d
The initial --with-gradle support disabled building the Android test
directories; everything was built from
mobile/android/app/build.gradle. That doesn't declare support files
that need to be packaged for Robocop tests. This patch stops building
instrumentation test APKs, which aren't used in automation under any
circumstance and which aren't packaged when building with Gradle; and
avoids building the Robocop APK by tweaking the Makefile. That gets
support files in place while not using moz.build in place of Gradle.
I would have declared the support files elsewhere, but there are path
requirements that I couldn't make work, so in robocop/ they stay.
MozReview-Commit-ID: KCpXvqzYBsY
--HG--
extra : rebase_source : b99be8fada7787ee473f68265824cca2250c70a1
extra : histedit_source : fa19d3ede51e14707f400ab8527d44f5bf550f85
This also adds a GRADLE_FLAGS environment variable for use in
automation.
Manually tested.
MozReview-Commit-ID: 8nDkqz2VnJn
--HG--
extra : rebase_source : 32626a7dc0c0a6a440e300d92c31670f14319325
extra : amend_source : fe134e25f079851b4c648b53a7a485ee20c15c18
Bug 1242213 removed the entire <activity-alias>. Sadly, users who
added the Firefox icon to their dock (for example, Samsung's Touchwiz
dock) will see the icon disappear when they upgrade, because the
intent filter disappears. (That is, the icon is connected to .App and
action MAIN, not to the package and action MAIN.)
This patch restores the .App <activity-alias> for action MAIN. It
doesn't add the launcher and other categories, which could lead to
multiple launcher icons. New users that add the Firefox icon to their
dock will use .BrowserApp, but sadly we'll need to maintain this alias
essentially forever to support existing dock icons.
MozReview-Commit-ID: 1o9XS5MEs1s
--HG--
extra : rebase_source : 8f4e1321da475bf2dcfca88c5807bf26c940b1c5
extra : amend_source : 3429ff0eaec0edb02e5e47678252bf5a2de74d3f
This permission was added in API 16, and is only enforced in API 19+.
(It's benign to add it to APKs installed on API 15.)
We want to declare it explicitly so that the bouncer APK and the main
APK have the same permission set. There appears to be some fanciness
with Gradle's implied permission system where the bouncer APK does not
request READ implicitly where-as the main APK does request READ
implicitly. This just makes things explicit (and uniform).
MozReview-Commit-ID: LBblAPa0duV
--HG--
extra : rebase_source : 616fa58ccd9063088abb14240265a20ac4e48da7
extra : histedit_source : abf6211896e99cc7b53f9c36714d056f0d44d63f
Pinned sites should be deleted directly, however I'm not confident enough
in my knowledge of sync to be certain that we won't end up with deleted
pinned sites in our table.
(We use normal bookmark deletion for removing pinned sites.)
MozReview-Commit-ID: SSLDkSXWlI
This fixes an edge case that is most likely to happen to new users if they pin a site followed
by removing one or more suggested sites. This results in the topsites table containing
less sites than needed, leading to some pinned sites being displayed in a higher
than expected position. This also broke unpinning since our code assumes
that a topsites physical position corresponds to its DB position (which prior
to this patch was not the case).
MozReview-Commit-ID: JgTUa55eXnz
This variable was renamed to be positive instead of negative but the value/usage wasn't adapted.
MozReview-Commit-ID: H1DeZMmvTXC
--HG--
extra : amend_source : a83d5ae3f1c7525031c17cdb4fa0f4950f6aa7c2
The battery level can decrease while we're trying to charge. Most commonly
many devices oscillate between 99% and 100% when fully charged. This situation
could also happen if the device is receiving less power than it's using. Therefore
we shouldn't treat this as an extraordinary situation.
MozReview-Commit-ID: eN3W9P5oty
--HG--
extra : amend_source : abd0231d8514a9bbe9d8ad21ed2de7dba68bbc91
The ordering in this commit is now correct, whereas the previously pushed
commit hadn't addressed the review comment.
MozReview-Commit-ID: Ga5lPgwpIxp
--HG--
extra : source : ca8d1e1caeefdcd45bd608adec7a10c3632ceab5
extra : amend_source : 39a2ef44130138eb664d4af679ffe49433749de7
I managed to land the unfixed version - we should be closing the cursor before adding
a LeakCanary watcher, or else we have a greater probability of a race condition.
MozReview-Commit-ID: JuP15Tkj1hz