1087940 - Make ro.build.version.incremental available through settings

Status: RESOLVED FIXED

(Core :: General, defect)

Description

[Leila]

This is to add support for adding incremental build number in Settings app.

Comment 1

[Leila]

Attachment: Bug_1087940-Make-ro.build.version.incremental-available-through.patch

Comment 2

[Leila]

Attachment: Make ro.build.version.incremental available through settings

This patch should be built together with gecko patch.

Comment 3

[Leila]

This path is to add incremental "Build Number" on Settings app.

deviceinfo has been used in gaia-ui-tests and ftu as following:

tests/python/gaia-ui-tests/gaiatest/tests/functional/settings/test_settings_device_info.py
tests/python/gaia-ui-tests/gaiatest/apps/settings/regions/device_info.py
gaia/apps/system/js/ftu_ping.js
apps/system/test/unit/ftu_ping_test.js

should I add the build_number to test and ftu in this bug or a new bug?

Comment 4

[Leila]

Attachment: https://github.com/mozilla-b2g/gaia/pull/25456

Comment 5

[[:fabrice] Fabrice Desré]

Why do we need that? Note that you are adding strings, and we are way past string freeze so we can't take that on 2.1.

Comment 6

[Leila]

Hi Fabrice,

This field is used to specify our SW version and we want to have the SW version visible in the settings, both for our end users and for internal users. Today we only see the Firefox OS version, gecko version and a build identifier which none of these can be set by us.
We are aware that it's late for 2.1, thus we are fine with getting it to 2.2(later version).

Thanks

Comment 7

[Arthur Chen [:arthurcc][inactive after 6/30]]

Comment on attachment 8510371 [details] [review]
https://github.com/mozilla-b2g/gaia/pull/25456

Newly added strings are only required to be added in the english language file. L10n team will work on the other languages.

Comment 8

[Arthur Chen [:arthurcc][inactive after 6/30]]

Comment on attachment 8510371 [details] [review]
https://github.com/mozilla-b2g/gaia/pull/25456

It seems that the build number does not always exist. Could you add logic determining the visibility of the item and related unit tests? Thanks.

Comment 9

[Leila]

Running gaia on top of gonk, "ro.build.version.incremental" is always set to $BUILD_NUMBER
(in android tree: build/tools/buildinfo.sh), which in turn is set to "eng.$(USER).$(shell date +%Y%m%d.%H%M%S)" (in build/core/version_defaults.mk), if nothing else is provided (we can always set BUILD_NUMBER as environment variable), so the build number should always exists. 
I will add the patch for unit test. Thanks.

Comment 10

[Arthur Chen [:arthurcc][inactive after 6/30]]

Shawn, regarding comment9, could you provide inputs from the aspect of the build process? Thanks!

Comment 11

[shawn ku [:sku]]

build/tools/buildinfo.sh shows "ro.build.version.incremental" is always there. Hence I agree with comment 9 to have a test case for preventing this.

Comment 12

[Arthur Chen [:arthurcc][inactive after 6/30]]

Comment on attachment 8510371 [details] [review]
https://github.com/mozilla-b2g/gaia/pull/25456

Thanks Leila and Shawn. The settings part looks good to me. Flagging Zac for reviewing the gaia ui tests.

Comment 13

[Zac C (:zac)]

Comment on attachment 8510371 [details] [review]
https://github.com/mozilla-b2g/gaia/pull/25456

r+, thanks, but can you land the Gecko patch before the Gaia patch? 
As otherwise of course, then the tests will fail until the Gecko patch matches :)

Comment 14

[Leila]

Fabrice, Can you help landing gecko patch before gaia? Thanks.

Comment 15

[[:fabrice] Fabrice Desré]

https://hg.mozilla.org/integration/b2g-inbound/rev/4b6a82a644b5

Comment 16

[Ryan VanderMeulen [:RyanVM]]

checkin-needed isn't needed for uplifts. We have separate queries for it.

Comment 17

[Ryan VanderMeulen [:RyanVM]]

Oh nevermind, it was for the Gaia patch. In the future, a bit of context to the request is appreciated :)

Master: https://github.com/mozilla-b2g/gaia/commit/8dc75432e27aa532faffc415b16f409e230ea80b

Comment 18

[Leila]

Thanks Ryan.

Comment 19

[Ryan VanderMeulen [:RyanVM]]

Argh, this bug somehow never got resolved when the Gecko patch landed on mozilla-central. Sorry for the huge delay here :(

https://hg.mozilla.org/releases/mozilla-b2g34_v2_1/rev/ff62b8fb461b

v2.1: https://github.com/mozilla-b2g/gaia/commit/88c44f2243a5ca1683587aca9faf29023974b96c

Comment 20

[Francesco Lodolo [:flod]]

Why are we pushing strings to a branch that has been string frozen for months?

Comment 21

[Ryan VanderMeulen [:RyanVM]]

Crap. I guess that means we should backout at least the Gaia patch? Not sure if the Gecko patch is worth keeping in by itself or not.

Comment 22

[[:fabrice] Fabrice Desré]

(In reply to Ryan VanderMeulen [:RyanVM UTC-5] from comment #21)
> Crap. I guess that means we should backout at least the Gaia patch? Not sure
> if the Gecko patch is worth keeping in by itself or not.

It's fine to keep it. No one will be harmed ;)

Comment 23

[Ryan VanderMeulen [:RyanVM]]

Backed out the Gaia change. Sorry for the churn.
v2.1: https://github.com/mozilla-b2g/gaia/commit/9ec3e0ad324105ad74e7489a7644f14ea8eab497

Comment 24

[Vincent Liu[:vliu]]

https://hg.mozilla.org/releases/mozilla-b2g34_v2_1s/rev/ff62b8fb461b