<feed xmlns='http://www.w3.org/2005/Atom'>
<title>platform_tools_base/bazel/coverage, branch studio-master-dev</title>
<subtitle>Unnamed repository; edit this file 'description' to name the repository.
</subtitle>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/'/>
<entry>
<title>Add coverage target for parallel device test report.</title>
<updated>2020-05-19T18:08:50+00:00</updated>
<author>
<name>Arif Sukoco</name>
<email>arifs@google.com</email>
</author>
<published>2020-05-18T19:07:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=8532277e4d6535d3323e8214971961d578043e20'/>
<id>8532277e4d6535d3323e8214971961d578043e20</id>
<content type='text'>
Test: manually run coverage/report.sh
Bug: None

Change-Id: I9c7cc1cbb04624dc683c22bab2017272a770ec38
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Test: manually run coverage/report.sh
Bug: None

Change-Id: I9c7cc1cbb04624dc683c22bab2017272a770ec38
</pre>
</div>
</content>
</entry>
<entry>
<title>Prune coverage master ignore list to only necessary items</title>
<updated>2020-05-08T06:24:38+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-05-05T23:18:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=810b499a1dc7ccfe054c584cbbec5fe8b65518b9'/>
<id>810b499a1dc7ccfe054c584cbbec5fe8b65518b9</id>
<content type='text'>
The use of baseline sources takes care of a lot of filtering for us
(test code/data, external projects that don't use our custom rules,
etc). This lets us drop a lot from the master filter without changing
what is filtered in practice.

Test: Checked that filtered merged baseline sources is the same
Bug: 153913236
Change-Id: I107b625db373f2cd5db49a62d90e18bdc6b343d7
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The use of baseline sources takes care of a lot of filtering for us
(test code/data, external projects that don't use our custom rules,
etc). This lets us drop a lot from the master filter without changing
what is filtered in practice.

Test: Checked that filtered merged baseline sources is the same
Bug: 153913236
Change-Id: I107b625db373f2cd5db49a62d90e18bdc6b343d7
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge baseline LCOV and use in all coverage reports</title>
<updated>2020-05-08T04:35:50+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-05-03T04:36:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=fa50dc42616906ce28dc667184fda538725f3830'/>
<id>fa50dc42616906ce28dc667184fda538725f3830</id>
<content type='text'>
Merge all individual baseline LCOVs into a single baseline LCOV, which
then is used as an additional input to all coverage report definitions.

This ensures that all coverage eligible files that meet the
include/exclude rules of a coverage report definition are actually
included in the report. Previously if some file was eligible, but never
actually touched by a test then it would be omitted due to lack of data.
With this change, such omissions will be detected and included in the
report as a file with zero coverage.

Change-Id: I1a4b14ebf8f75e342786cb7917e3503d4e95207e
Test: N/A
Bug: 153913236
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Merge all individual baseline LCOVs into a single baseline LCOV, which
then is used as an additional input to all coverage report definitions.

This ensures that all coverage eligible files that meet the
include/exclude rules of a coverage report definition are actually
included in the report. Previously if some file was eligible, but never
actually touched by a test then it would be omitted due to lack of data.
With this change, such omissions will be detected and included in the
report as a file with zero coverage.

Change-Id: I1a4b14ebf8f75e342786cb7917e3503d4e95207e
Test: N/A
Bug: 153913236
</pre>
</div>
</content>
</entry>
<entry>
<title>Use baseline filelist in results XML-&gt;LCOV conversion step</title>
<updated>2020-05-08T01:26:04+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-05-03T03:44:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=d9b73406660e6abf6c13cee1f68506bc124d418f'/>
<id>d9b73406660e6abf6c13cee1f68506bc124d418f</id>
<content type='text'>
This removes the master ignore list logic from the XML-LCOV conversion
ending the earlier duplication. Now the converter is provided a list of
coverage eligible files and drops any data from ineligible files.

As a side bonus, this enables the converter to run remotely removing the
local execution bottleneck that previously slowed down coverage
post-processing.

Test: N/A
Bug: 153913236
Change-Id: I254a74a6ba33fb5cbfa28f7b22af027d220c5481
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This removes the master ignore list logic from the XML-LCOV conversion
ending the earlier duplication. Now the converter is provided a list of
coverage eligible files and drops any data from ineligible files.

As a side bonus, this enables the converter to run remotely removing the
local execution bottleneck that previously slowed down coverage
post-processing.

Test: N/A
Bug: 153913236
Change-Id: I254a74a6ba33fb5cbfa28f7b22af027d220c5481
</pre>
</div>
</content>
</entry>
<entry>
<title>Revert Change Ic17e5fdd</title>
<updated>2020-05-08T01:00:00+00:00</updated>
<author>
<name>Dana Dahlstrom</name>
<email>dahlstrom@google.com</email>
</author>
<published>2020-05-08T00:00:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=3188ccc2ab9fea5e0abd5f47cec5f57b390b2f43'/>
<id>3188ccc2ab9fea5e0abd5f47cec5f57b390b2f43</id>
<content type='text'>
… following Change Ibbb7f1e0.

Change-Id: Ie50e05f32f7e66a63498fc10c901416a57a2d7bd
Issue: 155929139
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
… following Change Ibbb7f1e0.

Change-Id: Ie50e05f32f7e66a63498fc10c901416a57a2d7bd
Issue: 155929139
</pre>
</div>
</content>
</entry>
<entry>
<title>Find and merge all baseline coverage sourcefile lists</title>
<updated>2020-05-08T00:18:19+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-05-03T05:59:52+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=51157d603c2bceda3863d6464d2c7c095fcfa5c2'/>
<id>51157d603c2bceda3863d6464d2c7c095fcfa5c2</id>
<content type='text'>
Uses the WORKSPACE loop trick to find all baseline source file targets
by looking for their ourput and constructs the merger rule with all
the rules as the sources.

We cannot express such a dependency in Starlark normally due to a mix of
factors (no wildcards in BUILD files prevents using a genquery to find
the rules and even if we could do that we couldn't use the output file
to populate the `srcs` of a genrule anyway).
However, by using a WORKSPACE loop into `bazel-bin` to glob for files
that look like the rule outputs we can identify all the rules and
construct the merger genrule correctly.
The downside is that we need to do a targeted remove of the
rule outputs to ensure we don't pick up anything from deleted targets,
but we can mechanically enforce that in both places we invoke coverage
processing (tools/base/bazel/studio_coverage.sh and
tools/base/bazel/coverage/report.sh).

This is similar to using a script to construct a BUILD file and then
invoking a target in that BUILD file, but this method does not pollute
the workspace filetree.

Change-Id: I7132af7aea7f8b2a65b393801ca7962453cd02a9
Test: N/A
Bug: 153913236
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Uses the WORKSPACE loop trick to find all baseline source file targets
by looking for their ourput and constructs the merger rule with all
the rules as the sources.

We cannot express such a dependency in Starlark normally due to a mix of
factors (no wildcards in BUILD files prevents using a genquery to find
the rules and even if we could do that we couldn't use the output file
to populate the `srcs` of a genrule anyway).
However, by using a WORKSPACE loop into `bazel-bin` to glob for files
that look like the rule outputs we can identify all the rules and
construct the merger genrule correctly.
The downside is that we need to do a targeted remove of the
rule outputs to ensure we don't pick up anything from deleted targets,
but we can mechanically enforce that in both places we invoke coverage
processing (tools/base/bazel/studio_coverage.sh and
tools/base/bazel/coverage/report.sh).

This is similar to using a script to construct a BUILD file and then
invoking a target in that BUILD file, but this method does not pollute
the workspace filetree.

Change-Id: I7132af7aea7f8b2a65b393801ca7962453cd02a9
Test: N/A
Bug: 153913236
</pre>
</div>
</content>
</entry>
<entry>
<title>Duplicate coverage master ignore logic into standalone filter</title>
<updated>2020-05-07T23:23:18+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-05-03T03:39:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=e7c0e4bf64511c4319994f84385045471c887b8c'/>
<id>e7c0e4bf64511c4319994f84385045471c887b8c</id>
<content type='text'>
This lets us use the master ignore list in other places (specifically
for baseline coverage). This is just a temporary duplication. The logic
in the XML-&gt;LCOV conveter will be removed in the course of the baseline
coverage integration. I'm just splitting it out like this to make
following the process across the commit chain easier.

Test: N/A
Bug: 153913236
Change-Id: Ic0f9f85c1aefa992f60b3a2e07a3dddf10a227af
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This lets us use the master ignore list in other places (specifically
for baseline coverage). This is just a temporary duplication. The logic
in the XML-&gt;LCOV conveter will be removed in the course of the baseline
coverage integration. I'm just splitting it out like this to make
following the process across the commit chain easier.

Test: N/A
Bug: 153913236
Change-Id: Ic0f9f85c1aefa992f60b3a2e07a3dddf10a227af
</pre>
</div>
</content>
</entry>
<entry>
<title>Remove failing android-ndk_tests</title>
<updated>2020-05-07T03:00:00+00:00</updated>
<author>
<name>Dana Dahlstrom</name>
<email>dahlstrom@google.com</email>
</author>
<published>2020-05-07T02:00:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=e50e05f32f7e66a63498fc10c901416a57a2d7bd'/>
<id>e50e05f32f7e66a63498fc10c901416a57a2d7bd</id>
<content type='text'>
Change-Id: Ic17e5fdd934e2ca1de8e90577bf4010981b088ae
Issue: 155929139
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Change-Id: Ic17e5fdd934e2ca1de8e90577bf4010981b088ae
Issue: 155929139
</pre>
</div>
</content>
</entry>
<entry>
<title>Add test tags to coverage data in results processing</title>
<updated>2020-05-06T00:15:50+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-03-14T00:53:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=b63973ff9e491bdf089e75e084ca25bb434d0719'/>
<id>b63973ff9e491bdf089e75e084ca25bb434d0719</id>
<content type='text'>
Refactoring that splits lcov merger and test tagging functionality
so that the merger function can be reused in other stages.

Change-Id: Idb63a629a47c3c5dccc83214839acd7f6d98e0a3
Test: N/A
Bug: 153913236
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Refactoring that splits lcov merger and test tagging functionality
so that the merger function can be reused in other stages.

Change-Id: Idb63a629a47c3c5dccc83214839acd7f6d98e0a3
Test: N/A
Bug: 153913236
</pre>
</div>
</content>
</entry>
<entry>
<title>Remove leading "./" in LCOV and master ignore list</title>
<updated>2020-05-05T23:19:05+00:00</updated>
<author>
<name>Eric Reed</name>
<email>ericreed@google.com</email>
</author>
<published>2020-05-03T01:57:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.replicant.us/mirrors/AOSP/platform_tools_base/commit/?id=90359a4a03231ca19cf605eb8acc508d8fa616f7'/>
<id>90359a4a03231ca19cf605eb8acc508d8fa616f7</id>
<content type='text'>
This refactoring makes processing filepaths a bit simpler, which will
help keep things cleaner when adding baseline coverage

Test: N/A
Bug: N/A
Change-Id: I0924b6aba0fc488db0a08bddd2be5fa762bd2e5f
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This refactoring makes processing filepaths a bit simpler, which will
help keep things cleaner when adding baseline coverage

Test: N/A
Bug: N/A
Change-Id: I0924b6aba0fc488db0a08bddd2be5fa762bd2e5f
</pre>
</div>
</content>
</entry>
</feed>
