EngineeringProductivity/Projects/CodeCoverage: Difference between revisions
Jump to navigation
Jump to search
Klahnakoski (talk | contribs) (begin!) |
Klahnakoski (talk | contribs) (→Code Coverage: add note about updates) |
||
| (6 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
=Code Coverage= | =Code Coverage= | ||
March 1st, 2017 - This project is active, but this page is stale. Email me after Mar 8th if this message is still here: mailto://klahnakoski@mozilla.org | |||
==Introduction== | ==Introduction== | ||
| Line 6: | Line 8: | ||
The objective of the Code Coverage project is to provide daily code coverage details and summary statistics. The assumption is that | The objective of the Code Coverage project is to provide daily code coverage details and summary statistics. The assumption is that | ||
code coverage statistics and details can help judge code quality. We expect this information to be useful to | code coverage statistics and details can help judge code quality. We expect this information to be useful to | ||
* Release managers - Provide them with another metric to help measure risk a particular changeset, or | * Release managers - Provide them with another metric to help measure risk a particular changeset, or family of changesets, adds to a release | ||
* Module owners - Provide drill down coverage statistics so test coverage is focused on the most important part of the code | * Module owners - Provide drill down coverage statistics so test coverage is focused on the most important part of the code | ||
* Individual developers - To see if their new code | * Individual developers - To see if their new code require new tests to cover it, or if tests already cover their new code | ||
* Our own test-infrastructure team - We can reduce the number of machines dedicated to testing, if we know where tests overlap, and focus on tests that cover recent changes. | * Our own test-infrastructure team - We can reduce the number of machines dedicated to testing, if we know where tests overlap, and focus on tests that cover recent changes. | ||
| Line 15: | Line 17: | ||
* If code coverage is going down, what new code is not being covered? Is it OK to not be covered? | * If code coverage is going down, what new code is not being covered? Is it OK to not be covered? | ||
* If code coverage is going down, what tests have been dropped? Is that OK? | * If code coverage is going down, what tests have been dropped? Is that OK? | ||
* Should we alert when code coverage down "significantly"? | * Should we alert when code coverage goes down "significantly"? | ||
* When adding or change code in source file S, what tests cover that source file? What tests should we run? | * When adding or change code in source file S, what tests cover that source file? What tests should we run? | ||
| Line 26: | Line 28: | ||
* GUI objective - http://people.mozilla.org/~choller/firefox/coverage/mc-coverage-20130210-29dd80c95b7d/ | * GUI objective - http://people.mozilla.org/~choller/firefox/coverage/mc-coverage-20130210-29dd80c95b7d/ | ||
* | ==Code== | ||
* '''UI''' - https://github.com/chinhodado/codecoverage_presenter | |||
* '''ETL''' - https://github.com/klahnakoski/ActiveData-ETL/blob/dev/activedata_etl/transforms/jscov_to_es.py | |||
==Contacts== | |||
* jmaher@mozilla.com | |||
* klahnakoski@mozilla.com | |||
* gmierz1@live.ca | |||
* chmanchester@mozilla.com | |||
Many of us are live at #ateam @ irc.mozilla.org during usual North American work hours. | |||
==Reference docs == | ==Reference docs == | ||
Latest revision as of 01:05, 2 March 2017
Code Coverage
March 1st, 2017 - This project is active, but this page is stale. Email me after Mar 8th if this message is still here: mailto://klahnakoski@mozilla.org
Introduction
The objective of the Code Coverage project is to provide daily code coverage details and summary statistics. The assumption is that code coverage statistics and details can help judge code quality. We expect this information to be useful to
- Release managers - Provide them with another metric to help measure risk a particular changeset, or family of changesets, adds to a release
- Module owners - Provide drill down coverage statistics so test coverage is focused on the most important part of the code
- Individual developers - To see if their new code require new tests to cover it, or if tests already cover their new code
- Our own test-infrastructure team - We can reduce the number of machines dedicated to testing, if we know where tests overlap, and focus on tests that cover recent changes.
Generally, we would like to know
- Is code coverage going up or down?
- If code coverage is going down, what new code is not being covered? Is it OK to not be covered?
- If code coverage is going down, what tests have been dropped? Is that OK?
- Should we alert when code coverage goes down "significantly"?
- When adding or change code in source file S, what tests cover that source file? What tests should we run?
Getting Started
- Getting Started with Code Coverage - https://public.etherpad-mozilla.org/p/coverage_quickstart
User Interface
- GUI objective - http://people.mozilla.org/~choller/firefox/coverage/mc-coverage-20130210-29dd80c95b7d/
Code
- UI - https://github.com/chinhodado/codecoverage_presenter
- ETL - https://github.com/klahnakoski/ActiveData-ETL/blob/dev/activedata_etl/transforms/jscov_to_es.py
Contacts
- jmaher@mozilla.com
- klahnakoski@mozilla.com
- gmierz1@live.ca
- chmanchester@mozilla.com
Many of us are live at #ateam @ irc.mozilla.org during usual North American work hours.
Reference docs
The following is optional reading, for reference purposes only.
- Previous term CodeCoverage meeting notes (optional, reference only) - https://public.etherpad-mozilla.org/p/ucosp_meeting_notes