-
Notifications
You must be signed in to change notification settings - Fork 3.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ARROW-17322: [Docs] Documenting issue lifecycle for bugs and feature requests #13781
ARROW-17322: [Docs] Documenting issue lifecycle for bugs and feature requests #13781
Conversation
the issue and starting progress. Issues in this state should be unassigned. | ||
* **In progress** - At the time a contributor self-assigns an issue, the status | ||
should be set to In progress by clicking the **Start progress** button. All | ||
issues in this status should have an assignee - unassigned issues will be |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
NOTE: This details a newly proposed project policy - that any issues that have a status of "In Progress" but lack an assignee should be reverted to "Open" status. This prevents issues from being misrepresented as "In Progress" when there is nobody actually working the issue. This is likely to be automated, and while manual inspection of an issue in such a state may indicate somebody is likely working the issue (e.g., adding comments, etc.) but neglected (or cannot) self-assign, that may not be recognized in an automated process. Any automation should include a comment when changing the status from "In Progress" to "Open" for this reason, guiding any affected users on how to self-assign if appropriate.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
+1
* Implemented | ||
* Done | ||
|
||
* **Closed** - Another terminal status, Closed indicates the issue is complete, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
NOTE: Resolution codes have not been consistently applied historically. See this analysis for more details, and to comment on proposal to standardized and clean existing resolution code data. The resolution codes listed here eliminate several codes that have been previously used, and aims to standardize usage for clarity.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Right, the problem here is there are too many possible choices with no one being obviously better than the others in some situations :-)
The attached .zip file contains the complete HTML page to help with review: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @toddfarmer ! Minor comment from me
`create an ASF Jira account here <https://issues.apache.org/jira/secure/Signup!default.jspa>`_. | ||
Self-assigning issues to work requires | ||
the "Contributor" permission. This permission is set by project maintainers, | ||
and may be requested by emailing the dev mailing list. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we add the full email address? As a new user reporting an issue it might not be clear where to find the dev mailing list.
and may be requested by emailing the dev mailing list. | |
and may be requested by emailing the `dev@arrow.apache.org <https://mail-archives.apache.org/mod_mbox/arrow-dev/>`_ mailing list. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good idea! I changed this to point to the mailing list section of the communication page, just because I'm not sure of the mechanics of a non-subscriber emailing the dev@ list (will it go through? will they see any reply?).
Could you open a JIRA issue for this?
|
Happy to, thanks for the prompt: ARROW-17322. |
dd6d2e0
to
fe5de40
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
+1
Co-authored-by: Sutou Kouhei <[email protected]>
Co-authored-by: Sutou Kouhei <[email protected]>
Co-authored-by: Sutou Kouhei <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'll merge this after a few days to wait for comments from others.
Co-authored-by: Sutou Kouhei <[email protected]>
closed, including during review of pull requests. | ||
* **Resolved** - This is a terminal status indicating action has been taken | ||
on the issue, which is now considered completed. Issues in a resolved status | ||
may have the following resolution codes set: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should really always use "Fixed" for consistency, IMHO.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is that true for non-defect issues (e.g., feature requests)? If a status of "Resolved" should map to a resolution of "Fixed" 100% of the time, that's easy enough to automate (in addition to document).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, it is. The merge script does that. Besides, I don't think "Implemented" or "Done" would convey any useful subtlety compared to "Fixed".
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hmm, I'm not entirely sure what to make of ARROW-16790 in this context. The PR was merged, but the resolution is "Resolved" instead of "Fixed".
I've updated this page to direct usage of "Fixed" when resolving an issue. I scanned this list for examples where that's not set, and it seems to mostly be manually-set resolutions, either using a status of "Resolved" when it should be more properly be "Closed" (e.g., "Information Provided" or "Won't Fix" resolutions), or because the issue was resolved by work done on a different issue (and the PR referenced the other issue, not the one with the incorrect resolution).
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ARROW-16790 should ideally be marked "Fixed". But it seems that would require reopening the issue... let's not bother with this one perhaps.
Co-authored-by: Antoine Pitrou <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Two small comments regarding the self-assignment of an issue.
`create an ASF Jira account here <https://issues.apache.org/jira/secure/Signup!default.jspa>`_. | ||
Self-assigning issues to work requires | ||
the "Contributor" permission. This permission is set by project maintainers, | ||
and may be requested by emailing the :ref:`development mailing list <mailing_list>`. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There is no need for a contributor permission to self-assign Jira issues anymore, if I understand correctly.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated to reflect this change - thanks!
Co-authored-by: Antoine Pitrou <[email protected]>
Co-authored-by: Alenka Frim <[email protected]>
…equired for Resolved status
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the updates @toddfarmer !
Benchmark runs are scheduled for baseline = c90d07f and contender = d53bce2. d53bce2 is a master commit associated with this PR. Results will be available as each benchmark for each run completes. |
['Python', 'R'] benchmarks have high level of regressions. |
PR apache#13781 moved back the testing submodule to an old changeset by mistake, which subsequently broke multiple CI builds. (added a dummy C++ change to trigger CI)
PR #13781 moved back the testing submodule to an old changeset by mistake, which subsequently broke multiple CI builds. (added a dummy C++ change to trigger CI) Authored-by: Antoine Pitrou <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
…requests (apache#13781) These changes aim to document the issue lifecycle - how to create new issues (largely already existed) as well as how to engage with and understand issue state later on. This also serves to document proposed processes: 1. Assigned issues idle more than 90 days may be unassigned (discussed and approved on ML) 2. Unassigned issues in "In Progress" status should be set to "Open" 3. Establishing consistent usage of Status and Resolution fields. Authored-by: Todd Farmer <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
PR apache#13781 moved back the testing submodule to an old changeset by mistake, which subsequently broke multiple CI builds. (added a dummy C++ change to trigger CI) Authored-by: Antoine Pitrou <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
…requests (apache#13781) These changes aim to document the issue lifecycle - how to create new issues (largely already existed) as well as how to engage with and understand issue state later on. This also serves to document proposed processes: 1. Assigned issues idle more than 90 days may be unassigned (discussed and approved on ML) 2. Unassigned issues in "In Progress" status should be set to "Open" 3. Establishing consistent usage of Status and Resolution fields. Authored-by: Todd Farmer <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
PR apache#13781 moved back the testing submodule to an old changeset by mistake, which subsequently broke multiple CI builds. (added a dummy C++ change to trigger CI) Authored-by: Antoine Pitrou <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
PR apache#13781 moved back the testing submodule to an old changeset by mistake, which subsequently broke multiple CI builds. (added a dummy C++ change to trigger CI) Authored-by: Antoine Pitrou <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
These changes aim to document the issue lifecycle - how to create new issues (largely already existed) as well as how to engage with and understand issue state later on. This also serves to document proposed processes: