You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@netbeans.apache.org by "Gili (JIRA)" <ji...@apache.org> on 2018/01/02 11:34:00 UTC

[jira] [Comment Edited] (NETBEANS-241) Migrate Exception Reporter to Apache

    [ https://issues.apache.org/jira/browse/NETBEANS-241?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16307906#comment-16307906 ] 

Gili edited comment on NETBEANS-241 at 1/2/18 11:33 AM:
--------------------------------------------------------

Issue collectors might work but I believe we're missing a middle piece (#2 below). I'm aware of 3 pieces of the puzzle:

1. Analytics collection in the IDE.
2. Report tracker (as opposed to issue tracker which follows). Each report corresponds to a single occurrence of a bug in a user's IDE.
3. Issue tracker. Each issue corresponds to one or more reports in step 2.

The Reporter Tracker in step 2 consolidates related reports and does some intelligent things like:

1. Multiple reports with similar stack-traces get redirected to the same issue.
2. Only create a new issue once a bug has been reported multiple times, ideally by different users. Some reports are promoted into issues right away (e.g. NPEs) while others (IOExceptions) are not. I assume this is because some exceptions are more likely to be caused by user error than others (e.g. out of memory, out of diskspace, etc)
3. If a report corresponds to an issue that is marked as fixed, but the reporter is running a newer version of the IDE than the version the issue should be fixed in then the Exception Reporter automatically reopens the bug report with the relevant information.
4. If a report corresponds to an issue that is marked as fixed, but the reporter is running an older version of the IDE, the Exception Reporter points the user to the newer version so he can update.

So the question is: where does this piece sit? I don't believe JIRA has something to answer this question, though obviously nothing prevents us from wrapping this logic in a JIRA plugin. If we do so, other Apache (and non-Apache) projects stand to gain.


was (Author: cowwoc):
Issue collectors might work but I believe we're missing a middle piece (#2 below). I'm aware of 3 pieces of the puzzle:

1. Analytics collection in the IDE.
2. Report tracker (as opposed to issue tracker which follows). Each report corresponds to a single occurrence of a bug in a user's IDE.
3. Issue tracker. Each issue corresponds to one or more reports in step 2.

The Reporter Tracker in step 2 consolidates related reports and does some intelligent things like:

1. Multiple reports with similar stack-traces get redirected to the same issue.
2. Only create a new issue once a bug has been reported multiple times, ideally by different users. Some reports are promoted into issues right away (e.g. NPEs) while others (IOExceptions) are not. I assume this is because some exceptions are more likely to be caused by user error than others (e.g. out of memory, out of diskspace, etc)
3. If a report corresponds to an issue that is marked as fixed, but the reporter is running a newer version of the IDE than the version the issue should be fixed in then the Exception Reporter automatically reopens the bug report with the relevant information.
4. If a report corresponds to an issue that is marked as fixed, but the reporter is running an older version of the IDE, the Exception Reporter points the user to the newer version so he can update.

> Migrate Exception Reporter to Apache
> ------------------------------------
>
>                 Key: NETBEANS-241
>                 URL: https://issues.apache.org/jira/browse/NETBEANS-241
>             Project: NetBeans
>          Issue Type: Improvement
>          Components: ide - Logger
>            Reporter: Gili
>
> Expected behavior: Exception Reporter should file bugs in the new Apache JIRA issue tracker instead of the old Bugzilla system.
> It's not clear (to me at least) what metadata you need to store outside of individual bug reports, so that needs to be fleshed out because Apache might have provide the necessary infrastructure to host it.
> This issue is related to NETBEANS-164



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@netbeans.apache.org
For additional commands, e-mail: commits-help@netbeans.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists