db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lily Wei (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DERBY-4787) Track tasks needed to release Derby 10.6.2
Date Thu, 09 Sep 2010 23:09:33 GMT

    [ https://issues.apache.org/jira/browse/DERBY-4787?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12907817#action_12907817

Lily Wei commented on DERBY-4787:

Hi Kathey:
     Sorry,    I did not read your comment before I send the RELEASE-NOTES.html I created
yesterday. Thank you so much for helping me to get to this far on the release process. I couldn't
do it without you.
     The JSR section is not in the RELEASE-NOTES.html after Knut's comment and they had been
removed from the source code.

     As far as changing back to use .xml format, I am not sure how to change the ReleaseNoteGenerator.java
back to use .xml  format. I will take a look at the 10.5 branch code. At this point, I think
it is more constructive for me to focus on how to improve the release process. However, I
am not trying to say either .html or .xml format is better. I am focus on how to improve the
current generating a release process. 

    Overall, I was able to create the RELEASE-NOTES.html in a timely fashion. However, there
are two steps which I would suggest to spend more time to improve.

1.	Manual process to extract description if the JIRA issue has parent id.
How to solve it so it is more programmatic?

Right now,
The tag looks like this if JIRA issue has parent id:
                            <a href="http://issues.apache.org/jira/browse/DE
RBY-2925"> <img src="fixedBugsList1_files/issue_subtask.gif" alt="Sub-task" titl
e="Sub-task - The sub-task of the issue" width="16" align="absmiddle" border="0"
        <td class="nav issuekey">

    <a href="http://issues.apache.org/jira/browse/DERBY-2925">DERBY-2925</a>

        <td class="nav summary">
                                                    <a href="http://issues.apach
e.org/jira/browse/DERBY-2437" class="parentIssue" title="SYSCS_EXPORT_TABLE can
be used to overwrite derby files">DERBY-2437</a><br>
                            <img src="fixedBugsList1_files/link_out_bot.gif" alt
="" width="16" align="absmiddle" border="0" height="16">
                        <a href="http://issues.apache.org/jira/browse/DERBY-2925
" style="text-decoration: none;">Prevent export from overwriting existing files<

Change to:

                                <a href="http://issues.apache.org/jira/browse/DE
RBY-2925"> <img src="http://issues.apache.org/jira/images/icons/issue_subtask.gi
f" height="16" width="16" border="0" align="absmiddle" alt="Sub-task" title="Sub
-task - The sub-task of the issue"></a>
        <td class="nav issuekey">

    <a href="http://issues.apache.org/jira/browse/DERBY-2925">DERBY-2925</a>

        <td class="nav summary">
                        <a href="http://issues.apache.org/jira/browse/DERBY-2925
" style="text-decoration: none; ">Prevent export from overwriting existing files

so ParseKey() will extract the right description for JIRA like this when running genrelnotes
target in tools/release directory.

This is the ParseKey method in RelaseNoteGenerator.java on 10.6 branch:
        public String parseKey( TagReader tr ) throws Exception
            tr.position( "issues.apache.org/jira/browse/DERBY-", true );
            String keyString = tr.getUpTill( "\">", true );
            return keyString;
The pattern is fine with JIRA issue that does not have extra tag info to extract the key as
description. I cannot see a better pattern to extract the key than the one we already have.
Any suggestion on how to improve it is welcome. Rick suggest people have some ideas about

2.	I am regenerating a new .xml format and grep for releaseNote.html to find out the JIRA
that require release note for Issues are in the release note. 

For the current data I have for, I got the following from the .xml file:
$ grep releaseNote.html fixedBugsListxml.xml
            <attachment id="12454149" name="releaseNote.html" size="3891" author
="lilywei" created="Wed, 8 Sep 2010 16:11:35 -0400 (EDT)" />
            <attachment id="12447876" name="releaseNote.html" size="3891" author
="mamtas" created="Wed, 23 Jun 2010 16:17:43 -0400 (EDT)" />
            <attachment id="12447606" name="releaseNote.html" size="3979" author
="mamtas" created="Mon, 21 Jun 2010 12:38:06 -0400 (EDT)" />
            <attachment id="12447055" name="releaseNote.html" size="4018" author
="mamtas" created="Mon, 14 Jun 2010 15:10:16 -0400 (EDT)" />
                    <comment id="12584134" author="rhillegas" created="Tue, 1 Ap
r 2008 09:33:30 -0400 (EDT)"  >Re-attaching release note as releaseNote.html sin
ce this is the file name which the release notes generator looks for.</comment>
            <attachment id="12379029" name="releaseNote.html" size="4770" author
="rhillegas" created="Tue, 1 Apr 2008 09:33:30 -0400 (EDT)" />

>From there, I confirm the result id and put it into my getReleaseNoteAttachmentID method
in ReportParse.java.

Luckily, there are only two JIRA issue that need release note information on

I also think we can have more programmatic way to gather release note information on issue.
Any suggestion is welcome. Thanks!

> Track tasks needed to release Derby 10.6.2
> ------------------------------------------
>                 Key: DERBY-4787
>                 URL: https://issues.apache.org/jira/browse/DERBY-4787
>             Project: Derby
>          Issue Type: Task
>          Components: Miscellaneous
>    Affects Versions:
>            Reporter: Lily Wei
>            Assignee: Lily Wei
>         Attachments: DERBY-4787_patch_for_wronglinkadminguild.diff, DERBY-4787_patch_for_wronglinkadminguild.diff,
DERBY-4787_patch_for_wronglinkadminguild.diff, RELEASE-NOTES.html, RELEASE-NOTES.html, RELEASE-NOTES.html,
> Track tasks for 10.6.2 release

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message