CLJ tickets needing work on May 24 2012

[Andy Fingerhut]

Tickets needing work for CLJ project
Date: May 24, 2012

If you are a Clojure contributor, the tickets below could use some
work. Some are easy, some are not. If you've never added a patch
before, or want a refresher on the mechanics, see (Note 1) near the
bottom.

For the list of tickets that have patches, but they need updating,
keep reading below.

For the list of tickets that don't have any patches, see (Note 2) near
the bottom.

----------------------------------------------------------------------

Patches should be changed to desired git format:

CLJ-980 clojure--extend-type-doc-fix.diff

Patches below used to apply cleanly to latest master as of the date
given, but not with latest master as of the date of this report.

<none>


The patches below would be prescreened, except that their authors are
not on the list of Clojure contributors. The authors have been
contacted via email to ask if they are willing to sign CAs.

CLJ-450 clj-450-add-default-pred-arg-to-core-fns-patch.txt
CLJ-884 diff.patch
CLJ-980 clojure--extend-type-doc-fix.diff
CLJ-983 proxy_super.patch


The ones above are likely easy to update. The ones below are likely
more work, where part of the work could be figuring out what ought to
be done in the first place.


These have patches, and are marked with Fix Version of Release 1.5.
Rich has left comments on the existing patches describing changes he'd
like to see before including them.
CLJ-850 clj-850-type-hinted-fn-abstractmethoderror-patch3.txt


What should be done with these?
CLJ-703 (neither patch applies cleanly)
CLJ-840 (patch clj840-2.diff applies cleanly and passes tests)


Comments perhaps imply that authors or other contributors think more
work should be done to create new patches. Some of these might be
significant hunks of work. Not easy for me to tell.
CLJ-2 (only patch fails to apply cleanly)
CLJ-5 (one of two patches applied and built cleanly on May 4, 2012, but not on May 10, 2012)
CLJ-322 (patch compile-interop-1.patch applies and builds cleanly)
CLJ-445 (no patches apply cleanly)
CLJ-666 (only patch fails to apply cleanly)
CLJ-735 (patch protocolerr.diff applies and builds cleanly)
CLJ-792 (patch clj-792-reorg-reflector-patch2.txt applied cleanly in Feb 2012, but didn't pass all tests)
CLJ-803 (both patches apply and build cleanly, iatom.patch using patch(1))
CLJ-821 (only patch fails to apply cleanly)
CLJ-891 (only patch applies and builds cleanly)
CLJ-993 (several patches, lots of comments back and forth)
CLJ-994 (some tests pass on JDK 1.7 but fail on JDK 1.6 due to NoClassDefFoundError on jsr166y/ForkJoinTask)

----------------------------------------------------------------------

(Note 1)

Read the documentation under the headings "Development", "Adding
patches", and "Removing patches" on this page:

http://dev.clojure.org/display/design/JIRA+workflow

(Note 2)

As of the date of this report, there are 136 tickets matching a JIRA
filter called "Needs a Patch (all)". To see them, go to this link:

http://dev.clojure.org/jira/secure/ManageFilters.jspa#filterView=search

In the box labeled Author, type in Andy Fingerhut. Click on the
choice that pops up and the contents should change to my user account
name, jafingerhut. Click on the Search button at the bottom. Then
click on the filter called "Needs a Patch (all)". You can subscribe
to the filter, after which it should show up in your list of filters
under the Issues menu on this page:

http://dev.clojure.org/jira/browse/CLJ

Most "Needs a Patch" tickets have no patches attached. If it does
have a patch, then its Patch attribute needs to be changed. Read the
documentation on adding patches at the link in (Note 1) above.

Some of the older tickets have links to patches in the old Assembla
ticket system. If you add an attachment to the JIRA ticket, my
automated patch prescreening code will pick it up the next time I run
it (about once per week).

[Michel Alexandre Salim]

On 05/25/2012 10:45 AM, Andy Fingerhut wrote:
> Tickets needing work for CLJ project
> Date: May 24, 2012
>

[snip]

> As of the date of this report, there are 136 tickets matching a JIRA
> filter called "Needs a Patch (all)". To see them, go to this link:
>
> http://dev.clojure.org/jira/secure/ManageFilters.jspa#filterView=search
>

Should CLJ-118 be closed? It refers to a problem with the 1.0 release
(the problem is not fixed in the commit tagged 1.0 in Git, but on the
other hand it's for a version no one uses anymore).

http://dev.clojure.org/jira/browse/CLJ-118


Thanks,

--
Michel Alexandre Salim
�blog: http://identi.ca/hircus
http://twitter.com/hircus
GPG key ID: A36A937A

() ascii ribbon campaign - against html e-mail
/\ www.asciiribbon.org - against proprietary attachments

[Steve Miner]

On May 26, 2012, at 10:08 AM, Michel Alexandre Salim wrote:

> Should CLJ-118 be closed? It refers to a problem with the 1.0 release
> (the problem is not fixed in the commit tagged 1.0 in Git, but on the
> other hand it's for a version no one uses anymore).
>
> http://dev.clojure.org/jira/browse/CLJ-118

I think it should be closed, but I'm just a contributor not a screener. The JIRA workflow document seems to say that only screeners should be closing tickets.

http://dev.clojure.org/display/design/JIRA+workflow

I'd like to suggest that any contributor be allowed to close obsolete or duplicate tickets. (I think JIRA will actually let me close it -- it's just a question of following protocol.)

[Andy Fingerhut]

I may have overstepped my bounds as a mere contributor, but I have closed tickets as duplicate that I was sure were duplicates, with a very explicit comment about what I was doing and why, e.g.:

http://dev.clojure.org/jira/browse/CLJ-998
http://dev.clojure.org/jira/browse/CLJ-920

It is even better if you can get agreement from the person that created the ticket that it is a duplicate first, e.g.:

http://dev.clojure.org/jira/browse/CLJ-930


For closing tickets that we believe are obsolete, I have just added a section titled "Tickets that might be closable without any code changes" to my weekly "prescreened patch list" targeted at Clojure screeners. So far it contains only CLJ-118 and CLJ-820. I would personally feel more comfortable if these were closed by a screener.

The difference: Marking a ticket as a duplicate still leaves open another ticket that will continue to track the issue in the future, whereas closing a ticket that has no duplicates does not.

Andy

[Michel Alexandre Salim]

Hi Andy,

On 05/27/2012 09:15 AM, Andy Fingerhut wrote:
>
> For closing tickets that we believe are obsolete, I have just added a section titled "Tickets that might be closable without any code changes" to my weekly "prescreened patch list" targeted at Clojure screeners. So far it contains only CLJ-118 and CLJ-820. I would personally feel more comfortable if these were closed by a screener.
>
> The difference: Marking a ticket as a duplicate still leaves open another ticket that will continue to track the issue in the future, whereas closing a ticket that has no duplicates does not.
>

In JIRA, is there a way, when closing a ticket as a duplicate, to
indicate which ticket it duplicates? I'm used to Bugzilla where such
information is noted in both the closed ticket and the ticket it's
merged into.

[Andy Fingerhut]

The only way I am aware of is to write something in the comment. Strings of the form CLJ-<number> in comments appear to be automatically turned into links to that ticket. That creates a reference from the bug marked as a duplicate to the original.

If you also want a reference from the original to the duplicate, you can add a comment to the original ticket as well.

Andy

On May 28, 2012, at 5:07 AM, Michel Alexandre Salim wrote:

> Hi Andy,
>
> On 05/27/2012 09:15 AM, Andy Fingerhut wrote:
>>
>> For closing tickets that we believe are obsolete, I have just added a section titled "Tickets that might be closable without any code changes" to my weekly "prescreened patch list" targeted at Clojure screeners. So far it contains only CLJ-118 and CLJ-820. I would personally feel more comfortable if these were closed by a screener.
>>
>> The difference: Marking a ticket as a duplicate still leaves open another ticket that will continue to track the issue in the future, whereas closing a ticket that has no duplicates does not.
>>
> In JIRA, is there a way, when closing a ticket as a duplicate, to
> indicate which ticket it duplicates? I'm used to Bugzilla where such
> information is noted in both the closed ticket and the ticket it's
> merged into.
>
> Thanks,
>
> --
> Michel Alexandre Salim

> 琨log: http://identi.ca/hircus

> http://twitter.com/hircus
> GPG key ID: A36A937A
>
> () ascii ribbon campaign - against html e-mail
> /\ www.asciiribbon.org - against proprietary attachments
>

> --
> You received this message because you are subscribed to the Google Groups "Clojure Dev" group.
> To post to this group, send email to cloju...@googlegroups.com.
> To unsubscribe from this group, send email to clojure-dev...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/clojure-dev?hl=en.
>