GNATS feature requests

GNATS feature requests

[Bryce McKinlay]

Here are a few requests for GNATS features that would fix problems
with the GCC bug database which I find annoying...

1. Ability to notify multiple mailing lists of follow-ups to certain
PR categories. Currently, notification of new PRs in the libgcj & java
categories are sent to java-prs (as well as gcc-prs), but *follow-ups*
to those PRs do not go to the java-prs list. This is really annoying,
because I have to go to the mailing list pages to check for follow
ups, and often miss them, since I don't really want to subscribe to
gcc-prs ;-). Tom Tromey tells me he tried to configure GNATS to do
this, but it couldn't do it. I can't imagine that it would be
particularly difficult to implement this, and it would make life a
*lot* easier.

2. Notification of changes to fields like "synopsis" and "class-id".
If I rename a PR, or mark it as a duplicate, etc, it would be nice if
GNATS would send out a notification of that change too (or at least
add this information to a notification that it was already going to
send out). Currently it notifies changes to the status, owner, etc,
but for some reason not the other fields.

3. If I mark a bug as a duplicate, it would be really helpful if GNATS
would add the email of the originator of the duplicate PR to the
notification/contact list of the original/canonical PR (I suppose it
would need a way of specifying which PR the PR is a duplicate of to do
this.). It would also be cool if it would add notes like "PR xx was
marked as a duplicate of this PR" and "This is a duplicate of PR yy"
automatically. It may also be useful to have a "see-also" mechanism to
say that one PR is similar to another, but not necessarily a
duplicate.

regards

  [ bryce ]

Re: GNATS feature requests

[Milan Zamazal]

>>>>> "BM" == Bryce McKinlay <bryce@albatross.co.nz> writes:

    BM> Here are a few requests for GNATS features that would fix
    BM> problems with the GCC bug database which I find annoying...

    BM> 1. Ability to notify multiple mailing lists of follow-ups to
    BM> certain PR categories.

[...]

    BM> 2. Notification of changes to fields like "synopsis" and
    BM> "class-id".

[...]

    BM> 3. If I mark a bug as a duplicate, it would be really helpful if
    BM> GNATS would add the email of the originator of the duplicate PR
    BM> to the notification/contact list of the original/canonical PR

[...]

I think all the requests are reasonable.  I'll try to implement once
GNATS 4 is released (till that time, there are more critical tasks like
getting the system work well, to fix all important bugs from BTS, and
updating the manual).

Regards,

Milan Zamazal

-- 
I think any law that restricts independent use of brainpower is suspect.
                                               -- Kent Pitman in comp.lang.lisp

Re: GNATS feature requests

[Richard Earnshaw]

>     BM> 3. If I mark a bug as a duplicate, it would be really helpful if
>     BM> GNATS would add the email of the originator of the duplicate PR
>     BM> to the notification/contact list of the original/canonical PR

It would also be useful for the canonical PR to contain a reference to 
those PRs that are duplicates, just in case one of them happens to have 
additional pertinent information.  A field such as "duplicates: <pr-list>" 
 would do this.

Re: GNATS feature requests

[Bryce McKinlay]

On Monday, April 9, 2001, at 03:37 AM, Milan Zamazal wrote:

>>>>>> "BM" == Bryce McKinlay <bryce@albatross.co.nz> writes:
>
>     BM> Here are a few requests for GNATS features that would fix
>     BM> problems with the GCC bug database which I find annoying...
>
>     BM> 1. Ability to notify multiple mailing lists of follow-ups to
>     BM> certain PR categories.
>
> I think all the requests are reasonable.  I'll try to implement once
> GNATS 4 is released (till that time, there are more critical tasks like
> getting the system work well, to fix all important bugs from BTS, and
> updating the manual).

Thanks for looking at these. FWIW, Item 1. is the one that I find most 
annoying currently, and I suspect that it would be relativly simple to 
implement.

Re: GNATS feature requests

[Milan Zamazal]

>>>>> "RE" == Richard Earnshaw <rearnsha@arm.com> writes:

    RE> It would also be useful for the canonical PR to contain a
    RE> reference to those PRs that are duplicates, just in case one of
    RE> them happens to have additional pertinent information.  A field
    RE> such as "duplicates: <pr-list>" would do this.

Yes, there should be some way of linking of both the problems (kind of
virtual merge).

Milan Zamazal

-- 
Here is my advice, don't try to program the bleeding edge for the
general populace unless you really, really, really like migraines.
						   Neal H. Walfield