How to make fixes to Hudson core issues

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

How to make fixes to Hudson core issues

Benjamin Muschko
Hi,

I would be interested in implementing issue #229. My development environment is already set up and I made some changes to the Hudson core concerning this issue. Though I know that it's much easier to develop a plugin for Hudson my guess is that this issue rather has to be implemented in the core itself. How do you guys normally handle these kinds of changes and how is the process to get changes to the core into CVS?

Thanks,

Ben
Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

stephenconnolly
If you have commit access and feel confident, you could just commit
your changes...

If you are not feeling quite so confident, post a patch file to the
dev list and somebody with commit access will review and commit your
patch for you (or if you have commit access, once they've reviewed it
you can commit it yourself)

-Stephen

On Dec 5, 2007 3:03 PM, Benjamin Muschko <[hidden email]> wrote:

>
> Hi,
>
> I would be interested in implementing
> https://hudson.dev.java.net/issues/show_bug.cgi?id=229 issue #229 . My
> development environment is already set up and I made some changes to the
> Hudson core concerning this issue. Though I know that it's much easier to
> develop a plugin for Hudson my guess is that this issue rather has to be
> implemented in the core itself. How do you guys normally handle these kinds
> of changes and how is the process to get changes to the core into CVS?
>
> Thanks,
>
> Ben
> --
> View this message in context: http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14173502
> Sent from the Hudson dev mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Kohsuke Kawaguchi
Administrator
Stephen Connolly wrote:
> If you have commit access and feel confident, you could just commit
> your changes...
>
> If you are not feeling quite so confident, post a patch file to the
> dev list and somebody with commit access will review and commit your
> patch for you (or if you have commit access, once they've reviewed it
> you can commit it yourself)

+1

--
Kohsuke Kawaguchi
Sun Microsystems                   [hidden email]

smime.p7s (4K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Benjamin Muschko
In reply to this post by stephenconnolly
Sorry, but I don't have commit access. I attached the patch file. Please review the code and let me know in case you have questions, comments or concerns.

Thanks,

Ben

issue229_patch.txt

Stephen Connolly-2 wrote
If you have commit access and feel confident, you could just commit
your changes...

If you are not feeling quite so confident, post a patch file to the
dev list and somebody with commit access will review and commit your
patch for you (or if you have commit access, once they've reviewed it
you can commit it yourself)

-Stephen

On Dec 5, 2007 3:03 PM, Benjamin Muschko <benjamin.muschko@gmail.com> wrote:
>
> Hi,
>
> I would be interested in implementing
> https://hudson.dev.java.net/issues/show_bug.cgi?id=229 issue #229 . My
> development environment is already set up and I made some changes to the
> Hudson core concerning this issue. Though I know that it's much easier to
> develop a plugin for Hudson my guess is that this issue rather has to be
> implemented in the core itself. How do you guys normally handle these kinds
> of changes and how is the process to get changes to the core into CVS?
>
> Thanks,
>
> Ben
> --
> View this message in context: http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14173502
> Sent from the Hudson dev mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> For additional commands, e-mail: dev-help@hudson.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
For additional commands, e-mail: dev-help@hudson.dev.java.net
Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Kohsuke Kawaguchi
Administrator
One small issue with this is that this causes behavior changes to
people's projects. Since the current behavior is to use the -D option,
this needs to be the default.

But more importantly, shouldn't it be possible to run CVS normally,
and then go in later to update CVS/Entries to delete the sticky tag?
In that way, you get best of both worlds and we don't need to ask
users to do any configuration.

2007/12/5, Benjamin Muschko <[hidden email]>:

>
> Sorry, but I don't have commit access. I attached the patch file. Please
> review the code and let me know in case you have questions, comments or
> concerns.
>
> Thanks,
>
> Ben
>
> http://www.nabble.com/file/p14185500/issue229_patch.txt issue229_patch.txt
>
>
> Stephen Connolly-2 wrote:
> >
> > If you have commit access and feel confident, you could just commit
> > your changes...
> >
> > If you are not feeling quite so confident, post a patch file to the
> > dev list and somebody with commit access will review and commit your
> > patch for you (or if you have commit access, once they've reviewed it
> > you can commit it yourself)
> >
> > -Stephen
> >
> > On Dec 5, 2007 3:03 PM, Benjamin Muschko <[hidden email]>
> > wrote:
> >>
> >> Hi,
> >>
> >> I would be interested in implementing
> >> https://hudson.dev.java.net/issues/show_bug.cgi?id=229 issue #229 . My
> >> development environment is already set up and I made some changes to the
> >> Hudson core concerning this issue. Though I know that it's much easier to
> >> develop a plugin for Hudson my guess is that this issue rather has to be
> >> implemented in the core itself. How do you guys normally handle these
> >> kinds
> >> of changes and how is the process to get changes to the core into CVS?
> >>
> >> Thanks,
> >>
> >> Ben
> >> --
> >> View this message in context:
> >> http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14173502
> >> Sent from the Hudson dev mailing list archive at Nabble.com.
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [hidden email]
> >> For additional commands, e-mail: [hidden email]
> >>
> >>
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [hidden email]
> > For additional commands, e-mail: [hidden email]
> >
> >
> >
>
> --
> View this message in context: http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14185500
> Sent from the Hudson dev mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>


--
Kohsuke Kawaguchi

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Benjamin Muschko
Yes, you're right. Let me summarize: The checkout command will always use the -D option. For the update command there should be an option -A to reset the sticky tags. In conclusion the text for this CVS option would rather be "Reset Sticky Tags". Is that correct?

kohsuke wrote
One small issue with this is that this causes behavior changes to
people's projects. Since the current behavior is to use the -D option,
this needs to be the default.

But more importantly, shouldn't it be possible to run CVS normally,
and then go in later to update CVS/Entries to delete the sticky tag?
In that way, you get best of both worlds and we don't need to ask
users to do any configuration.

2007/12/5, Benjamin Muschko <benjamin.muschko@gmail.com>:
>
> Sorry, but I don't have commit access. I attached the patch file. Please
> review the code and let me know in case you have questions, comments or
> concerns.
>
> Thanks,
>
> Ben
>
> http://www.nabble.com/file/p14185500/issue229_patch.txt issue229_patch.txt
>
>
> Stephen Connolly-2 wrote:
> >
> > If you have commit access and feel confident, you could just commit
> > your changes...
> >
> > If you are not feeling quite so confident, post a patch file to the
> > dev list and somebody with commit access will review and commit your
> > patch for you (or if you have commit access, once they've reviewed it
> > you can commit it yourself)
> >
> > -Stephen
> >
> > On Dec 5, 2007 3:03 PM, Benjamin Muschko <benjamin.muschko@gmail.com>
> > wrote:
> >>
> >> Hi,
> >>
> >> I would be interested in implementing
> >> https://hudson.dev.java.net/issues/show_bug.cgi?id=229 issue #229 . My
> >> development environment is already set up and I made some changes to the
> >> Hudson core concerning this issue. Though I know that it's much easier to
> >> develop a plugin for Hudson my guess is that this issue rather has to be
> >> implemented in the core itself. How do you guys normally handle these
> >> kinds
> >> of changes and how is the process to get changes to the core into CVS?
> >>
> >> Thanks,
> >>
> >> Ben
> >> --
> >> View this message in context:
> >> http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14173502
> >> Sent from the Hudson dev mailing list archive at Nabble.com.
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> >> For additional commands, e-mail: dev-help@hudson.dev.java.net
> >>
> >>
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> > For additional commands, e-mail: dev-help@hudson.dev.java.net
> >
> >
> >
>
> --
> View this message in context: http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14185500
> Sent from the Hudson dev mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> For additional commands, e-mail: dev-help@hudson.dev.java.net
>
>


--
Kohsuke Kawaguchi

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
For additional commands, e-mail: dev-help@hudson.dev.java.net
Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Kohsuke Kawaguchi
Administrator
No, the approach that I'm trying to suggest doesn't require any option.

We'd always check out and update with the -D option, so that we get
consistent copy that's bound by the build timestamp, but once CVS does
its job, update CVS/Entries so as not to have the trace of the sticky
date.

Oh, wait, I guess we still need a checkbox not to put -D in case of
checking out a tag (or is there a way in CVS for us to figure out
whether something is a tag or a branch?)

2007/12/6, Benjamin Muschko <[hidden email]>:

>
> Yes, you're right. Let me summarize: The checkout command will always use the
> -D option. For the update command there should be an option -A to reset the
> sticky tags. In conclusion the text for this CVS option would rather be
> "Reset Sticky Tags". Is that correct?
>
>
> kohsuke wrote:
> >
> > One small issue with this is that this causes behavior changes to
> > people's projects. Since the current behavior is to use the -D option,
> > this needs to be the default.
> >
> > But more importantly, shouldn't it be possible to run CVS normally,
> > and then go in later to update CVS/Entries to delete the sticky tag?
> > In that way, you get best of both worlds and we don't need to ask
> > users to do any configuration.
> >
> > 2007/12/5, Benjamin Muschko <[hidden email]>:
> >>
> >> Sorry, but I don't have commit access. I attached the patch file. Please
> >> review the code and let me know in case you have questions, comments or
> >> concerns.
> >>
> >> Thanks,
> >>
> >> Ben
> >>
> >> http://www.nabble.com/file/p14185500/issue229_patch.txt
> >> issue229_patch.txt
> >>
> >>
> >> Stephen Connolly-2 wrote:
> >> >
> >> > If you have commit access and feel confident, you could just commit
> >> > your changes...
> >> >
> >> > If you are not feeling quite so confident, post a patch file to the
> >> > dev list and somebody with commit access will review and commit your
> >> > patch for you (or if you have commit access, once they've reviewed it
> >> > you can commit it yourself)
> >> >
> >> > -Stephen
> >> >
> >> > On Dec 5, 2007 3:03 PM, Benjamin Muschko <[hidden email]>
> >> > wrote:
> >> >>
> >> >> Hi,
> >> >>
> >> >> I would be interested in implementing
> >> >> https://hudson.dev.java.net/issues/show_bug.cgi?id=229 issue #229 . My
> >> >> development environment is already set up and I made some changes to
> >> the
> >> >> Hudson core concerning this issue. Though I know that it's much easier
> >> to
> >> >> develop a plugin for Hudson my guess is that this issue rather has to
> >> be
> >> >> implemented in the core itself. How do you guys normally handle these
> >> >> kinds
> >> >> of changes and how is the process to get changes to the core into CVS?
> >> >>
> >> >> Thanks,
> >> >>
> >> >> Ben
> >> >> --
> >> >> View this message in context:
> >> >>
> >> http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14173502
> >> >> Sent from the Hudson dev mailing list archive at Nabble.com.
> >> >>
> >> >> ---------------------------------------------------------------------
> >> >> To unsubscribe, e-mail: [hidden email]
> >> >> For additional commands, e-mail: [hidden email]
> >> >>
> >> >>
> >> >
> >> > ---------------------------------------------------------------------
> >> > To unsubscribe, e-mail: [hidden email]
> >> > For additional commands, e-mail: [hidden email]
> >> >
> >> >
> >> >
> >>
> >> --
> >> View this message in context:
> >> http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14185500
> >> Sent from the Hudson dev mailing list archive at Nabble.com.
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [hidden email]
> >> For additional commands, e-mail: [hidden email]
> >>
> >>
> >
> >
> > --
> > Kohsuke Kawaguchi
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [hidden email]
> > For additional commands, e-mail: [hidden email]
> >
> >
> >
>
> --
> View this message in context: http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14191926
> Sent from the Hudson dev mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>


--
Kohsuke Kawaguchi

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Benjamin Muschko
OK, how would you suggest efficiently reseting the sticky date from the CVS/Entries files after updating the source? I just know the -A option in CVS which we don't want to use.

Yes, some people wanted to check out tags and this only works without the -D option.  I'm not aware of a way in CVS to find out if a label is a tag or a branch. Some posts suggested a radio button that let's you decide whether you want to use a tag or a branch. What do you think about that?

kohsuke wrote
No, the approach that I'm trying to suggest doesn't require any option.

We'd always check out and update with the -D option, so that we get
consistent copy that's bound by the build timestamp, but once CVS does
its job, update CVS/Entries so as not to have the trace of the sticky
date.

Oh, wait, I guess we still need a checkbox not to put -D in case of
checking out a tag (or is there a way in CVS for us to figure out
whether something is a tag or a branch?)

2007/12/6, Benjamin Muschko <benjamin.muschko@gmail.com>:
>
> Yes, you're right. Let me summarize: The checkout command will always use the
> -D option. For the update command there should be an option -A to reset the
> sticky tags. In conclusion the text for this CVS option would rather be
> "Reset Sticky Tags". Is that correct?
>
>
> kohsuke wrote:
> >
> > One small issue with this is that this causes behavior changes to
> > people's projects. Since the current behavior is to use the -D option,
> > this needs to be the default.
> >
> > But more importantly, shouldn't it be possible to run CVS normally,
> > and then go in later to update CVS/Entries to delete the sticky tag?
> > In that way, you get best of both worlds and we don't need to ask
> > users to do any configuration.
> >
> > 2007/12/5, Benjamin Muschko <benjamin.muschko@gmail.com>:
> >>
> >> Sorry, but I don't have commit access. I attached the patch file. Please
> >> review the code and let me know in case you have questions, comments or
> >> concerns.
> >>
> >> Thanks,
> >>
> >> Ben
> >>
> >> http://www.nabble.com/file/p14185500/issue229_patch.txt
> >> issue229_patch.txt
> >>
> >>
> >> Stephen Connolly-2 wrote:
> >> >
> >> > If you have commit access and feel confident, you could just commit
> >> > your changes...
> >> >
> >> > If you are not feeling quite so confident, post a patch file to the
> >> > dev list and somebody with commit access will review and commit your
> >> > patch for you (or if you have commit access, once they've reviewed it
> >> > you can commit it yourself)
> >> >
> >> > -Stephen
> >> >
> >> > On Dec 5, 2007 3:03 PM, Benjamin Muschko <benjamin.muschko@gmail.com>
> >> > wrote:
> >> >>
> >> >> Hi,
> >> >>
> >> >> I would be interested in implementing
> >> >> https://hudson.dev.java.net/issues/show_bug.cgi?id=229 issue #229 . My
> >> >> development environment is already set up and I made some changes to
> >> the
> >> >> Hudson core concerning this issue. Though I know that it's much easier
> >> to
> >> >> develop a plugin for Hudson my guess is that this issue rather has to
> >> be
> >> >> implemented in the core itself. How do you guys normally handle these
> >> >> kinds
> >> >> of changes and how is the process to get changes to the core into CVS?
> >> >>
> >> >> Thanks,
> >> >>
> >> >> Ben
> >> >> --
> >> >> View this message in context:
> >> >>
> >> http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14173502
> >> >> Sent from the Hudson dev mailing list archive at Nabble.com.
> >> >>
> >> >> ---------------------------------------------------------------------
> >> >> To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> >> >> For additional commands, e-mail: dev-help@hudson.dev.java.net
> >> >>
> >> >>
> >> >
> >> > ---------------------------------------------------------------------
> >> > To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> >> > For additional commands, e-mail: dev-help@hudson.dev.java.net
> >> >
> >> >
> >> >
> >>
> >> --
> >> View this message in context:
> >> http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14185500
> >> Sent from the Hudson dev mailing list archive at Nabble.com.
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> >> For additional commands, e-mail: dev-help@hudson.dev.java.net
> >>
> >>
> >
> >
> > --
> > Kohsuke Kawaguchi
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> > For additional commands, e-mail: dev-help@hudson.dev.java.net
> >
> >
> >
>
> --
> View this message in context: http://www.nabble.com/How-to-make-fixes-to-Hudson-core-issues-tf4950233.html#a14191926
> Sent from the Hudson dev mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
> For additional commands, e-mail: dev-help@hudson.dev.java.net
>
>


--
Kohsuke Kawaguchi

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hudson.dev.java.net
For additional commands, e-mail: dev-help@hudson.dev.java.net
Reply | Threaded
Open this post in threaded view
|

Re: How to make fixes to Hudson core issues

Kohsuke Kawaguchi
Administrator
2007/12/7, Benjamin Muschko <[hidden email]>:
> OK, how would you suggest efficiently reseting the sticky date from the
> CVS/Entries files after updating the source? I just know the -A option in
> CVS which we don't want to use.

I was thinking just opening that file and modifying it, without using cvs CLI.

> Yes, some people wanted to check out tags and this only works without the -D
> option.  I'm not aware of a way in CVS to find out if a label is a tag or a
> branch. Some posts suggested a radio button that let's you decide whether
> you want to use a tag or a branch. What do you think about that?

I try to have Hudson figure things out than asking users to configure
something, but I guess you are right that in this case a radio button
or a checkbox would be unavoidable.

--
Kohsuke Kawaguchi

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]