From: Robert M. <rob...@gm...> - 2013-10-16 19:22:11
|
Hi, I'm watching commits fly by in master, and I'm just curious where we stand. Paul - did you finish porting the changes that you wanted to? Is it time for me to start finding out what breaks? Robert -- http://robert.muntea.nu/ |
From: Paul R. <pa...@ma...> - 2013-10-16 19:27:55
|
Yes mainly. Draged myself had discussion last night on db installer. Will have a patch to implement thAT discussion in next 48 h I've still got a 20k line diff to reduce. I don't think dregads merged in his oracle branch yet If theres any data type issues with db stuff would be good to let me know rather than fixing as still want to look at A few things around that. Paul . On Oct 16, 2013 8:22 PM, "Robert Munteanu" <rob...@gm...> wrote: > Hi, > > I'm watching commits fly by in master, and I'm just curious where we stand. > > Paul - did you finish porting the changes that you wanted to? Is it > time for me to start finding out what breaks? > > Robert > > -- > http://robert.muntea.nu/ > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > |
From: Robert M. <rob...@gm...> - 2013-10-16 19:43:22
|
Can we count on you to give us a heads-up when it's safe to start looking at stuff and file bugs? On Wed, Oct 16, 2013 at 10:27 PM, Paul Richards <pa...@ma...> wrote: > Yes mainly. > > Draged myself had discussion last night on db installer. Will have a patch > to implement thAT discussion in next 48 h > > I've still got a 20k line diff to reduce. I don't think dregads merged in > his oracle branch yet > > If theres any data type issues with db stuff would be good to let me know > rather than fixing as still want to look at A few things around that. > > Paul > . > > On Oct 16, 2013 8:22 PM, "Robert Munteanu" <rob...@gm...> > wrote: >> >> Hi, >> >> I'm watching commits fly by in master, and I'm just curious where we >> stand. >> >> Paul - did you finish porting the changes that you wanted to? Is it >> time for me to start finding out what breaks? >> >> Robert >> >> -- >> http://robert.muntea.nu/ >> >> >> ------------------------------------------------------------------------------ >> October Webinars: Code for Performance >> Free Intel webinars can help you accelerate application performance. >> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >> from >> the latest Intel processors and coprocessors. See abstracts and register > >> >> http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk >> _______________________________________________ >> mantisbt-dev mailing list >> man...@li... >> https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > -- http://robert.muntea.nu/ |
From: Victor B. <vb...@gm...> - 2013-10-17 04:00:50
|
Please let us know what "db installer" / database changes (incl. Oracle changes) are planned? I'm seeing more changes related to db (e.g. db_num_rows, db_prepare_int, installer, etc) than I expected given that we were thinking that changes to all APIs relating to db will come with the re-visting of the ADODB dependency post this release. That may be OK as long as there is agreement, but just curious about the scope here. I've also noticed that the build is broken -- my parsing of it is that it is related to some of the db related checkins. It would be great if we can get this healthy again. https://travis-ci.org/mantisbt/mantisbt/builds On Wed, Oct 16, 2013 at 12:43 PM, Robert Munteanu <rob...@gm... > wrote: > Can we count on you to give us a heads-up when it's safe to start > looking at stuff and file bugs? > > On Wed, Oct 16, 2013 at 10:27 PM, Paul Richards <pa...@ma...> > wrote: > > Yes mainly. > > > > Draged myself had discussion last night on db installer. Will have a > patch > > to implement thAT discussion in next 48 h > > > > I've still got a 20k line diff to reduce. I don't think dregads merged > in > > his oracle branch yet > > > > If theres any data type issues with db stuff would be good to let me know > > rather than fixing as still want to look at A few things around that. > > > > Paul > > . > > > > On Oct 16, 2013 8:22 PM, "Robert Munteanu" <rob...@gm...> > > wrote: > >> > >> Hi, > >> > >> I'm watching commits fly by in master, and I'm just curious where we > >> stand. > >> > >> Paul - did you finish porting the changes that you wanted to? Is it > >> time for me to start finding out what breaks? > >> > >> Robert > >> > >> -- > >> http://robert.muntea.nu/ > >> > >> > >> > ------------------------------------------------------------------------------ > >> October Webinars: Code for Performance > >> Free Intel webinars can help you accelerate application performance. > >> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > >> from > >> the latest Intel processors and coprocessors. See abstracts and > register > > >> > >> > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > >> _______________________________________________ > >> mantisbt-dev mailing list > >> man...@li... > >> https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > > > > > > > ------------------------------------------------------------------------------ > > October Webinars: Code for Performance > > Free Intel webinars can help you accelerate application performance. > > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > > from > > the latest Intel processors and coprocessors. See abstracts and register > > > > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > > _______________________________________________ > > mantisbt-dev mailing list > > man...@li... > > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > > > > > > -- > http://robert.muntea.nu/ > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > |
From: Damien R. <dr...@ma...> - 2013-10-17 07:47:34
|
On 2013-10-17 06:00, Victor Boctor wrote: > Please let us know what "db installer" / database changes (incl. Oracle > changes) are planned? I'm seeing more changes related to db (e.g. > db_num_rows, db_prepare_int, installer, etc) than I expected given that > we were thinking that changes to all APIs relating to db will come with > the re-visting of the ADODB dependency post this release. That may be > OK as long as there is agreement, but just curious about the scope here. What I still have in the works: - Oracle branch, 99% ready to push (found one regression issue yesterday in testing, which I need to fix) - Enhancements to install.php (90% ready) - PostgreSQL branch, 90% done (needs testing) I expect to be done with most of the above today if all goes well. > I've also noticed that the build is broken -- my parsing of it is that > it is related to some of the db related checkins. It would be great if > we can get this healthy again. I have fixed these issues locally yesterday, but got tied up and did not have time to complete tests and push. > On Wed, Oct 16, 2013 at 12:43 PM, Robert Munteanu wrote: > > Can we count on you to give us a heads-up when it's safe to start > looking at stuff and file bugs? Will do. D. |
From: Paul R. <pa...@ma...> - 2013-10-17 20:30:15
|
Right, I think I'm not going to bother trying to merge these and leave them for db changes in 1.4 And now the 'fun' part: When can we branch master into master-1.3? I said in a round about way to victor/damien before [and maybe to list - not sure], that I need mssql changes done by end of October. So to do that properly, I've got 13 days to test db api changes, and need somewhere to put them. It would strike me we could branch 1.3 off master to do this now we've got most of everyone's commits in. I've still got concerns that I'm not happy with promoting the use of non-mysql databases in 1.3 [even though support is greatly improved] as a new feature where we may do things differently from 1.4. Can we leave publicising the oracle support off the 1.3 release notes? On Thu, Oct 17, 2013 at 9:03 PM, Paul Richards <pa...@ma...> wrote: > OK, Didn't realise you had installer fixes on top of your oracle stuff - > as I had a bunch of installer fixes too to now work out how to merge ;/ > > Paul > > > On Thu, Oct 17, 2013 at 6:05 PM, Damien Regad <dr...@ma...> wrote: > >> Hi all >> >> I merged 2 branches into master today: Oracle and Installer fixes. >> >> Resolved a whopping 40 issues or so in the tracker :-) >> >> I noticed that Travis reports a successful build, but in fact they are >> failed. http://www.mantisbt.org/bugs/view.php?id=16506 >> >> I think it's an issue with the pre-build script needing some >> adjustments, I'll look at it later. >> >> I'm almost done with merging my local branches - I still have some >> testing to do on the PostgreSQL and a few minor issues here and there. >> >> Nevertheless, as far as I'm concerned you should be able to start testing. >> >> D >> >> >> >> >> ------------------------------------------------------------------------------ >> October Webinars: Code for Performance >> Free Intel webinars can help you accelerate application performance. >> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >> from >> the latest Intel processors and coprocessors. See abstracts and register > >> >> http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk >> _______________________________________________ >> mantisbt-dev mailing list >> man...@li... >> https://lists.sourceforge.net/lists/listinfo/mantisbt-dev >> > > |
From: Robert M. <rob...@gm...> - 2013-10-17 22:07:29
|
On Thu, Oct 17, 2013 at 8:05 PM, Damien Regad <dr...@ma...> wrote: > Hi all > > I merged 2 branches into master today: Oracle and Installer fixes. > > Resolved a whopping 40 issues or so in the tracker :-) Nice! > > I noticed that Travis reports a successful build, but in fact they are > failed. http://www.mantisbt.org/bugs/view.php?id=16506 > > I think it's an issue with the pre-build script needing some > adjustments, I'll look at it later. I had a little time on my hands and fixed it. > > I'm almost done with merging my local branches - I still have some > testing to do on the PostgreSQL and a few minor issues here and there. Maybe it's a good idea to issue pull requests when merging large changes? This way travis can vote on the pull request and signal any build failures. Damien, Paul, you're doing the heavy lifting right now, WDYT about that? Robert > > Nevertheless, as far as I'm concerned you should be able to start testing. > > D > > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev -- http://robert.muntea.nu/ |
From: Robert M. <rob...@gm...> - 2013-10-18 07:30:50
|
On Fri, Oct 18, 2013 at 1:47 AM, Damien Regad <dr...@ma...> wrote: > On 2013-10-18 00:07, Robert Munteanu wrote: >> I had a little time on my hands and fixed it. > > Thanks, beat me to it ;-) > > I was about to commit something a possibly more elegantly solution, by > passing adequate data to crypto_master_salt post param; the new > installer will then automatically add the entry in the config file for you. This is only a solution to unblock the tests. Is there a bug to add support for this param to the installer? I'll create a bug if there isn't. Robert > >> Maybe it's a good idea to issue pull requests when merging large >> changes? This way travis can vote on the pull request and signal any >> build failures. >> >> Damien, Paul, you're doing the heavy lifting right now, WDYT about that? > > Moving forward, probably the right way to do things as discussed with > Victor earlier. But in this case it's probably more efficient to just go > ahead and commit, and fix later. > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev -- http://robert.muntea.nu/ |
From: Damien R. <dr...@ma...> - 2013-10-18 07:49:51
|
On 2013-10-18 09:30, Robert Munteanu wrote: >> I was about to commit something a possibly more elegantly solution, by >> passing adequate data to crypto_master_salt post param; the new >> installer will then automatically add the entry in the config file for you. > > This is only a solution to unblock the tests. Is there a bug to add > support for this param to the installer? I'll create a bug if there > isn't. The param and corresponding code is already implemented, as part of the enhancements to the installer I pushed yesterday. I was thinking of adding something like &crypto_master_salt=$(base64 /dev/urandom |head -c 50) to the data options of the curl statement. In theory that should do the trick, but I did not test it yet D |
From: Robert M. <rob...@gm...> - 2013-10-18 07:32:42
|
On Fri, Oct 18, 2013 at 1:59 AM, Damien Regad <dr...@ma...> wrote: > On 2013-10-18 00:07, Robert Munteanu wrote: >> I had a little time on my hands and fixed it. > > BTW if you've still got time on your hands, maybe you can investigate > why all of a sudden we get builds running for 45 minutes, with timeout > errors like > > 1) IssueAddTest::testCreateIssueWithMiscNote > SoapFault: Maximum execution time of 30 seconds exceeded > /home/travis/build/mantisbt/mantisbt/tests/soap/SoapBase.php:134 > > when the same test script runs fine locally (the whole think completes > under 1 minute) That would be http://www.mantisbt.org/bugs/view.php?id=16507 , and it affects all bugnote additions, not only using SOAP. Robert > > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev -- http://robert.muntea.nu/ |
From: Damien R. <dr...@ma...> - 2013-10-18 08:10:19
|
On 2013-10-18 09:32, Robert Munteanu wrote: > On Fri, Oct 18, 2013 at 1:59 AM, Damien Regad <dr...@ma...> wrote: >> 1) IssueAddTest::testCreateIssueWithMiscNote >> SoapFault: Maximum execution time of 30 seconds exceeded >> /home/travis/build/mantisbt/mantisbt/tests/soap/SoapBase.php:134 >> >> when the same test script runs fine locally (the whole think completes >> under 1 minute) > > > That would be http://www.mantisbt.org/bugs/view.php?id=16507 , and it > affects all bugnote additions, not only using SOAP. Thanks for opening the issue; I added a couple of notes in there I just don't get why it's working fine for me here, but not for you or Travis. |
From: Damien R. <dr...@ma...> - 2013-10-17 17:06:02
|
Hi all I merged 2 branches into master today: Oracle and Installer fixes. Resolved a whopping 40 issues or so in the tracker :-) I noticed that Travis reports a successful build, but in fact they are failed. http://www.mantisbt.org/bugs/view.php?id=16506 I think it's an issue with the pre-build script needing some adjustments, I'll look at it later. I'm almost done with merging my local branches - I still have some testing to do on the PostgreSQL and a few minor issues here and there. Nevertheless, as far as I'm concerned you should be able to start testing. D |
From: Paul R. <pa...@ma...> - 2013-10-17 20:04:03
|
OK, Didn't realise you had installer fixes on top of your oracle stuff - as I had a bunch of installer fixes too to now work out how to merge ;/ Paul On Thu, Oct 17, 2013 at 6:05 PM, Damien Regad <dr...@ma...> wrote: > Hi all > > I merged 2 branches into master today: Oracle and Installer fixes. > > Resolved a whopping 40 issues or so in the tracker :-) > > I noticed that Travis reports a successful build, but in fact they are > failed. http://www.mantisbt.org/bugs/view.php?id=16506 > > I think it's an issue with the pre-build script needing some > adjustments, I'll look at it later. > > I'm almost done with merging my local branches - I still have some > testing to do on the PostgreSQL and a few minor issues here and there. > > Nevertheless, as far as I'm concerned you should be able to start testing. > > D > > > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > |
From: Damien R. <dr...@ma...> - 2013-10-17 22:16:52
|
On 2013-10-17 22:30, Paul Richards wrote: > When can we branch master into master-1.3? IMO it would be a mistake to do this too early. We should at least wait until we reach a release candidate state before branching. > I said in a round about way to victor/damien before [and maybe to list - > not sure], that I need mssql changes done by end of October. So to do > that properly, I've got 13 days to test db api changes, and need > somewhere to put them. It would strike me we could branch 1.3 off master > to do this now we've got most of everyone's commits in. I would prefer if you actually gave a try to Mantis as it is now with the latest ADODB, in theory (I can't test myself) MSSQL should be much better now. If you really want to move forward with this, I'd recommend you keep it in a private fork/branch for now, and most importantly, keep it up-to-date by frequently merging from master. > Can we leave > publicising the oracle support off the 1.3 release notes? That's one of the major improvements we can "sell" to the users, I think it would be a mistake not to publicize it. |
From: Roland B. <ro...@at...> - 2013-10-18 08:42:22
|
> Damien Regad <dr...@ma...> hat am 18. Oktober 2013 um 00:16 > geschrieben: > > On 2013-10-17 22:30, Paul Richards wrote: > > When can we branch master into master-1.3? > > IMO it would be a mistake to do this too early. We should at least wait > until we reach a release candidate state before branching. +1 as we would not get one more issue like http://www.mantisbt.org/bugs/view.php?id=15721 and another tag like http://www.mantisbt.org/bugs/tag_view_page.php?tag_id=205 with more than 400 attached issues. |
From: Damien R. <dr...@ma...> - 2013-10-17 22:48:01
|
On 2013-10-18 00:07, Robert Munteanu wrote: > I had a little time on my hands and fixed it. Thanks, beat me to it ;-) I was about to commit something a possibly more elegantly solution, by passing adequate data to crypto_master_salt post param; the new installer will then automatically add the entry in the config file for you. > Maybe it's a good idea to issue pull requests when merging large > changes? This way travis can vote on the pull request and signal any > build failures. > > Damien, Paul, you're doing the heavy lifting right now, WDYT about that? Moving forward, probably the right way to do things as discussed with Victor earlier. But in this case it's probably more efficient to just go ahead and commit, and fix later. |
From: Damien R. <dr...@ma...> - 2013-10-17 23:00:13
|
On 2013-10-18 00:07, Robert Munteanu wrote: > I had a little time on my hands and fixed it. BTW if you've still got time on your hands, maybe you can investigate why all of a sudden we get builds running for 45 minutes, with timeout errors like 1) IssueAddTest::testCreateIssueWithMiscNote SoapFault: Maximum execution time of 30 seconds exceeded /home/travis/build/mantisbt/mantisbt/tests/soap/SoapBase.php:134 when the same test script runs fine locally (the whole think completes under 1 minute) |
From: Robert M. <rob...@gm...> - 2013-10-18 08:47:47
|
On Fri, Oct 18, 2013 at 1:16 AM, Damien Regad <dr...@ma...> wrote: > On 2013-10-17 22:30, Paul Richards wrote: >> When can we branch master into master-1.3? > > IMO it would be a mistake to do this too early. We should at least wait > until we reach a release candidate state before branching. +1 . Do we really want to start development on 1.4 until 1.3 is stable and has had some good exposure? Remember that some people ( myself included ) simply stay away from the .0 or even .1 releases for really important stuff. I'd advise keeping focus on 1.3 for a bit longer. > >> I said in a round about way to victor/damien before [and maybe to list - >> not sure], that I need mssql changes done by end of October. So to do >> that properly, I've got 13 days to test db api changes, and need >> somewhere to put them. It would strike me we could branch 1.3 off master >> to do this now we've got most of everyone's commits in. > > I would prefer if you actually gave a try to Mantis as it is now with > the latest ADODB, in theory (I can't test myself) MSSQL should be much > better now. > > If you really want to move forward with this, I'd recommend you keep it > in a private fork/branch for now, and most importantly, keep it > up-to-date by frequently merging from master. +1 > >> Can we leave >> publicising the oracle support off the 1.3 release notes? > > That's one of the major improvements we can "sell" to the users, I think > it would be a mistake not to publicize it. +1 . Unless we know it's broken, we should publicize it. Robert > > > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev -- http://robert.muntea.nu/ |
From: Victor B. <vb...@gm...> - 2013-10-23 05:24:26
|
My responses to some of Paul's comments: - Forking for 1.3.x - We should fork for 1.3.x once the first release candidate (or alpha?) version is released. In other words, fork as part of cutting the release. That should keep us focused on 1.3.x until the first iteration is out. While we are iterating, we want to keep merging between master and master-1.3.x easy. - MS SQL / ADODB: a. I agree that we should evaluate progress made on 1.3.x and advertise achieved improvements. That doesn't mean we can't look into other options for 1.4 and have even better improvements. b. The place to keep MS SQL is a fork under your GitHub account. As per discussion with Paul earlier, I would like the master-2.x branch to move to Paul's github account, same applies for the new branch that should contain current master + db changes. Remember, just keep it a single purpose branch - not Paul's goodness branch :) c. When to merge into master? i. First, the team has reviewed and agreed about the high level change and the approach. (DL discussion then pull request) ii. Second, as late as possible to reduce merge pain - we don't want every bug fix in 1.3.x to be costly to merge into master due to the scope of the DB API call changes. Hence, I recommend getting this merged pre-forking of 1.4.x. That should come after the first couple of stable 1.3.x releases are out. Thanks, -Victor On Fri, Oct 18, 2013 at 1:47 AM, Robert Munteanu <rob...@gm...>wrote: > On Fri, Oct 18, 2013 at 1:16 AM, Damien Regad <dr...@ma...> wrote: > > On 2013-10-17 22:30, Paul Richards wrote: > >> When can we branch master into master-1.3? > > > > IMO it would be a mistake to do this too early. We should at least wait > > until we reach a release candidate state before branching. > > +1 . Do we really want to start development on 1.4 until 1.3 is stable > and has had some good exposure? Remember that some people ( myself > included ) simply stay away from the .0 or even .1 releases for really > important stuff. > > I'd advise keeping focus on 1.3 for a bit longer. > > > > >> I said in a round about way to victor/damien before [and maybe to list - > >> not sure], that I need mssql changes done by end of October. So to do > >> that properly, I've got 13 days to test db api changes, and need > >> somewhere to put them. It would strike me we could branch 1.3 off master > >> to do this now we've got most of everyone's commits in. > > > > I would prefer if you actually gave a try to Mantis as it is now with > > the latest ADODB, in theory (I can't test myself) MSSQL should be much > > better now. > > > > If you really want to move forward with this, I'd recommend you keep it > > in a private fork/branch for now, and most importantly, keep it > > up-to-date by frequently merging from master. > > +1 > > > > >> Can we leave > >> publicising the oracle support off the 1.3 release notes? > > > > That's one of the major improvements we can "sell" to the users, I think > > it would be a mistake not to publicize it. > > +1 . Unless we know it's broken, we should publicize it. > > Robert > > > > > > > > > > > > ------------------------------------------------------------------------------ > > October Webinars: Code for Performance > > Free Intel webinars can help you accelerate application performance. > > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > > the latest Intel processors and coprocessors. See abstracts and register > > > > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > > _______________________________________________ > > mantisbt-dev mailing list > > man...@li... > > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > > > > -- > http://robert.muntea.nu/ > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > |
From: Robert M. <rob...@gm...> - 2013-10-23 13:23:58
|
On Wed, Oct 23, 2013 at 8:24 AM, Victor Boctor <vb...@gm...> wrote: > My responses to some of Paul's comments: > > - Forking for 1.3.x - We should fork for 1.3.x once the first release > candidate (or alpha?) version is released. In other words, fork as part of > cutting the release. That should keep us focused on 1.3.x until the first > iteration is out. While we are iterating, we want to keep merging between > master and master-1.3.x easy. +1 for forking after the RC. Even though I expect to cut a couple of releases soon for the 1.3.0 final release, for fixing any regressions. > > - MS SQL / ADODB: > a. I agree that we should evaluate progress made on 1.3.x and advertise > achieved improvements. That doesn't mean we can't look into other options > for 1.4 and have even better improvements. > b. The place to keep MS SQL is a fork under your GitHub account. As per > discussion with Paul earlier, I would like the master-2.x branch to move to > Paul's github account, same applies for the new branch that should contain > current master + db changes. Remember, just keep it a single purpose branch > - not Paul's goodness branch :) > c. When to merge into master? > i. First, the team has reviewed and agreed about the high level change and > the approach. (DL discussion then pull request) > ii. Second, as late as possible to reduce merge pain - we don't want every > bug fix in 1.3.x to be costly to merge into master due to the scope of the > DB API call changes. Hence, I recommend getting this merged pre-forking of > 1.4.x. That should come after the first couple of stable 1.3.x releases are > out. +1 Robert > > Thanks, > -Victor > > > > On Fri, Oct 18, 2013 at 1:47 AM, Robert Munteanu <rob...@gm...> > wrote: >> >> On Fri, Oct 18, 2013 at 1:16 AM, Damien Regad <dr...@ma...> wrote: >> > On 2013-10-17 22:30, Paul Richards wrote: >> >> When can we branch master into master-1.3? >> > >> > IMO it would be a mistake to do this too early. We should at least wait >> > until we reach a release candidate state before branching. >> >> +1 . Do we really want to start development on 1.4 until 1.3 is stable >> and has had some good exposure? Remember that some people ( myself >> included ) simply stay away from the .0 or even .1 releases for really >> important stuff. >> >> I'd advise keeping focus on 1.3 for a bit longer. >> >> > >> >> I said in a round about way to victor/damien before [and maybe to list >> >> - >> >> not sure], that I need mssql changes done by end of October. So to do >> >> that properly, I've got 13 days to test db api changes, and need >> >> somewhere to put them. It would strike me we could branch 1.3 off >> >> master >> >> to do this now we've got most of everyone's commits in. >> > >> > I would prefer if you actually gave a try to Mantis as it is now with >> > the latest ADODB, in theory (I can't test myself) MSSQL should be much >> > better now. >> > >> > If you really want to move forward with this, I'd recommend you keep it >> > in a private fork/branch for now, and most importantly, keep it >> > up-to-date by frequently merging from master. >> >> +1 >> >> > >> >> Can we leave >> >> publicising the oracle support off the 1.3 release notes? >> > >> > That's one of the major improvements we can "sell" to the users, I think >> > it would be a mistake not to publicize it. >> >> +1 . Unless we know it's broken, we should publicize it. >> >> Robert >> >> > >> > >> > >> > >> > >> > ------------------------------------------------------------------------------ >> > October Webinars: Code for Performance >> > Free Intel webinars can help you accelerate application performance. >> > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >> > from >> > the latest Intel processors and coprocessors. See abstracts and register >> > > >> > >> > http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk >> > _______________________________________________ >> > mantisbt-dev mailing list >> > man...@li... >> > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev >> >> >> >> -- >> http://robert.muntea.nu/ >> >> >> ------------------------------------------------------------------------------ >> October Webinars: Code for Performance >> Free Intel webinars can help you accelerate application performance. >> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most >> from >> the latest Intel processors and coprocessors. See abstracts and register > >> >> http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk >> _______________________________________________ >> mantisbt-dev mailing list >> man...@li... >> https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > > > > ------------------------------------------------------------------------------ > October Webinars: Code for Performance > Free Intel webinars can help you accelerate application performance. > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most > from > the latest Intel processors and coprocessors. See abstracts and register > > http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk > _______________________________________________ > mantisbt-dev mailing list > man...@li... > https://lists.sourceforge.net/lists/listinfo/mantisbt-dev > -- http://robert.muntea.nu/ |