Обсуждение: Planning for 1.14 Beta 1
Hi, We need to start planning for 1.14 beta 1, to be released roughly in line with PostgreSQL 9.1 beta 1 which is likely to be just a few short weeks away. Does anyone have any outstanding patches or WIP for this release? -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Hi, Le 04/04/2011 16:59, Dave Page a écrit : > [...] > We need to start planning for 1.14 beta 1, to be released roughly in > line with PostgreSQL 9.1 beta 1 which is likely to be just a few short > weeks away. > > Does anyone have any outstanding patches or WIP for this release? > Nope. I'm only working on the dialogs right now. There are quite a lot of tickets to work on, but I don't want to start big tickets for this release. -- Guillaume http://www.postgresql.fr http://dalibo.com
Le 04/04/2011 21:53, Guillaume Lelarge a écrit : > Hi, > > Le 04/04/2011 16:59, Dave Page a écrit : >> [...] >> We need to start planning for 1.14 beta 1, to be released roughly in >> line with PostgreSQL 9.1 beta 1 which is likely to be just a few short >> weeks away. >> >> Does anyone have any outstanding patches or WIP for this release? >> > > Nope. I'm only working on the dialogs right now. There are quite a lot > of tickets to work on, but I don't want to start big tickets for this > release. > Hmm, nights help you remember some stuff. So, just to let you know the things we don't support yet: * 7.4: operator class (the properties dialog only) * 8.3: operator family (the properties dialog only) * 9.1: * per column collation support * security label * hostname/host suffixes in pg_hba.conf Operator class and operator family are not showstoppers, at least to me. I would love to have them, but they are definitely not a priority. The security label feature isn't hard to implement. I previously thought we would need to add yet another tab, which bores me to death, but I had another idea while working on the dialogs. Security label behaves more or less like comments, so I intend to add support just like support for comments was done. It should be a quick patch to write. I'm not sure about the host-name/suffix stuff. There is a bug to fix in the pg_hba.conf editor, could be part of it. I still didn't work on the "per column collation support" because Tom was working really hard on fixing/implementing it. But I think we need that one before beta1. Maybe the only showstopper right now. So, here are the stuff I can do quickly: * finish the work on the dialogs (probably thursday) * add the security label support (friday) The two of them and my two latest patches not yet commited could be available at the end of the week. And, unless someone works on it, I can start the work on the "per column collation" patch, starting next week. It doesn't feel to be a hard patch to write, even if it touches many objects. -- Guillaume http://www.postgresql.fr http://dalibo.com
Hi On Tue, Apr 5, 2011 at 9:24 AM, Guillaume Lelarge <guillaume@lelarge.info> wrote: > Hmm, nights help you remember some stuff. So, just to let you know the > things we don't support yet: > > * 7.4: operator class (the properties dialog only) > > * 8.3: operator family (the properties dialog only) We intentionally never wrote those as they would take more work than at the time we could justify. For completeness they would be nice to have, but they're certainly not a show stopper. > * 9.1: > * per column collation support > * security label > * hostname/host suffixes in pg_hba.conf > > Operator class and operator family are not showstoppers, at least to me. > I would love to have them, but they are definitely not a priority. Oh, yeah. That's what I get for replying as I read :-p > The security label feature isn't hard to implement. I previously thought > we would need to add yet another tab, which bores me to death, but I had > another idea while working on the dialogs. Security label behaves more > or less like comments, so I intend to add support just like support for > comments was done. It should be a quick patch to write. OK. > I'm not sure about the host-name/suffix stuff. There is a bug to fix in > the pg_hba.conf editor, could be part of it. OK. > I still didn't work on the "per column collation support" because Tom > was working really hard on fixing/implementing it. But I think we need > that one before beta1. Maybe the only showstopper right now. I certainly think we should aim to include that - it's probably going to be more popular than security labels. > So, here are the stuff I can do quickly: > * finish the work on the dialogs (probably thursday) > * add the security label support (friday) > > The two of them and my two latest patches not yet commited could be > available at the end of the week. Sounds good. > And, unless someone works on it, I can start the work on the "per column > collation" patch, starting next week. It doesn't feel to be a hard patch > to write, even if it touches many objects. I suspect not. Anyway - that's great, thanks. It looks like we're shooting for a release of beta 1 on the 18th, which means I'll need to be able to wrap beta 1 first thing in the morning on Friday 15th. -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
On Tue, Apr 5, 2011 at 11:01, Dave Page <dpage@pgadmin.org> wrote: > Hi > > On Tue, Apr 5, 2011 at 9:24 AM, Guillaume Lelarge > <guillaume@lelarge.info> wrote: >> The security label feature isn't hard to implement. I previously thought >> we would need to add yet another tab, which bores me to death, but I had >> another idea while working on the dialogs. Security label behaves more >> or less like comments, so I intend to add support just like support for >> comments was done. It should be a quick patch to write. > > OK. > >> I'm not sure about the host-name/suffix stuff. There is a bug to fix in >> the pg_hba.conf editor, could be part of it. > > OK. > >> I still didn't work on the "per column collation support" because Tom >> was working really hard on fixing/implementing it. But I think we need >> that one before beta1. Maybe the only showstopper right now. > > I certainly think we should aim to include that - it's probably going > to be more popular than security labels. Probably by at least a factor 1,000 more popular. Actually, I'd be willing to be a lot more than just a factor 1,000... If we only get one of them, focus *really* should be on the collations one. But it half the seclables one is already written, then yeah, finishing that one off first makes sense :-) -- Magnus Hagander Me: http://www.hagander.net/ Work: http://www.redpill-linpro.com/
Le 05/04/2011 10:24, Guillaume Lelarge a écrit : > Le 04/04/2011 21:53, Guillaume Lelarge a écrit : > [...] > The security label feature isn't hard to implement. I previously thought > we would need to add yet another tab, which bores me to death, but I had > another idea while working on the dialogs. Security label behaves more > or less like comments, so I intend to add support just like support for > comments was done. It should be a quick patch to write. > And I was wrong. You can have multiple seclabels for one object (one per provider). So it needs its own tab, and it won't be easy. > I'm not sure about the host-name/suffix stuff. There is a bug to fix in > the pg_hba.conf editor, could be part of it. > > I still didn't work on the "per column collation support" because Tom > was working really hard on fixing/implementing it. But I think we need > that one before beta1. Maybe the only showstopper right now. > Guess I'm gonna work on this one now :) -- Guillaume http://www.postgresql.fr http://dalibo.com
I would like to have "experimental support for wxWidgets 2.9" as a feature for this release - it might encourage some curious person to build against 2.9. I'll make sure that support hasn't broken tomorrow. It needs to be maintained going forward. -- Peter Geoghegan http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training and Services
On Sun, Apr 10, 2011 at 11:53 AM, Peter Geoghegan <peter@2ndquadrant.com> wrote: > I would like to have "experimental support for wxWidgets 2.9" as a > feature for this release - it might encourage some curious person to > build against 2.9. I'll make sure that support hasn't broken tomorrow. > It needs to be maintained going forward. Yes, we should definitely mention this prominently in the release notes etc. -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Le 10/04/2011 12:53, Peter Geoghegan a écrit : > I would like to have "experimental support for wxWidgets 2.9" as a > feature for this release - it might encourage some curious person to > build against 2.9. I'll make sure that support hasn't broken tomorrow. > It needs to be maintained going forward. > Sure. FYI, I have four patches to push. I intended to do that tuesday evening, but I can do that on monday evening if you prefer. This way, you could work tuesday with almost all the code for 1.14. -- Guillaume http://www.postgresql.fr http://dalibo.com
Le 10/04/2011 19:10, Dave Page a écrit : > On Sun, Apr 10, 2011 at 11:53 AM, Peter Geoghegan <peter@2ndquadrant.com> wrote: >> I would like to have "experimental support for wxWidgets 2.9" as a >> feature for this release - it might encourage some curious person to >> build against 2.9. I'll make sure that support hasn't broken tomorrow. >> It needs to be maintained going forward. > > Yes, we should definitely mention this prominently in the release notes etc. > +1 -- Guillaume http://www.postgresql.fr http://dalibo.com
Le 10/04/2011 22:58, Guillaume Lelarge a écrit : > Le 10/04/2011 12:53, Peter Geoghegan a écrit : >> I would like to have "experimental support for wxWidgets 2.9" as a >> feature for this release - it might encourage some curious person to >> build against 2.9. I'll make sure that support hasn't broken tomorrow. >> It needs to be maintained going forward. >> > > Sure. > > FYI, I have four patches to push. I intended to do that tuesday evening, > but I can do that on monday evening if you prefer. This way, you could > work tuesday with almost all the code for 1.14. > Well, actually, it seems we have two more weeks. 9.1 beta 1 is delayed, so I guess 1.14 beta 1 is delayed too. Which gives me some time to work on seclabels. -- Guillaume http://www.postgresql.fr http://dalibo.com
On 10 April 2011 21:58, Guillaume Lelarge <guillaume@lelarge.info> wrote: > FYI, I have four patches to push. I intended to do that tuesday evening, > but I can do that on monday evening if you prefer. This way, you could > work tuesday with almost all the code for 1.14. Okay, let's do that. -- Peter Geoghegan http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training and Services