Обсуждение: Stamping of 8.4
Tom did most of the 8.4 stamping and I incremented the library minor version numbers. I have updated the library bump wording in RELEASE_CHANGES: o Bump minor library versions, major if appropriate (see below) I also updated this item description: o update config.guess and config.sub at the start of beta I see this was done already by Peter during beta: date: 2007/11/15 20:21:04; author: petere; state: Exp; lines: +25 -9Update config.guess and config.sub so I removed the "at the start of beta" from above. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Sorry, please ignore me. I got "start of beta" confused with "start of development". --------------------------------------------------------------------------- Bruce Momjian wrote: > Tom did most of the 8.4 stamping and I incremented the library minor > version numbers. I have updated the library bump wording in > RELEASE_CHANGES: > > o Bump minor library versions, major if appropriate (see below) > > I also updated this item description: > > o update config.guess and config.sub at the start of beta > > I see this was done already by Peter during beta: > > date: 2007/11/15 20:21:04; author: petere; state: Exp; lines: +25 -9 > Update config.guess and config.sub > > so I removed the "at the start of beta" from above. > > -- > Bruce Momjian <bruce@momjian.us> http://momjian.us > EnterpriseDB http://postgres.enterprisedb.com > > + If your life is a hard drive, Christ can be your backup. + > > ---------------------------(end of broadcast)--------------------------- > TIP 4: Have you searched our list archives? > > http://archives.postgresql.org -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Bruce Momjian wrote: > Tom did most of the 8.4 stamping and I incremented the library minor > version numbers. I have updated the library bump wording in > RELEASE_CHANGES: > > o Bump minor library versions, major if appropriate (see below) I admit I am surprised by the library bump. We haven't done anything to the libraries yet, so why bump the versions? -- Alvaro Herrera http://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support
Alvaro Herrera wrote: > Bruce Momjian wrote: > > Tom did most of the 8.4 stamping and I incremented the library minor > > version numbers. I have updated the library bump wording in > > RELEASE_CHANGES: > > > > o Bump minor library versions, major if appropriate (see below) > > I admit I am surprised by the library bump. We haven't done anything to > the libraries yet, so why bump the versions? It is standard practice because we always modify the library in some way, and if we don't do the minor version libraries will not favor the new version. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Alvaro Herrera <alvherre@commandprompt.com> writes:
> I admit I am surprised by the library bump. We haven't done anything to
> the libraries yet, so why bump the versions?
Minor version bump is done unconditionally for each release cycle.
Major version bump is as-needed (ABI break).
regards, tom lane
Tom Lane wrote: > Alvaro Herrera <alvherre@commandprompt.com> writes: > > I admit I am surprised by the library bump. We haven't done anything to > > the libraries yet, so why bump the versions? > > Minor version bump is done unconditionally for each release cycle. Oh, I see. I will update RELEASE_CHANGES to that effect. -- Alvaro Herrera http://www.CommandPrompt.com/ PostgreSQL Replication, Consulting, Custom Development, 24x7 support
Alvaro Herrera wrote: > Tom Lane wrote: > > Alvaro Herrera <alvherre@commandprompt.com> writes: > > > I admit I am surprised by the library bump. We haven't done anything to > > > the libraries yet, so why bump the versions? > > > > Minor version bump is done unconditionally for each release cycle. > > Oh, I see. I will update RELEASE_CHANGES to that effect. I think I did that. Does it need more? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
* Tom Lane (tgl@sss.pgh.pa.us) wrote:
> Alvaro Herrera <alvherre@commandprompt.com> writes:
> > I admit I am surprised by the library bump. We haven't done anything to
> > the libraries yet, so why bump the versions?
>
> Minor version bump is done unconditionally for each release cycle.
This isn't a *huge* deal, but I'm not sure it's actually appropriate.
We should bump the minor version when we actually add something new to
the library (which is probably just about every time we do a major
version, but still).
Thanks,
Stephen
Stephen Frost wrote: -- Start of PGP signed section. > * Tom Lane (tgl@sss.pgh.pa.us) wrote: > > Alvaro Herrera <alvherre@commandprompt.com> writes: > > > I admit I am surprised by the library bump. We haven't done anything to > > > the libraries yet, so why bump the versions? > > > > Minor version bump is done unconditionally for each release cycle. > > This isn't a *huge* deal, but I'm not sure it's actually appropriate. > We should bump the minor version when we actually add something new to > the library (which is probably just about every time we do a major > version, but still). The problem is the risk of forgetting during development. When we break an API it is obvious, but improvements are so regular you can't remember when you first do it for each interface. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
Stephen Frost <sfrost@snowman.net> writes:
> * Tom Lane (tgl@sss.pgh.pa.us) wrote:
>> Minor version bump is done unconditionally for each release cycle.
> This isn't a *huge* deal, but I'm not sure it's actually appropriate.
> We should bump the minor version when we actually add something new to
> the library (which is probably just about every time we do a major
> version, but still).
Only if there were exactly zero changes would the minor version bump
be unnecessary. I doubt that's ever happened.
regards, tom lane
Bruce Momjian wrote: > Alvaro Herrera wrote: > > Tom Lane wrote: > > > Alvaro Herrera <alvherre@commandprompt.com> writes: > > > > I admit I am surprised by the library bump. We haven't done anything to > > > > the libraries yet, so why bump the versions? > > > > > > Minor version bump is done unconditionally for each release cycle. > > > > Oh, I see. I will update RELEASE_CHANGES to that effect. > > I think I did that. Does it need more? Yeah, you didn't touch the "Minor Version" subsection of "Library Version Changes". -- Alvaro Herrera http://www.CommandPrompt.com/ The PostgreSQL Company - Command Prompt, Inc.
Bruce Momjian <bruce@momjian.us> writes:
> Stephen Frost wrote:
>> We should bump the minor version when we actually add something new to
>> the library (which is probably just about every time we do a major
>> version, but still).
> The problem is the risk of forgetting during development. When we break
> an API it is obvious, but improvements are so regular you can't remember
> when you first do it for each interface.
We could possibly do the bump at the end of the cycle (eg, just before
beta) if no major bump has happened meanwhile. However, this would
complicate life for developers. I believe one of the arguments for the
immediate minor bump was so that you could tell a development library
from the previous release version, and (if your platform lets you) even
install them in parallel.
regards, tom lane
Tom Lane wrote: > Bruce Momjian <bruce@momjian.us> writes: > > Stephen Frost wrote: > >> We should bump the minor version when we actually add something new to > >> the library (which is probably just about every time we do a major > >> version, but still). > > > The problem is the risk of forgetting during development. When we break > > an API it is obvious, but improvements are so regular you can't remember > > when you first do it for each interface. > > We could possibly do the bump at the end of the cycle (eg, just before > beta) if no major bump has happened meanwhile. However, this would > complicate life for developers. I believe one of the arguments for the > immediate minor bump was so that you could tell a development library > from the previous release version, and (if your platform lets you) even > install them in parallel. Yes, a late bump would invalidate a lot of installations running CVS in testing. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +