Обсуждение: DRAFT beta release announcement
All,
Please submit revisions and corrections in the next 48 hours, thanks:
==============================
DATE: The first beta release of PostgreSQL version 9.0 is now
available. Version 9.0 is the first version of PostgreSQL to inclue
built-in binary database replication, consisting of two features, "Hot
Standby" and "Streaming Replication". This new replication, as well as
several other radical new features, will drive adoption of PostgreSQL by
new users and in new types of applications.
This release is a beta version. This means that it is expected to have
bugs, issues, and missing documentation. 9.0 beta is being released so
that our users will find those issues and help eliminate them in time
for the final release. All PostgreSQL users are emphatically requested
to download and test version 9.0 to help us achieve a faster and more
trouble-free 9.0 release.
In addition to the new replication, a number of new features will allow
developers and DBAs to broaden their use of PostgreSQL, including:
* Support for 64-bit Windows
* Redesigned LISTEN/NOTIFY built-in event notifications
* The DO() statement, allowing users to execute ad-hoc procedural statements
* Conditional and Column Triggers
* Numerous PL/Perl improvements
* Uniqueness constraints for non-scalar data
* Improved key-value data support
The full list of over 200 changes is available in the release notes:
Users interested in taking advantage of these new features should
download and test porting their applications to PostgreSQL 9.0 beta now.
Because of the many new major features, as well as overhauled internal
code, in version 9.0, there are expected to be a number of
backwards-compatibility issues.
Source code, as well as binary installers for many platforms, is
available from the PostgreSQL Web Site:
* Source:
* Windows Installer:
* Binaries for other platforms:
* Release Notes:
* Participating in Testing:
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
-----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 > DATE: The first beta release of PostgreSQL version 9.0 is now > available. Version 9.0 is the first version of PostgreSQL to inclue s/inclue/include/; yum install aspell :) > built-in binary database replication, consisting of two features, "Hot > Standby" and "Streaming Replication". This new replication, as well as > several other radical new features, will drive adoption of PostgreSQL by > new users and in new types of applications. "other radical new features" is a strong claim: can we back it up? > This release is a beta version. This means that it is expected to have > bugs, issues, and missing documentation. 9.0 beta is being released so > that our users will find those issues and help eliminate them in time > for the final release. All PostgreSQL users are emphatically requested > to download and test version 9.0 to help us achieve a faster and more > trouble-free 9.0 release. Conflating ideas here: the "faster" is how quickly 9.0 is released, and the "trouble-free" refers to 9.0 itself. The sentence currently says that users will make 9.0 itself faster. > * Redesigned LISTEN/NOTIFY built-in event notifications Redesigned is not a power statement. Perhaps "improved"? > * Conditional and Column Triggers Column-based? > * Numerous PL/Perl improvements So vague it should be at the end of the list > * Uniqueness constraints for non-scalar data Could be written better. * Improved key-value data support Very vague: can we be more specific? > Because of the many new major features, as well as overhauled internal > code, in version 9.0, there are expected to be a number of > backwards-compatibility issues. Remove the first comma. Is the "overhauled internal code" any more than any other release? Sentence still reads funny even with the comma removed... Might also want to mention how to give feedback (I presume -bugs?) - -- Greg Sabino Mullane greg@turnstep.com End Point Corporation http://www.endpoint.com/ PGP Key: 0x14964AC8 201004271342 http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8 -----BEGIN PGP SIGNATURE----- iEYEAREDAAYFAkvXIpwACgkQvJuQZxSWSsgqfACfa7KuE3wzXH8+bFr/CqYb++A3 6aEAmwbySUlFcEXx4qzwwsNqAlc3SV+s =dfaM -----END PGP SIGNATURE-----
On Tue, 2010-04-27 at 10:29 -0700, Josh Berkus wrote: > In addition to the new replication, a number of new features will allow > developers and DBAs to broaden their use of PostgreSQL, including: > > * Support for 64-bit Windows This should be something like, "64-bit Support on Windows", we have always has support for 64-bit Windows. > * Redesigned LISTEN/NOTIFY built-in event notifications > * The DO() statement, allowing users to execute ad-hoc procedural statements DO() support, allowing for inline? execution of procedural statements. Not quite sure about this one. I don't like the word ad-hoc. Dynamic? Sincerely, Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 503.667.4564 Consulting, Training, Support, Custom Development, Engineering
On Tue, April 27, 2010 19:45, Joshua D. Drake wrote: > On Tue, 2010-04-27 at 10:29 -0700, Josh Berkus wrote: >> * The DO() statement, allowing users to execute ad-hoc procedural statements > > DO() support, allowing for inline? execution of procedural statements. > > Not quite sure about this one. I don't like the word ad-hoc. Dynamic? > That other db uses the term 'anonymous block' [1] Maybe not so strong from an advocacy viewpoint, but it's a good description. [1] http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14251/adfns_packages.htm#i1006270
Greg Sabino Mullane wrote: > "other radical new features" is a strong claim: can we back it up? > I just read the release notes again, and I don't think a statement that strong really works there either. There are tons of great features of various sizes here, none of which besides the replication stuff really are "radical" though. "Totally awesome", sure, perhaps even "tubular", but not quite "radical" I think. The description of the replication features needed to be punched up even more though, it wasn't clear what the new capabilities were and why there were so important. Below is a candidate second draft. Along with the above, I did some wording updates to reflect two Gregs worth of ideas and the other comments on the list so far, changes were too numerous to be worth enumerating individually. I cleared up the DO description without using any buzzwords too. And I specifically highlighted the internal PL/PgSQL modifications in this version as ones expected to introduce backward compatibility issues. That's something I think could use a little migration guide in the docs, because it's not really obvious what the implications of the changes noted in the release notes are. === DATE: The first beta release of PostgreSQL version 9.0 is now available. Version 9.0 is the first version of PostgreSQL to include built-in real-time binary database replication with query scale-out, consisting of two features, "Hot Standby" and "Streaming Replication". Combining that with its other major features, this release will expand adoption of PostgreSQL by new users and in new types of applications. This release is a beta version. This means that it is expected to have bugs, issues, and missing documentation. 9.0 beta is being released so that our users will find those issues and allow eliminating them before the final release. All PostgreSQL users are emphatically requested to download and test version 9.0 to help us produce a timely and more trouble-free 9.0 release. In addition to the expanded replication features, a number of new features allow developers and DBAs to broaden their use of PostgreSQL, including: * 64-bit support on Windows * Improved LISTEN/NOTIFY allows fast internal database event messaging * The DO() statement executes procedural code without needing to declare a function * Conditional and SQL-compliant per-column triggers * Support for Python 3 in in PL/Python and numerous PL/Perl improvements * Uniqueness constraints for non-scalar data * Improved key-value data support The full list of over 200 changes is available in the release notes: http://developer.postgresql.org/pgdocs/postgres/release-9-0.html Users interested in taking advantage of these new features should download and test porting their applications to PostgreSQL 9.0 beta now. Because version 9.0 includes many new major features as well as overhauled internal code, such as changes to the syntax allowed by the PL/PgSQL language, backwards compatiblility issues are expected. Source code, as well as binary installers for many platforms, is available from the PostgreSQL Web Site: * Source: * Windows Installer: * Binaries for other platforms: * Release Notes: * Participating in Testing: -- Greg Smith 2ndQuadrant US Baltimore, MD PostgreSQL Training, Services and Support greg@2ndQuadrant.com www.2ndQuadrant.us
On Tue, 2010-04-27 at 14:38 -0400, Greg Smith wrote:
> DATE: The first beta release of PostgreSQL version 9.0 is now
> available. Version 9.0 is the first version of PostgreSQL to include
> built-in real-time binary database replication with query scale-out,
> consisting of two features, "Hot Standby" and "Streaming Replication".
> Combining that with its other major features, this release will expand
> adoption of PostgreSQL by new users and in new types of applications.
How about "Along with it's other major features..."?
I like the focus on replication and the terminology in the second
sentence.
> This release is a beta version. This means that it is expected to have
> bugs, issues, and missing documentation. 9.0 beta is being released so
> that our users will find those issues and allow eliminating them before
> the final release. All PostgreSQL users are emphatically requested to
> download and test version 9.0 to help us produce a timely and more
> trouble-free 9.0 release.
I would remove or replace the word "emphatically". It sounds a little
overstated, especially because the sentence is passively worded.
I'd also remove the word "more" and just say "trouble-free 9.0 release".
Regards,
Jeff Davis
On Tue, 2010-04-27 at 10:29 -0700, Josh Berkus wrote:
> * Uniqueness constraints for non-scalar data
If that's describing Exclusion Constraints, I don't know if that's the
best description.
If you're looking for a quick, one-line description, perhaps:
"Non-scalar/spatial index constraints (Exclusion Constraints)"
Without the words "Exclusion Constraints," nobody would even know where
to look in the documentation. Something about GiST could be mentioned,
but that might make it sound like it only works on GiST.
Also, "Deferrable UNIQUE" is a nice feature that doesn't need much
explanation, and would seem to fit on a beta announcement.
Regards,
Jeff Davis
jd@commandprompt.com ("Joshua D. Drake") writes:
>> * The DO() statement, allowing users to execute ad-hoc procedural statements
>
> DO() support, allowing for inline? execution of procedural statements.
>
> Not quite sure about this one. I don't like the word ad-hoc. Dynamic?
Hmm. How about...
DO() enables users to execute procedural statements without requiring
assigning function names.
That's clearer, though a bit wordier.
--
"cbbrowne","@","gmail.com"
http://linuxdatabases.info/info/slony.html
"I am a bomb technician. If you see me running, try to keep up..."
On 4/27/10 2:47 PM, Chris Browne wrote:
> jd@commandprompt.com ("Joshua D. Drake") writes:
>>> * The DO() statement, allowing users to execute ad-hoc procedural statements
>> DO() support, allowing for inline? execution of procedural statements.
>>
>> Not quite sure about this one. I don't like the word ad-hoc. Dynamic?
>
> Hmm. How about...
>
> DO() enables users to execute procedural statements without requiring
> assigning function names.
>
> That's clearer, though a bit wordier.
I'd really prefer something that fits into a bullet without wrapping.
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
On Tue, 2010-04-27 at 10:29 -0700, Josh Berkus wrote: > In addition to the new replication, a number of new features will allow > developers and DBAs to broaden their use of PostgreSQL, including: > > * Support for 64-bit Windows This should be something like, "64-bit Support on Windows", we have always has support for 64-bit Windows. > * Redesigned LISTEN/NOTIFY built-in event notifications > * The DO() statement, allowing users to execute ad-hoc procedural statements DO() support, allowing for inline? execution of procedural statements. Not quite sure about this one. I don't like the word ad-hoc. Dynamic? Sincerely, Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 503.667.4564 Consulting, Training, Support, Custom Development, Engineering
On tis, 2010-04-27 at 10:29 -0700, Josh Berkus wrote: > * The DO() statement, allowing users to execute ad-hoc procedural > statements The DO statement doesn't have any parentheses in its syntax.
On Tue, 2010-04-27 at 10:29 -0700, Josh Berkus wrote: > In addition to the new replication, a number of new features will allow > developers and DBAs to broaden their use of PostgreSQL, including: > > * Support for 64-bit Windows > * Redesigned LISTEN/NOTIFY built-in event notifications > * The DO() statement, allowing users to execute ad-hoc procedural statements > * Conditional and Column Triggers > * Numerous PL/Perl improvements > * Uniqueness constraints for non-scalar data > * Improved key-value data support I think the join removal feature is of critical importance, much more so than many of the above items. I might describe it like in *one* of these ways * Automatic join removal to optimise complex SQL generated by OR mapping * Major new optimizations for complex automatically generated SQL -- Simon Riggs www.2ndQuadrant.com
On Tue, Apr 27, 2010 at 6:20 PM, Josh Berkus <josh@agliodbs.com> wrote:
On 4/27/10 2:47 PM, Chris Browne wrote:I'd really prefer something that fits into a bullet without wrapping.
> jd@commandprompt.com ("Joshua D. Drake") writes:
>>> * The DO() statement, allowing users to execute ad-hoc procedural statements
>> DO() support, allowing for inline? execution of procedural statements.
>>
>> Not quite sure about this one. I don't like the word ad-hoc. Dynamic?
>
> Hmm. How about...
>
> DO() enables users to execute procedural statements without requiring
> assigning function names.
>
> That's clearer, though a bit wordier.
+1 for anonymous blocks. It's a common term in the industry, and may raise more eyebrows than 'ad-hoc' or 'dynamic'
--Scott
Sent via pgsql-advocacy mailing list (pgsql-advocacy@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-advocacy
> I think the join removal feature is of critical importance, much more so
> than many of the above items.
I just don't have any figures on actual performance benefit from this,
which will be the *first* thing a reporter asks.
> I might describe it like in *one* of these ways
>
> * Automatic join removal to optimise complex SQL generated by OR mapping
>
> * Major new optimizations for complex automatically generated SQL
How about
* Optimization of ORM-generated queries through automatic join removal
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
Josh Berkus wrote: > How about > * Optimization of ORM-generated queries through automatic join removal It might be worthwhile to reword this so it's more obvious that the feature is good for things other than just ORM optimization, maybe like this: * Automatic join removal optimization, effective for ORM-generated queries -- Greg Smith 2ndQuadrant US Baltimore, MD PostgreSQL Training, Services and Support greg@2ndQuadrant.com www.2ndQuadrant.us
On Wed, 2010-04-28 at 14:38 -0700, Josh Berkus wrote: > > I think the join removal feature is of critical importance, much more so > > than many of the above items. > > I just don't have any figures on actual performance benefit from this, > which will be the *first* thing a reporter asks. "Orders of magnitude" saving. It's not *faster* it just doesn't do the work at all any more. > > I might describe it like in *one* of these ways > > > > * Automatic join removal to optimise complex SQL generated by OR mapping > > > > * Major new optimizations for complex automatically generated SQL > > How about > > * Optimization of ORM-generated queries through automatic join removal Good. All 3 do it for me, but I think I'm too close to judge. -- Simon Riggs www.2ndQuadrant.com
On Wed, Apr 28, 2010 at 5:07 PM, Greg Smith <greg@2ndquadrant.com> wrote: > Josh Berkus wrote: >> >> How about >> * Optimization of ORM-generated queries through automatic join removal > > It might be worthwhile to reword this so it's more obvious that the feature > is good for things other than just ORM optimization, maybe like this: > > * Automatic join removal optimization, effective for ORM-generated queries > > this sounds better -- Atentamente, Jaime Casanova Soporte y capacitación de PostgreSQL Asesoría y desarrollo de sistemas Guayaquil - Ecuador Cel. +59387171157
All, Here's what I have now: ========================= DATE: The first beta release of PostgreSQL version 9.0 is now available. Version 9.0 is the first version of PostgreSQL to include built-in real-time binary database replication with query scale-out, consisting of two features, "Hot Standby" and "Streaming Replication". Combined with its other major features, this release will expand adoption of PostgreSQL by new users and in new types of applications. This release is a beta version. This means that it is expected to have bugs, issues, and missing documentation. 9.0 beta is being released so that our users will find those issues and allow eliminating them before the final release. The PostgreSQL Global Development Group requests that all users download and test version 9.0 to help us produce a timely and trouble-free 9.0 release. In version 9.0, a large number of new features will allow developers and DBAs to broaden their use of PostgreSQL, including: * New binary replication * 64-bit support on Windows * Improved LISTEN/NOTIFY allows fast internal database event messaging * Anonymous procedure blocks with the DO statement * Conditional and SQL-compliant per-column triggers * Support for Python 3 in in PL/Python and numerous PL/Perl improvements * Uniqueness constraints for non-scalar data (exclusion constraints) * Improved key-value data support * Automatic join removal, optimizing for ORM-generated queries The full list of over 200 changes is available in the release notes: http://developer.postgresql.org/pgdocs/postgres/release-9-0.html Users interested in taking advantage of these new features should download and test porting their applications to PostgreSQL 9.0 beta now. Because version 9.0 includes many new major features as well as overhauled internal code, such as changes to the syntax allowed by the PL/PgSQL language, backwards compatiblility issues are expected. Source code, as well as binary installers for many platforms, is available from the PostgreSQL Web Site: * Beta Information Page: http://www.postgresql.org/developer/beta * Source: http://www.postgresql.org/ftp/source/9.0beta1 * Windows Installer: http://www.enterprisedb.com/products/pgdownload.do * Binaries for other platforms: http://www.postgresql.org/ftp/binary/9.0beta1 * Release Notes: http://developer.postgresql.org/pgdocs/postgres/release-9-0.html -- -- Josh Berkus PostgreSQL Experts Inc. http://www.pgexperts.com
On tor, 2010-04-29 at 14:48 -0700, Josh Berkus wrote: > * Conditional and SQL-compliant per-column triggers That seems pretty confusing. Are the per-column triggers conditional and SQL-compliant? (Answer: No, they're two separate features.) Given that our trigger support isn't really that SQL-conforming in the first place, I suggest you write just * Conditional and per-column triggers
On Wed, Apr 28, 2010 at 11:15:52PM +0100, Simon Riggs wrote: > On Wed, 2010-04-28 at 14:38 -0700, Josh Berkus wrote: > > > I think the join removal feature is of critical importance, much more so > > > than many of the above items. > > > > I just don't have any figures on actual performance benefit from this, > > which will be the *first* thing a reporter asks. > > "Orders of magnitude" saving. It's not *faster* it just doesn't do the > work at all any more. > > > > I might describe it like in *one* of these ways > > > > > > * Automatic join removal to optimise complex SQL generated by OR mapping > > > > > > * Major new optimizations for complex automatically generated SQL > > > > How about > > > > * Optimization of ORM-generated queries through automatic join removal > > Good. All 3 do it for me, but I think I'm too close to judge. Is it worth considering that more than just ORMs are responsible for machine-generated SQL, and just saying "machine-generated" instead of "ORM-generated"? Provided, that is, that the extra four characters don't make the line wrap :) -- Joshua Tolley / eggyknap End Point Corporation http://www.endpoint.com
Вложения
On 5/4/10 4:49 AM, Jussi Mikkola wrote: > Hi, > > http://www.tietokone.fi/uutiset/avoin_postgresql_tietokanta_uudistuu > > Tietokone magazine, the largest computer magazine in Finland, has noticed the beta release already. (With a little help;-) > > All the positive text about PostgreSQL is written by them. Let's see, if it gets into the printed magazine too. Wow, great! Good work, Jussi. -- -- Josh Berkus PostgreSQL Experts Inc. http://www.pgexperts.com