Cutting branch for 1.30_00
Yanick Champoux
Trying to semi-stick to my insane goal of an official release a
month, I've cut a 1.31_00 development release branch for DBD::Oracle.
In the GitHub repo, that would be the candidate_1.31_00 branch, in
the svn repo it's rev14935
The changes for this release are minor. You can look in the
Changelog for the full list, but the highlights are:
* Makefile.PL options are now documented.
* explain, which was supposed to be removed as of 1.29, has been moved
to the examples directory.
* before, we were gagging the diag() output of the tests. I've removed
that odd behavior, and changed lots of 'diag()'s into 'notes()', as to
keep the verbosity of automatic tests to a sane level.
* couple of Makefile.PL/tests bug fixes.
* Oraperl's doc was mentioning that it would be removed in a future
version. I drew a line in the sand and fixed that future version to be
v1.38 (which, if all go well, we should hit in January 2012).
* removed a bevy of old configuration files dating from 2006 and earlier.
Could you be so kind as to give that pre-release a spin? Unless
there are reports of b0rkeness, I'll release the dev release on CPAN
Monday(ish), and the official follow-up release 1.32 at the end of the
month.
And, of course, if there's anything you don't agree with, now is
the time to raise your hand.
Joy,
`/anick
--
Yanick Champoux, Senior Perl Developer
The Pythian Group - love your data
http://www.pythian.com
--
The best compliment you could give Pythian for our service is a referral.
Martin J. Evans
> Hi all,
>
> Trying to semi-stick to my insane goal of an official release a month, I've cut a 1.31_00 development release branch for DBD::Oracle.
>
> In the GitHub repo, that would be the candidate_1.31_00 branch, in the svn repo it's rev14935
>
> The changes for this release are minor. You can look in the Changelog for the full list, but the highlights are:
>
> * Makefile.PL options are now documented.
>
> * explain, which was supposed to be removed as of 1.29, has been moved to the examples directory.
>
> * before, we were gagging the diag() output of the tests. I've removed that odd behavior, and changed lots of 'diag()'s into 'notes()', as to keep the verbosity of automatic tests to a sane level.
>
> * couple of Makefile.PL/tests bug fixes.
>
> * Oraperl's doc was mentioning that it would be removed in a future version. I drew a line in the sand and fixed that future version to be v1.38 (which, if all go well, we should hit in January 2012).
>
> * removed a bevy of old configuration files dating from 2006 and earlier.
>
>
> Could you be so kind as to give that pre-release a spin? Unless there are reports of b0rkeness, I'll release the dev release on CPAN Monday(ish), and the official follow-up release 1.32 at the end of the month.
>
> And, of course, if there's anything you don't agree with, now is the time to raise your hand.
>
> Joy,
> `/anick
>
I usually reserve note for noteworthy output not test progress or operations e.g.,
t/31lob_extended.t ...... 1/31 # procedure p_DBD_Oracle_drop_me dropped
# table dbd_ora__drop_me dropped
t/55nested.t ............ 1/29 # Fetched 4900 rows (select): 0.0282959938049316 secs.
# Fetched 4900 rows (nested cursor): 0.0298049449920654 secs.
pollutes the test output and tells someone not a working on DBD::Oracle itself what exactly?
on the other hand:
t/40ph_type.t ........... 1/19 Placeholder behaviour for ora_type=1 VARCHAR2 (the default) varies with Oracle version.
Oracle 7 didn't strip trailing spaces, Oracle 8 did, until 9.2.x
Your system doesn't. If that seems odd, let us know.
Seems noteworthy to me.
It is easy to run prove -vb t/x.t to get verbose output if you want it but most people don't want it - they just want a it works or does not work.
Martin
Martin J. Evans
>
> Trying to semi-stick to my insane goal of an official release a month, I've cut a 1.31_00 development release branch for DBD::Oracle.
>
> In the GitHub repo, that would be the candidate_1.31_00 branch, in the svn repo it's rev14935
>
> The changes for this release are minor. You can look in the Changelog for the full list, but the highlights are:
>
> * Makefile.PL options are now documented.
>
> * explain, which was supposed to be removed as of 1.29, has been moved to the examples directory.
>
> * before, we were gagging the diag() output of the tests. I've removed that odd behavior, and changed lots of 'diag()'s into 'notes()', as to keep the verbosity of automatic tests to a sane level.
>
> * couple of Makefile.PL/tests bug fixes.
>
> * Oraperl's doc was mentioning that it would be removed in a future version. I drew a line in the sand and fixed that future version to be v1.38 (which, if all go well, we should hit in January 2012).
>
> * removed a bevy of old configuration files dating from 2006 and earlier.
>
>
> Could you be so kind as to give that pre-release a spin? Unless there are reports of b0rkeness, I'll release the dev release on CPAN Monday(ish), and the official follow-up release 1.32 at the end of the month.
>
> And, of course, if there's anything you don't agree with, now is the time to raise your hand.
>
> Joy,
> `/anick
>
Martin
Yanick Champoux
> Actually, I don't like those diag to note changes much as it pollutes
> the normal test output.
You do realize that diag is more verbose than note, right?
diag: Prints a diagnostic message which is guaranteed not to interfere
with test output.
note: Like diag(), except the message will not be seen when the test is
run in a harness.
(from the Test::More pod)
I think that the change that is more the pressure point here is
commit 13a9e16ff51f1c0f5,
where I ungagged diag().
From what I can see, somebody realized at some point that diag()
was outputting an insane amount of information, and disabled all diag()
messages unless TEST_VERBOSE is set to true (or, as you mentioned, call
prove with '-v'). I removed that part, because it feels wrong to me to
change the default behavior of Test::More to accommodate an...
interpretative use of 'diag()' (the expression "two wrongs don't make a
right" comes to mind :-) ). And I changed lots of the diag()s that
were obviously not relevant diagnostics into notes.
And while I removed some of the noisier and less useful note()s
already, there are some that are still around as diag()s that could be
filtered out. I intend to do that as an ongoing maintenance activity.
> I usually reserve note for noteworthy output not test progress or
> operations e.g.,
>
> t/31lob_extended.t ...... 1/31 # procedure p_DBD_Oracle_drop_me dropped
> # table dbd_ora__drop_me dropped
> t/55nested.t ............ 1/29 # Fetched 4900 rows (select):
> 0.0282959938049316 secs.
> # Fetched 4900 rows (nested cursor): 0.0298049449920654 secs.
>
> pollutes the test output and tells someone not a working on
> DBD::Oracle itself what exactly?
Absolutely nothing useful. I totally agree, and that's going to
disappear as soon as I have a few more cycles to do it.
> It runs all tests 100% on the 3 clients and servers I have access to
> which covers a variety of Linuxes but and Oracle 11.*
Woohoo! \o/
Cheers,
`/anick
--
Yanick Champoux, Senior Perl Developer
The Pythian Group - love your data
http://www.pythian.com
--
Pythian at Oracle OpenWorld: 8 sessions packed with hot tips, real-world experiences and valuable insight. bit.ly/pythianoow11