Re: [Zope3-dev] Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
Stuart Bishop wrote: silence it (and open a bug report at the same time). The noisier test output is, the more likely you are to miss relevant information. Totally agreed, I was kinda shocked at how many deprecation errors Zope 2.9 ships with :-( Chris -- Simplistix - Content Management, Zope Python Consulting - http://www.simplistix.co.uk ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope3-dev] Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
Stephan Richter wrote: On Tuesday 21 March 2006 02:54, Chris Withers wrote: I particularly hate the fact that no real effort was put into backwards compatibility, not to mention those silly weird sort-of-fifty-dots-per-line thing that doesn't actually work. I think this is not fair. Jim has tried very, very hard to convert all the functionality in a backward-compatible way. ...except one of the most commonly used parameters ;-) cheers, Chris -- Simplistix - Content Management, Zope Python Consulting - http://www.simplistix.co.uk ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope3-dev] Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
Philipp von Weitershausen wrote: Tres Seaver wrote: I'm not sure what Chris meant, but the change to the visual output of the testrunner when running with dots seems gratuitous to me, as well -- I don't see any benefit to the indented, narrower output, Me neither, for what it's worth. Okay, Tres, can you give me a line number I can poke at to remove this? It seems like Jim doesn't care too much, and I see 3 people at least who don't like it. Zope 2.9 broke the 'confiugre-make' dance in several ways, due (I think) to the choice to bolt on^H^H^H^H^H^H^Hretrofit zpkg. Sort of. It didn't break configure make. It's just make install that was broken. I think Tres was assuming that was an integral part of it ;-) I still don't understand why people whine about make install being gone. The point of a checkout is that you have a full functional SVN working copy, not an installation source. If you want to install things, use a TGZ archive which lets you do make install perfectly fine. I've never installed Zope anywhere except on production servers anyway, and there you should obviously use releases. Tres and Jens have already made the comments I was going to, but just to note that I strongly agree with them... cheers, Chris -- Simplistix - Content Management, Zope Python Consulting - http://www.simplistix.co.uk ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 22 Mar 2006, at 06:15, Philipp von Weitershausen wrote: I still don't understand why people whine about make install being gone. The point of a checkout is that you have a full functional SVN working copy, not an installation source. If you want to install things, use a TGZ archive which lets you do make install perfectly fine. I've never installed Zope anywhere except on production servers anyway, and there you should obviously use releases. If you absolutely must use make install from a checkout (perhaps because you want to install the trunk somewhere), then you can make a TGZ first using zpkg. Though I still don't see the point of it. This is just like Stefan Richter's continuing arguments against things he just doesn't use, thus they must be useless. It doesn't wash. The configure/make/make install dance was the canonical and quick way to install from a checkout _or_ a tarball for years. This should not break, period, whether you're talking about a tarball or a checkout. How can you even do development work when you never install Zope except for on a production server, I don't get it. What the point of a checkout is should be left up to the individual user, by the way. With Zope before 2.9 I get both worlds: It is a pristine SVN working copy, and with the ability to run configure/make/ make install in a *separate* directory without leaving any artifacts in the source tree, it acted as a perfect installation source at the same time. Unfortunately that use case, installing from the source tree into a different place, was broken in 2.9 as well and I sorely miss it. Now I have to copy the whole source tree to the other place before running configure/make and then futz around to get the equivalenet of the broken make install. Extremely annoying. jens -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (Darwin) iD8DBQFEIRwPRAx5nvEhZLIRAkFbAKCO4NJrXJbnRbTHClssCD3oA3pNpACgjAog CukwEUhM6ZIWPW3f9U/q0jU= =HUtg -END PGP SIGNATURE- ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope3-dev] Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
Jim Fulton wrote:
From the old testrunner, which I miss *a lot*, I could ensure I am
indeed running a specific module by doing...
Yup, this is one of the things I like least from the Zope 3 world.
What happened to proposals and community agreement before inflicting
big changes on other people who're trying to help out?
Oh cut the crap.
Hmm, I'm confused by this. If there's a proposal, my bad, point me at
it. If there isn't, well, it's kinda odd to receive abuse for pointing
out that you aren't sticking to your own processes...
The new test runner tries very hard to be backward
compatible.
...but misses one of the most common use cases from the old one, and you
didn't seem particularly fussed about fixing this :-S
This breakage was not intentional. It was a bug. There is an
easy work around: just use the -m option.
It can't be that hard to put in some syntactic sugar to support this. I
was going to give it a shot myself but I ran out of time, and I worry
about things like the regex matching the old testrunner used to dowhen
using the missing option.
I particularly hate the fact that no real effort was put into
backwards compatibility, not to mention those silly weird
sort-of-fifty-dots-per-line thing that doesn't actually work.
What the heck are you talking about? What doesn't work?
Here's a literal screen dump:
C:\Zope\2.9iC:\Zope\2.9.1\bin\python.exe C:\Zope\2.9.1\bin\test.py
--config-file C:\Zope\2.9i\etc\zope.conf --keepbytecode
Parsing C:\Zope\2.9i\etc\zope.conf
Running tests at level 1
Running unit tests:
Running:
.C:\Zope\2.9.1\lib\python\OFS\Application.py:598:DeprecationWarning:
The zLOG package is deprecated and will be removed in Zope 2.11. Use the
Python logging module instead.
('New disk product detected, determining if we need '
.
.
Ran 63 tests with 0 failures and 0 errors in 6.009 seconds.
C:\Zope\2.9i
It looks bizarre having that carriage return in the middle of the row of
dots. What's the point of the change that Tres added his patch to avoid
seeing?
Chris
--
Simplistix - Content Management, Zope Python Consulting
- http://www.simplistix.co.uk
___
Zope-CMF maillist - Zope-CMF@lists.zope.org
http://mail.zope.org/mailman/listinfo/zope-cmf
See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope3-dev] Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
On Wed, Mar 22, 2006 at 06:25:41PM +0100, Martijn Faassen wrote: Anyway, a release and the development situation looking similar helps people actually work on the same codebase and structure, and not having to learn different ways of doing things as soon as they switch. Forcing context switches on people isn't a good idea. +1 -- Paul Winkler http://www.slinkp.com ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
Jens Vagelpohl wrote: From the old testrunner, which I miss *a lot*, I could ensure I am indeed running a specific module by doing... Yup, this is one of the things I like least from the Zope 3 world. What happened to proposals and community agreement before inflicting big changes on other people who're trying to help out? I particularly hate the fact that no real effort was put into backwards compatibility, not to mention those silly weird sort-of-fifty-dots-per-line thing that doesn't actually work. My other big bug bear is what I've read about make-make-install no longer working in 2.9. Why on earth was this just broken without any thought to the vast number of people who rely on this as a simple way to quickly get a new release onto a number of boxes? I've been very happy to see the majority of the stuff in Zope 3, but it seems some things like this are getting rushed through without a lot of thought and consideration, which seems very un-zope-3 to me :-( Chris -- Simplistix - Content Management, Zope Python Consulting - http://www.simplistix.co.uk ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jim Fulton wrote: Chris Withers wrote: Jens Vagelpohl wrote: From the old testrunner, which I miss *a lot*, I could ensure I am indeed running a specific module by doing... Yup, this is one of the things I like least from the Zope 3 world. What happened to proposals and community agreement before inflicting big changes on other people who're trying to help out? Oh cut the crap. The new test runner tries very hard to be backward compatible. The old test runner was increasingly unmaintaiable and had a host of bugs of it's own. I can't tell what you snipped, but I'm guessing that it was the breakage of supplying a module name as a positional argument. This breakage was not intentional. It was a bug. There is an easy work around: just use the -m option. I particularly hate the fact that no real effort was put into backwards compatibility, not to mention those silly weird sort-of-fifty-dots-per-line thing that doesn't actually work. I'm not sure what Chris meant, but the change to the visual output of the testrunner when running with dots seems gratuitous to me, as well - -- I don't see any benefit to the indented, narrower output, especially because it's prettiness gets fouled up anyway by deprecation warnings, etc. I objected to the change enough to file collector issues: http://www.zope.org/Collectors/Zope/1958 http://www.zope.org/Collectors/Zope3-dev/493 What the heck are you talking about? What doesn't work? Zope 2.9 broke the 'confiugre-make' dance in several ways, due (I think) to the choice to bolt on^H^H^H^H^H^H^Hretrofit zpkg. E.g.: $ cd /tmp $ svn co svn+ssh://svn.zope.org/repos/main/Zope/trunk zt $ cd zt $ ./configure --prefix=/tmp/zt-sw make make install ... running install_scripts error: cannot copy tree 'build/scripts-2.4': not a directory make: *** [install] Error 1 $ make clobber inplace ... $ ls -laF bin ls: bin: No such file or directory (The 'bin' directory was, in Zope 2.7, 2.8, the location for the scripts, created during an inplace build). Tres. - -- === Tres Seaver +1 202-558-7113 [EMAIL PROTECTED] Palladion Software Excellence by Designhttp://palladion.com -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFEH/jP+gerLs4ltQ4RAhlOAKDBNhLZNaLvUn9bHOTBiIjBz2CYLQCeI6gp WD21iz7SSHQfCTOd0PxT22s= =k4TY -END PGP SIGNATURE- ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
Re: [Zope-CMF] Fighting the Zope 2.9 testrunner
Tres Seaver wrote: I'm not sure what Chris meant, but the change to the visual output of the testrunner when running with dots seems gratuitous to me, as well -- I don't see any benefit to the indented, narrower output, Me neither, for what it's worth. Zope 2.9 broke the 'confiugre-make' dance in several ways, due (I think) to the choice to bolt on^H^H^H^H^H^H^Hretrofit zpkg. Sort of. It didn't break configure make. It's just make install that was broken. I still don't understand why people whine about make install being gone. The point of a checkout is that you have a full functional SVN working copy, not an installation source. If you want to install things, use a TGZ archive which lets you do make install perfectly fine. I've never installed Zope anywhere except on production servers anyway, and there you should obviously use releases. If you absolutely must use make install from a checkout (perhaps because you want to install the trunk somewhere), then you can make a TGZ first using zpkg. Though I still don't see the point of it. Philipp ___ Zope-CMF maillist - Zope-CMF@lists.zope.org http://mail.zope.org/mailman/listinfo/zope-cmf See http://collector.zope.org/CMF for bug reports and feature requests
