Hi, I have the feeling that the MANIFEST file is not always generated from the MANIFEST.in, especially if MANIFEST.in has not changed since the previous generation. Is this the right behaviour? I was recently bitten by this when shipping a bugfix release: new files had been added which should have made it to the tarball, but since the MANIFEST file was not regenerated, this did not happen. Is regenerating the MANIFEST file so expensive that this tiny optimisation is necessary? My experience with most package is that the time gained is not that big -- especially compared to the time spent tracking the missing files ;o) -- Alexandre Fayolle LOGILAB, Paris (France). http://www.logilab.com http://www.logilab.fr http://www.logilab.org Développement logiciel avancé - Intelligence Artificielle - Formations
Alexandre Fayolle wrote:
Hi,
I have the feeling that the MANIFEST file is not always generated from the MANIFEST.in, especially if MANIFEST.in has not changed since the previous generation. Is this the right behaviour?
I was recently bitten by this when shipping a bugfix release: new files had been added which should have made it to the tarball, but since the MANIFEST file was not regenerated, this did not happen.
Is regenerating the MANIFEST file so expensive that this tiny optimisation is necessary? My experience with most package is that the time gained is not that big -- especially compared to the time spent tracking the missing files ;o)
I always delete the MANIFEST before building a new release. That seems to do the trick. -- Marc-Andre Lemburg eGenix.com Professional Python Software directly from the Source (#1, Sep 25 2003)
Python/Zope Products & Consulting ... http://www.egenix.com/ mxODBC, mxDateTime, mxTextTools ... http://python.egenix.com/
::: Try mxODBC.Zope.DA for Windows,Linux,Solaris,FreeBSD for free ! ::::
On Thu, Sep 25, 2003 at 10:50:56AM +0200, M.-A. Lemburg wrote:
I always delete the MANIFEST before building a new release. That seems to do the trick.
Of course, this does the trick. But doing so also feels a lot like working around a bug or an unwanted feature. If people agree on that, I can work on a patch to change this behaviour. Unless of course some distutil users actually rely on this. -- Alexandre Fayolle LOGILAB, Paris (France). http://www.logilab.com http://www.logilab.fr http://www.logilab.org Développement logiciel avancé - Intelligence Artificielle - Formations
On Thu, 25 Sep 2003, Alexandre Fayolle wrote:
On Thu, Sep 25, 2003 at 10:50:56AM +0200, M.-A. Lemburg wrote:
I always delete the MANIFEST before building a new release. That seems to do the trick.
Of course, this does the trick. But doing so also feels a lot like working around a bug or an unwanted feature.
If people agree on that, I can work on a patch to change this behaviour. Unless of course some distutil users actually rely on this.
-- Alexandre Fayolle
The sdist command has an option --force-manifest http://www.python.org/doc/current/dist/source-dist.html#SECTION0005200000000... You could write it in one of your prefered config files for distutils http://www.python.org/doc/current/inst/config-syntax.html So there is no need to make a patch for it. (Except maybe to better document this option so people find it better ;-) Kind regards Rene Liebscher
On Thu, Sep 25, 2003 at 11:48:59AM +0200, Rene Liebscher wrote:
The sdist command has an option --force-manifest http://www.python.org/doc/current/dist/source-dist.html#SECTION0005200000000...
You could write it in one of your prefered config files for distutils http://www.python.org/doc/current/inst/config-syntax.html
So there is no need to make a patch for it. (Except maybe to better document this option so people find it better ;-)
My sincerest apologies for not having reread the fine manual, and thanks a lot for pointing this out to me. I still think that the default behaviour is not the best one, but being able to write a setup.cfg file and forget about it is an acceptable trade off. -- Alexandre Fayolle LOGILAB, Paris (France). http://www.logilab.com http://www.logilab.fr http://www.logilab.org Développement logiciel avancé - Intelligence Artificielle - Formations
On Thu, 2003-09-25 at 06:01, Alexandre Fayolle wrote:
I still think that the default behaviour is not the best one, but being able to write a setup.cfg file and forget about it is an acceptable trade off.
It agree that the default behavior seems problematic. I always use sdist --force-manifest when building a release. What's the use case for ignoring files in a source distribution that would otherwise be included? It doesn't sound like an obvious default. Jeremy
participants (4)
-
Alexandre Fayolle -
Jeremy Hylton -
M.-A. Lemburg -
Rene Liebscher