Hi, 1) When running 'setup.py bdist_egg' the subdirectories build/ and dists/ are created. Do people typically add and mange these directories and their contents within subversion as well? Or do you typically have a separate place, outside of your subversion sandbox for building distributions? 2) I'm using epydoc to generate documentation for my module. Is it appropriate to include the resulting html/ directory and it's contents in the egg in some way and if so, how? -Val ------------------------------------------------------ Val Schmidt CCOM/JHC University of New Hampshire Chase Ocean Engineering Lab 24 Colovos Road Durham, NH 03824 e: vschmidt [AT] ccom.unh.edu m: 614.286.3726
At 11:55 AM 11/24/2008 -0500, Val Schmidt wrote:
Hi,
1) When running 'setup.py bdist_egg' the subdirectories build/ and dists/ are created. Do people typically add and mange these directories and their contents within subversion as well? Or do you typically have a separate place, outside of your subversion sandbox for building distributions?
It's usually a good idea to exclude those subdirectories using the svn:ignore property.
2) I'm using epydoc to generate documentation for my module. Is it appropriate to include the resulting html/ directory and it's contents in the egg in some way and if so, how?
No; that should usually be only included in your source distribution, unless the package itself (or something it plugs into) will use the documentation at runtime. (For example, a Zope "product" or a Trac plugin that need to include HTML help to be displayed when using the product or plugin.)
Thanks! I also noticed that my SOURCES.txt file contains the contents of mymodulepy.egg-info (which got added to my subversion repository when I initially added the project directory). I suspect I don't want this either. So does it seem like a reasonable working method to develop in one's subversion sandbox, check everything in, then checkout a copy of the project somewhere else to build distributions? This might help to make sure one doesn't inadvertently get the extra files and directories created during the build process into subversion and inadvertently in your distribution package. -Val On Nov 24, 2008, at 12:57 PM, Phillip J. Eby wrote:
At 11:55 AM 11/24/2008 -0500, Val Schmidt wrote:
Hi,
1) When running 'setup.py bdist_egg' the subdirectories build/ and dists/ are created. Do people typically add and mange these directories and their contents within subversion as well? Or do you typically have a separate place, outside of your subversion sandbox for building distributions?
It's usually a good idea to exclude those subdirectories using the svn:ignore property.
2) I'm using epydoc to generate documentation for my module. Is it appropriate to include the resulting html/ directory and it's contents in the egg in some way and if so, how?
No; that should usually be only included in your source distribution, unless the package itself (or something it plugs into) will use the documentation at runtime. (For example, a Zope "product" or a Trac plugin that need to include HTML help to be displayed when using the product or plugin.)
------------------------------------------------------ Val Schmidt CCOM/JHC University of New Hampshire Chase Ocean Engineering Lab 24 Colovos Road Durham, NH 03824 e: vschmidt [AT] ccom.unh.edu m: 614.286.3726
At 07:56 AM 11/25/2008 -0500, Val Schmidt wrote:
Thanks!
I also noticed that my SOURCES.txt file contains the contents of mymodulepy.egg-info (which got added to my subversion repository when I initially added the project directory). I suspect I don't want this either.
That depends on the project; in some cases, files are put manually into .egg-info, or there are reasons to keep the generated files under revision control. The .egg-info directory is also automatically included in source distribution files.
So does it seem like a reasonable working method to develop in one's subversion sandbox, check everything in, then checkout a copy of the project somewhere else to build distributions? This might help to make sure one doesn't inadvertently get the extra files and directories created during the build process into subversion and inadvertently in your distribution package.
I'm not clear on how you can accidentally get something into subversion. In years of developing multiple packages with setuptools and subversion, I can't recall that I've ever accidentally typed "svn add build" or "svn add foobar.egg-info". ;-) Of course, when I start a new project, I start by svn add-ing the new, empty directory, and then explicitly add files as I go. And when I start getting annoyed by svn stat showing a bunch of "?"'s, I go ahead and set svn:ignore as needed. It's a lot less work than what you're describing, and even if I go too far with a wildcard add, there's always "svn rm" before checkin. (I always svn diff before checkin, so it's hard for an accidental add to go unnoticed.)
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Phillip J. Eby wrote:
At 07:56 AM 11/25/2008 -0500, Val Schmidt wrote:
Thanks!
I also noticed that my SOURCES.txt file contains the contents of mymodulepy.egg-info (which got added to my subversion repository when I initially added the project directory). I suspect I don't want this either.
That depends on the project; in some cases, files are put manually into .egg-info, or there are reasons to keep the generated files under revision control. The .egg-info directory is also automatically included in source distribution files.
So does it seem like a reasonable working method to develop in one's subversion sandbox, check everything in, then checkout a copy of the project somewhere else to build distributions? This might help to make sure one doesn't inadvertently get the extra files and directories created during the build process into subversion and inadvertently in your distribution package.
I'm not clear on how you can accidentally get something into subversion. In years of developing multiple packages with setuptools and subversion, I can't recall that I've ever accidentally typed "svn add build" or "svn add foobar.egg-info". ;-)
Of course, when I start a new project, I start by svn add-ing the new, empty directory, and then explicitly add files as I go. And when I start getting annoyed by svn stat showing a bunch of "?"'s, I go ahead and set svn:ignore as needed.
It's a lot less work than what you're describing, and even if I go too far with a wildcard add, there's always "svn rm" before checkin. (I always svn diff before checkin, so it's hard for an accidental add to go unnoticed.)
The usual way I get such stuff into SVN is via 'svn import': I then have to 'svn rm' it (usually after seeing the checkin message). Tres. - -- =================================================================== Tres Seaver +1 540-429-0999 tseaver@palladion.com Palladion Software "Excellence by Design" http://palladion.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFJLC3c+gerLs4ltQ4RAng/AKCt1OTqM146bGsmGieljDfgCN5ZIQCePbGE s5H3JsxqbPdigXk9zq5rivI= =fztv -----END PGP SIGNATURE-----
participants (3)
-
Phillip J. Eby -
Tres Seaver -
Val Schmidt