
Something that just struck me as I kept running bdist_egg --exclude-source-files is that it seems like "bdist" would imply exclusion of source files. Perhaps there could be an sdist_egg that creates an egg that includes all of the source... Kevin

At 11:40 AM 8/9/2005 -0400, Kevin Dangoor wrote:
Something that just struck me as I kept running bdist_egg --exclude-source-files is that it seems like "bdist" would imply exclusion of source files.
Source is important for tracebacks and debugging, especially if an egg is installed unpacked. The --exclude-source-files option is intended to allow space conservation when creating applications for distribution, or for "IP protection" uses. These are both fairly specialized build situations, not the common case for distributing an open source library.

On 8/9/05, Phillip J. Eby <pje@telecommunity.com> wrote:
At 11:40 AM 8/9/2005 -0400, Kevin Dangoor wrote:
Something that just struck me as I kept running bdist_egg --exclude-source-files is that it seems like "bdist" would imply exclusion of source files.
Source is important for tracebacks and debugging, especially if an egg is installed unpacked. The --exclude-source-files option is intended to allow space conservation when creating applications for distribution, or for "IP protection" uses. These are both fairly specialized build situations, not the common case for distributing an open source library.
And it now occurs to me that "sdist_egg" doesn't make sense either, because the egg has compiled C modules and such. Your explanation about the debugging value of the source also explains the --always-unzip easy_install option, which I thought much about. (I happen to fall into the "space conservation for distributing applications" category, which I agree is somewhat specialized.) Thanks for the explanation. Kevin
participants (2)
-
Kevin Dangoor
-
Phillip J. Eby