Fix build #98
No reviewers
Labels
No labels
bug
duplicate
enhancement
help wanted
in progress
invalid
question
under discussion
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: coderkun/mcg#98
Loading…
Reference in a new issue
No description provided.
Delete branch "jfly/fix-build"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
When doing a
python setup.py build
on my machine, I found thatbuild/lib
would not end up with a compiled gresource file until thesecond invocation of
python setup.py build
.Before:
Note how there is no data directory at all. Now check out what happens
on the second build:
That's because the first build generated the compiled schemas and
resources (you can see evidence of that in
git status
), and then thesecond build was able to copy the gresource file over according to the
package_data
rules. The fix I've introduced here is to just do thecompilations before we call
super(...).run(...)
. There might bebetter ways of doing this, I'm not very familiar with packaging gtk
python applications.
Things were even worse for the gschemas.compiled file: in addition to
the ordering issue it's not even mentioned in
data_files
, so even ifit does exist, it doesn't have a chance to get copied over when
installed. So I've added it to the
data_files
section. I don't know ifthat'll play nicely or not with the existing
--no-compile-schemas
flag.
mentioned in merge request jflysohigh/mcg!1
Thanks for your patch, @jflysohigh, it looks good to me. I will test it next week or at the weekend.
Please note that I am also working on Meson (see #32), which will overhaul the build process.
Thanks, @coderkun. And good luck with meson!
The changes look good, I merged them into the main branch. I only removed the removal of a blank line.
Pull request closed