[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#33497: [PATCH] Re: bug#33497: Our CMake package has no documentation
From: |
Maxim Cournoyer |
Subject: |
bug#33497: [PATCH] Re: bug#33497: Our CMake package has no documentation |
Date: |
Wed, 20 Mar 2019 12:06:36 -0400 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/26.1 (gnu/linux) |
Marius Bakke <address@hidden> writes:
> Maxim Cournoyer <address@hidden> writes:
[...]
>>> I'm not very comfortable with pulling python-sphinx into the dependency
>>> closure of CMake, because then we can't update it or its dependencies
>>> outside of the 'core-updates' cycle. It could also cause circular
>>> dependency issues down the road.
>>>
>>> Would it make sense to build the documentation as a separate package?
>>> In that case it can go on the master branch.
>>
>> I don't like the idea of having a separate package for the documentation
>> of cmake because it goes against the expectations of Guix users (package
>> comes with its manpage, and extra doc can be installed as extra output
>> of the same package).
>>
>> Maybe we could have a "cmake-minimal" package we'd keep hidden and use
>> by default as part of the cmake-build-system, which wouldn't include the
>> doc, and the regular, user facing cmake would be the one in this patch?
>
> This sounds reasonable to me. Can you send a patch? :-)
Here it comes, attached! Thanks to Ricardo on #guix for helping me
resolve a quoting issue :-).
To validate it works, I had to disable some failing test suites, using:
diff --git a/gnu/packages/libffi.scm b/gnu/packages/libffi.scm
index f47f7623b4..259ea23a67 100644
--- a/gnu/packages/libffi.scm
+++ b/gnu/packages/libffi.scm
@@ -106,26 +106,26 @@ conversions for values passed between the two languages.")
(guix build python-build-system))
#:phases
(modify-phases %standard-phases
- (replace 'check
- (lambda _
- (setenv "PYTHONPATH"
- (string-append
- (getenv "PYTHONPATH")
- ":" (getcwd) "/build/"
- (car (scandir "build" (cut string-prefix? "lib." <>)))))
+ ;; (replace 'check
+ ;; (lambda _
+ ;; (setenv "PYTHONPATH"
+ ;; (string-append
+ ;; (getenv "PYTHONPATH")
+ ;; ":" (getcwd) "/build/"
+ ;; (car (scandir "build" (cut string-prefix? "lib."
<>)))))
- ;; XXX The "normal" approach of setting CC and friends does
- ;; not work here. Is this the correct way of doing things?
- (substitute* "testing/embedding/test_basic.py"
- (("c = distutils\\.ccompiler\\.new_compiler\\(\\)")
- (string-append "c = distutils.ccompiler.new_compiler();"
- "c.set_executables(compiler='gcc',"
- "compiler_so='gcc',linker_exe='gcc',"
- "linker_so='gcc -shared')")))
- (substitute* "testing/cffi0/test_ownlib.py"
- (("'cc testownlib") "'gcc testownlib"))
- (invoke "py.test" "-v" "c/" "testing/")
- #t))
+ ;; ;; XXX The "normal" approach of setting CC and friends does
+ ;; ;; not work here. Is this the correct way of doing things?
+ ;; (substitute* "testing/embedding/test_basic.py"
+ ;; (("c = distutils\\.ccompiler\\.new_compiler\\(\\)")
+ ;; (string-append "c = distutils.ccompiler.new_compiler();"
+ ;; "c.set_executables(compiler='gcc',"
+ ;; "compiler_so='gcc',linker_exe='gcc',"
+ ;; "linker_so='gcc -shared')")))
+ ;; (substitute* "testing/cffi0/test_ownlib.py"
+ ;; (("'cc testownlib") "'gcc testownlib"))
+ ;; (invoke "py.test" "-v" "c/" "testing/")
+ ;; #t))
(add-before 'check 'disable-failing-test
;; This is assumed to be a libffi issue:
;;
https://bitbucket.org/cffi/cffi/issues/312/tests-failed-with-armv8
diff --git a/gnu/packages/python.scm b/gnu/packages/python.scm
index 0a483fb1db..67de01573e 100644
--- a/gnu/packages/python.scm
+++ b/gnu/packages/python.scm
@@ -127,7 +127,8 @@
"tk")) ;tkinter; adds 50 MiB to the closure
(build-system gnu-build-system)
(arguments
- `(#:test-target "test"
+ `(#:tests? #f
+ #:test-target "test"
#:configure-flags
(list "--enable-shared" ;allow embedding
"--with-system-ffi" ;build ctypes
0001-cmake-Generate-documentation.patch
Description: Text Data
Thanks!
Maxim