Created attachment 208509 [details] repocopy python37-->python38 Repocopy lang/python37 --> lang/python38 and update to 3.8.0
Need to integrate this with https://reviews.freebsd.org/D19822
> Need to integrate this with https://reviews.freebsd.org/D19822 Umm, why? This patch supersedes that review completely, it could just be dropped.
"if / where necessary" I didn't review each patchset
Any reason why --enable-optimizations isn't used?
I've tested it with some ports and it works great. Anything delaying the commit?
Review, confirmation of QA & python@ approval on attachment 208509 [details] as is normal
A commit references this bug: Author: wen Date: Sat Nov 2 23:40:42 UTC 2019 New revision: 516377 URL: https://svnweb.freebsd.org/changeset/ports/516377 Log: - Repocopy lang/python37 --> python38 and update to 3.8.0 PR: 241416 Submitted by: wen@(myself) Reviewed by: koobs@, amdmi3@ Changes: head/Mk/Uses/python.mk head/Mk/bsd.default-versions.mk head/lang/Makefile head/lang/python-doc-html/distinfo head/lang/python38/ head/lang/python38/Makefile head/lang/python38/Makefile.version head/lang/python38/distinfo head/lang/python38/files/patch-Makefile.pre.in head/lang/python38/files/patch-issue20210 head/lang/python38/pkg-plist
(In reply to commit-hook from comment #7) Note: I didn't review this, and still have questions for understanding the removal of files/patch-Makefile.pre.in
(In reply to Kubilay Kocak from comment #8) Sorry for add you to reviewer. But files/patch-Makefile.pre.in had not been removed. wen
There are problems: - _sqlite3 _tkinter _gdbm are no longer disabled - with DEBUG turned on it fails to package
(In reply to Antoine Brodin from comment #10) Thank you ! I shall fix it later. wen
(In reply to Wen Heping from comment #9) Apologies, I meant files/patch-Makefile.pre.in *modification*, with removal of the block that creates symlinks for non-ABI-suffixed files: # Description: Create symlinks for non-ABI-suffixed python*-config and ABI-suffixed python-*.pc # Submitted by: antoine@ (r358029) # TODO: Upstream Diff: https://svnweb.freebsd.org/ports/head/lang/python38/files/patch-Makefile.pre.in?r1=516377&r2=516375&pathrev=516377
(In reply to Kubilay Kocak from comment #12) Thank you again! The removal of the patch(files/patch-Makefile.pre.in modification) which creatd by antoine@ will result in package error, I shall re-test it later. wen
(In reply to Antoine Brodin from comment #10) See bug 243937 for a fix for disabling extensions.
Upgrading Python to 3.8.10 as a port in developer mode displays the following errors: ====> Running Q/A tests (stage-qa) Error: /usr/local/lib/python3.8/lib-dynload/_gdbm.cpython-38.so is linked to /usr/local/lib/libgdbm.so.6 from databases/gdbm but it is not declared as a dependency Warning: you need LIB_DEPENDS+=libgdbm.so:databases/gdbm Error: /usr/local/lib/python3.8/lib-dynload/_sqlite3.cpython-38.so is linked to /usr/local/lib/libsqlite3.so.0 from databases/sqlite3 but it is not declared as a dependency Warning: you need USES+=sqlite Error: /usr/local/lib/python3.8/lib-dynload/_tkinter.cpython-38_failed.so is linked to /usr/local/lib/libtk86.so.1 from x11-toolkits/tk86 but it is not declared as a dependency Warning: you need USES+=tk Error: /usr/local/lib/python3.8/lib-dynload/_tkinter.cpython-38_failed.so is linked to /usr/local/lib/libtcl86.so.1 from lang/tcl86 but it is not declared as a dependency Warning: you need USES+=tcl Error: /usr/local/lib/python3.8/lib-dynload/_tkinter.cpython-38_failed.so is linked to /usr/local/lib/libX11.so.6 from x11/libX11 but it is not declared as a dependency Warning: you need USE_XORG+=x11 readelf: Not an ELF file. Warning: /usr/ports/lang/python38/work/stage/usr/local/lib/libpython3.8.so.1.0-gdb.py doesn't have a SONAME. Warning: pkg(8) will not register it as being provided by the port. Warning: If another port depend on it, pkg will not be able to know where it comes from. Warning: It is directly in /usr/local/lib, it is probably used by other ports. That means that python3.8 from the port and python3.87 from a package might be different.
Status on this PR?
I think this can be closed. This port exists. Most if not all of the problems raised here have been fixed one way or another. Call it fixed or overcome by events or whatever. If there are specific issues with the port, they should be addressed in a separate bug.
(In reply to John Hein from comment #17) Now I closed this PR. Thank you !