Bug 240030
| Summary: | dns/py-tldextract: Update to 2.2.1 | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | Ports & Packages | Reporter: | Adam Jimerson <vendion> | ||||||
| Component: | Individual Port(s) | Assignee: | Dan Langille <dvl> | ||||||
| Status: | Closed FIXED | ||||||||
| Severity: | Affects Only Me | CC: | koobs, python, vendion | ||||||
| Priority: | --- | Flags: | bugzilla:
maintainer-feedback?
(dvl) |
||||||
| Version: | Latest | ||||||||
| Hardware: | Any | ||||||||
| OS: | Any | ||||||||
| Attachments: |
|
||||||||
I'm still new to writing/updating port makefiles, although this is something I would like to get into. If I missed something or did something wrong let me know so I can correct. Jim, The patch is a shar file, which is proper for a new port. For an update to an existing port, a diff is what should be supplied. See https://www.freebsd.org/doc/en/books/porters-handbook/porting-submitting.html Suggestion: submit a new patch as a diff. For the supplied patch: * I tested it, it works * the pkg-plist file is not required because Makefile contains: distutils There is a small discussion about that here: https://www.freebsd.org/doc/en_US.ISO8859-1/books/porters-handbook/using-python.html but note the formatting for Table 6.26 is mildly wonky. Good start, thank you. Thanks for the feedback Dan, I was unsure about the best way to the patch as my ports tree is from portsnap and thus not a subversion repo so the subversion commands didn't make sense which is why I did a shar file. I can re-roll my changes using a proper diff, if using the diff utility instead of subversions diff subcommand is acceptable. Or should I do this https://wiki.freebsd.org/PortsSubversionPrimer#Anonymous_Checkout to get a copy of the ports tree that is managed by subversion (this is not an issue for me if this is the best way)? Thanks for the link to the "Using Python" chapter I overlooked that section, I will drop the pkg-plist file from my changes then since it's not needed in this case. As I said in my original email I'm still new to this process, and was bound to make mistakes. (In reply to Adam Jimerson from comment #3) Assuming a clean ports tree, try this cp -a py-tldextract py-tldextract.orig Make your changes in py-tldextract Produce your patch via: diff -ruN py-tldextract.orig py-tldextract Created attachment 206791 [details]
New py-tldextract patch
Attached is the a new patch to update this port as a proper diff
(In reply to Adam Jimerson from comment #5) It applies cleanly. testport next. [dan@r710-01:~/WorkingCopyPortsTree] $ patch < ~/py-tldextract.diff Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |--- dns/py-tldextract/Makefile (revision 509606) |+++ dns/py-tldextract/Makefile (working copy) -------------------------- Patching file dns/py-tldextract/Makefile using Plan A... Hunk #1 succeeded at 2 with fuzz 1. Hmm... The next patch looks like a unified diff to me... The text leading up to this was: -------------------------- |--- dns/py-tldextract/distinfo (revision 509606) |+++ dns/py-tldextract/distinfo (working copy) -------------------------- Patching file dns/py-tldextract/distinfo using Plan A... Hunk #1 succeeded at 1. done Test port passes. A commit references this bug: Author: dvl Date: Thu Aug 22 17:02:56 UTC 2019 New revision: 509611 URL: https://svnweb.freebsd.org/changeset/ports/509611 Log: Update to 2.2.1 PR: 240030 Submitted by: Adam Jimerson <vendion@gmail.com> Changes: head/dns/py-tldextract/Makefile head/dns/py-tldextract/distinfo (In reply to Dan Langille from comment #6) Nice to know, I ended up just checking out a read-only copy of the ports repo and doing it there. Thanks again for your help here I would have been lost otherwise ;), is there anything else that I need to do here or is it all on your end now? Congratulations. Thank you. (In reply to Adam Jimerson from comment #9) If you'd like support in your porting, we have dedicated IRC channels: #freebsd-ports and #freebsd-python on freenode IRC, where we're happy to help ramp you up and answer questions @Dan Note also that this port is missing dependency declarations, from setup.py: INSTALL_REQUIRES = ["setuptools", "idna", "requests>=2.1.0", "requests-file>=1.4"] Full changelog (1.7.5 -> 2.2.1): https://github.com/john-kurkowski/tldextract/blob/master/CHANGELOG.md Since the package also installs console_scripts [1], the port should (must) use USE_PYTHON=concurrent. The missing dependencies change (commit) should also be merged to quarterly as that branch version will also be affected. It's a shame these issues weren't picked up prior to, or during the the 2.2.1 update as they could have been committed independently and merged to the quarterly branch without the 2.2.1 update. Personally, I'd merge the fixes and the 2.2.1 update to quarterly, as long as the current port version of its only dependency (dns/py-dns-lexicon) supports 2.2.1 too. [1] entry_points={ 'console_scripts': [ 'tldextract = tldextract.cli:main', ] }, (In reply to Kubilay Kocak from comment #12) If I was to submit a new patch to correct the missing Deps should I create a new PR or would re-opening this one and submitting a new patch here be the correct flow? (In reply to Kubilay Kocak from comment #11) Thanks for that tip, I will look into them. (In reply to Adam Jimerson from comment #13) If you create a new PR, I'd be happy to work on it with you. (In reply to Kubilay Kocak from comment #12) The issue with a MFH of the current version is the introduction of new features. With just 40 days or so until the new branch, this issue goes away, given people have been living / coping with it for at least more than the latest quarterly branch was created. I think a MFH should not be done because of the above. Happy to be told I'm wrong though. |
Created attachment 206775 [details] Updated tldextract files This contains an updated Makefile and distinfo file for tldextract 2.2.1 as well as adds the pkg-plist file which was missing.