Bug 209807 - security/py-pyotp: Update to 2.1.1
Summary: security/py-pyotp: Update to 2.1.1
Status: Closed FIXED
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: Any Any
: --- Affects Only Me
Assignee: Dmitry Marakasov
URL:
Keywords: patch, patch-ready
Depends on:
Blocks:
 
Reported: 2016-05-28 12:12 UTC by VK
Modified: 2016-05-28 22:07 UTC (History)
0 users

See Also:

Attachments
Update py-pyotp to 2.1.1 (1.28 KB, patch)
2016-05-28 12:12 UTC, VK 		vlad-fbsd: maintainer-approval+
Details | Diff
Update py-pyotp to 2.1.1 (1.76 KB, patch)
2016-05-28 14:12 UTC, VK 		vlad-fbsd: maintainer-approval+
Details | Diff
Update py-pyotp to 2.1.1 (1.64 KB, patch)
2016-05-28 21:15 UTC, VK 		vlad-fbsd: maintainer-approval+
Details | Diff
Show Obsolete (2) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description VK freebsd_triage 2016-05-28 12:12:12 UTC 
Created attachment 170748 [details]
Update py-pyotp to 2.1.1

Update to latest released v2.1.1.

Portlint doesn't pass, complains about the new TIMESTAMP line added by `make makesum`. I'll check if that's a bug to be reported. Otherwise port de/installs fine in a clean jail.
Comment 1 Dmitry Marakasov freebsd_committer freebsd_triage 2016-05-28 12:55:11 UTC 
Why do you use commit hash while the repository has tags? Also, please add WWW: to pkg-descr
Comment 2 VK freebsd_triage 2016-05-28 14:12:53 UTC 
Created attachment 170750 [details]
Update py-pyotp to 2.1.1

New patch with WWW in pkg-descr.
Comment 3 VK freebsd_triage 2016-05-28 14:16:11 UTC 
(In reply to Dmitry Marakasov from comment #1)

Because tags are not really snapshots of code state in time, like a commit hash. I've seen ports installation fail on hash/size because new code would be committed under the same tag. Am I using this wrong in the Makefile? Still new at this... Thanks.
Comment 4 Dmitry Marakasov freebsd_committer freebsd_triage 2016-05-28 19:44:29 UTC 
> Because tags are not really snapshots of code state in time, like a commit hash.

They are.

> I've seen ports installation fail on hash/size because new code would be committed under the same tag.

You may be confusing tags and branches. Nothing may be `committed under tag', but tag may be slipped, which happens rarely and is basically the same as distfile reroll. Tags, however, is what specify releases, not commit hashes, and ports should use them. It also reduces maintenance cost.
Comment 5 VK freebsd_triage 2016-05-28 21:14:11 UTC 
My point, maybe poorly expressed, was that people tag commits as they wish (and sometimes retag newer commits under previous tags), whereas a commit hash can't change for that specific snapshot. A tag is tied to a commit, so I was just trying to be more precise.

But I'll take your advice, new patch coming up.
Comment 6 VK freebsd_triage 2016-05-28 21:15:56 UTC 
Created attachment 170768 [details]
Update py-pyotp to 2.1.1

Removed commit hash in favor of tagname. Removed account name and repository name, as they default to port name.

Passes portlint, port test in a clean jail.
Comment 7 commit-hook freebsd_committer freebsd_triage 2016-05-28 22:07:25 UTC 
A commit references this bug:

Author: amdmi3
Date: Sat May 28 22:07:11 UTC 2016
New revision: 416072
URL: https://svnweb.freebsd.org/changeset/ports/416072

Log:
  - Update to 2.1.1
  - Switch to using github tags
  - Add WWW
  - While here, add LICENSE_FILE

  PR:		209807
  Submitted by:	vlad-fbsd@acheronmedia.com (maintainer)

Changes:
  head/security/py-pyotp/Makefile
  head/security/py-pyotp/distinfo
  head/security/py-pyotp/pkg-descr