On 2021-08-04 06:41, Jon Turney wrote:
On 03/08/2021 17:14, Brian Inglis wrote:
On 2021-08-03 09:56, Jon Turney wrote:
On 02/08/2021 18:19, Cygwin libidn2 Maintainer wrote:
The following packages have been upgraded in the Cygwin distribution:
* libidn 1.38
* libidn12 1.38
* libidn-devel 1.38
* libidn-doc 1.38
* mingw64-x86_64-libidn 1.38
* mingw64-i686-libidn 1.38
and the following package has been obsoleted from the Cygwin
distribution:
* libidn11 1.33
I've reverted that obsoletion, by removing 'obsoletes: libidn11' from
the hint for libidn12, since it apparently still has some uses.
... and removed the empty libidn11-1.1.38-1 package (generated by
cygport for compatibility with obsolete versions of setup)
... and added a 'replace-versions: 1.38-1' hint to libidn11 (in case
someone installed the above before I remembered to remove them)
Can users just rerun Cygwin Setup so that it will update setup.ini and
reinstall cygidn-11.dll?
Yes, that should fix any broken installs.
Is obsoleting previous dlls something that we should not do on a
package ABI break?
Correct, do not do that.
In this context, 'package A obsoletes package B' means 'package B
provides everything that package A did, so if A is installed, uninstall
A and install B'.
This behaviour is not unique to Cygwin packaging.
How should maintainers handle such situations in cygport?
You don't need to mention the old soversion in the updated cygport at all.
(a heuristic in calm identifies old soversions, and exempts them from
the (annoying) "all install packages from a source package must have a
unique current version" check)
(Yes, that means that those old soversions, and the corresponding
source, linger in the repository indefinitely. yselkowitz would
occasionally manually locate old soversions which aren't required by any
other package (or which could be made so with some rebuilds), and purge
them from the repo, but ... that service is no longer running :))
Thanks Jon for the fix.
*Sorry* to all users impacted to have caused the issue.
--
Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada
This email may be disturbing to some readers as it contains
too much technical detail. Reader discretion is advised.
[Data in binary units and prefixes, physical quantities in SI.]
--
Problem reports: https://cygwin.com/problems.html
FAQ: https://cygwin.com/faq/
Documentation: https://cygwin.com/docs.html
Unsubscribe info: https://cygwin.com/ml/#unsubscribe-simple
