On Fri 26 Nov 2021 at 09:40:59 +0100, Dr. Thomas Orgis wrote: > On the upside: If we could get the clean bootstrap process for pip > running, it would be nice if we could employ its help to actually > define the pkgsrc packages. After all, it's mostly a translation of the > metadata the Python world already organizes. I've lightly been thinking about such a process: generating the pkgsrc entry data automatically from pip/pypi/whatever/some other language's packaging system. And then use the normal pkgsrc way of building our own binary pkg from that. We could even invent some way to collect such auto-generated source packages somewhere (for sharing, or cross-checking, or rebuilding, or whatever). And we could even package all the files from a source package (i.e. Makefile, PLIST, DESCR, patches/*, etc) into a binary pkg. That would make a binary package re-buildable. We might even do that for all packages, not only auto-generated ones. It's usually just a few short text files, after all, that describe the whole package. (Somehow that last idea feels a bit similar to Debian packages, yet is still different; Debian mixes its packaging instructions / metadata into source trees, not resulting binary packages. This makes Debian package version management very confusing, if it's done by somebody else than the developer of the sources). Just some wild ideas. > Mabe even having a mechanism to snapshot the state of PiPI matching a > pkgsrc release [...] and then having a way to install Python packages > outside of the curated space of pkgsrc [...] via pip and ad-hoc > package creation from the PyPI metadata in a reproducible manner. So > installing packages in a controlled way, managed by pkgsrc, but > without having them all actually manually packaged in pkgsrc CVS. -Olaf. -- ___ "Buying carbon credits is a bit like a serial killer paying someone else to \X/ have kids to make his activity cost neutral." -The BOFH falu.nl@rhialto
Attachment:
signature.asc
Description: PGP signature