On 2020-12-06 06:16:21 Carlos E. R. wrote:
On Sunday, 2020-12-06 at 00:44 -0600, J Leslie Turriff wrote:
On 2020-12-04 05:38:57 Carlos E. R. wrote:
On 03/12/2020 22.32, J Leslie Turriff wrote:
On 2020-11-29 14:22:01 Carlos E. R. wrote:
On 29/11/2020 21.11, Dave Howorth wrote:
On Sun, 29 Nov 2020 20:07:03 +0100 "Carlos E. R." <> wrote: > On 29/11/2020 17.05, Axel Braun wrote: >> Am Sonntag, 29. November 2020, 14:38:33 CET schrieb Carlos E. R.: >>> On 29/11/2020 14.21, Axel Braun wrote:
...
The point of OBS is to provide a well-defined environment so that compilations etc *ARE* reproducible.
Ah, understood, thanks :-)
But that has never blocked me from compiling something I wanted. Can't say I even noticed the issue.
I suspect that it would be the users of the RPM that you create who might see problems from that. :-)
There are no users save me alone.
I mean, if you were to publish your RPM for others to use.
But I don't, ever.
The goal of using "checkinstall" is not to publish rpms, but to run "make install" and produce an rpm in/for my machine, and install that one instead, so that the rpm database knows that the package is installed, and so doesn't insist in installing the published package if it exists. And also facilitates uninstalling the produced package easily, because many devs create the "make install" procedure and forget about "make uninstall".
The instant one wants to publish the rpm, the road changes and OBS becomes the correct way.
In fact, I have something installed in my machine using "make install" because I could not find checkinstall. I forget which thing,
Hm. I would never have guessed that "checkinstall" is a tool for converting a tarball to an RPM. This is the sort of thing that drives me crazy: "I've got a great idea for a helpful tool! What should I call it? 'tar2rpm'? No, too obvious..." Leslie --