From 1be510f9529cb082f802408b472a77d074b394c0 Mon Sep 17 00:00:00 2001 From: Nicolas Vigier Date: Sun, 14 Apr 2013 13:46:12 +0000 Subject: Add zarb MLs html archives --- zarb-ml/mageia-dev/2013-February/023184.html | 101 +++++++++++++++++++++++++++ 1 file changed, 101 insertions(+) create mode 100644 zarb-ml/mageia-dev/2013-February/023184.html (limited to 'zarb-ml/mageia-dev/2013-February/023184.html') diff --git a/zarb-ml/mageia-dev/2013-February/023184.html b/zarb-ml/mageia-dev/2013-February/023184.html new file mode 100644 index 000000000..3a238218f --- /dev/null +++ b/zarb-ml/mageia-dev/2013-February/023184.html @@ -0,0 +1,101 @@ + + + + [Mageia-dev] A question about BuildRequires and other RPM questions. + + + + + + + + + +

[Mageia-dev] A question about BuildRequires and other RPM questions.

+ AL13N + alien at rmail.be +
+ Thu Feb 28 16:59:10 CET 2013 +

+
+ +
Op donderdag 28 februari 2013 15:43:22 schreef Robert Wood:
+> On 28/02/13 14:25, Guillaume Rousse wrote:
+> > There is not much choice, beside building from a specified build
+> > environment, which is usually the minimal installation for your target
+> > distribution + rpmbuild. Most distributions use dedicated systems to
+> > automatically deploy such clean environment before eac package build
+> > attempt (iurt, mock, etc...).
+> 
+> So, I should maybe install a minimal install on a virtual machine? Then
+> when it needs packages I can add those to the dependencies list? It
+> strikes me the problem with that would be that once you've added a load
+> of packages for one RPM, then started on another, you'd be back to the
+> same problem as before and need to keep reinstalling Mageia for each RPM
+> you do. There must be a better way than that.
+
+i myself have made a script that uses a minimal install btrfs subvolume, makes 
+a snapshots for this package, then builds the rpm in a screen session, then 
+removes the snapshot again.
+
+i use this script mostly for testbuilding on my local machine.
+
+in the past i just submitted it to the buildsystem (to updates_testing) when i 
+was done working on it, and if it failed, i added some extra buildrequires, 
+until it built, after that i did a final test
+
+> > Build dependencies are usually specified in installation instructions.
+> > For humans, of course. You may also try to parse the outpout of
+> > ./configure (or equivalent) script. In both case, there is not garanty
+> > then every build dependency will get specified.
+> 
+> It sounds like there is a load of trial and error then? I am sure I must
+> misunderstand, building reliable RPMs must be a relatively
+> straightforward thing once you're up and running surely?
+
+trial and error for buildrequires yes, mostly for new packages, if you're 
+updateing packages, you sort of know more about the package and if it has 
+extra builddependencies, the BS will tell you this.
+
+so, in short for new packages you can have some trial and error (usually no 
+more than 5 or 6 tries, wrt buildrequires).
+
+it's not such a big deal...
+
+ + + + + + +
+

+ +
+More information about the Mageia-dev +mailing list
+ -- cgit v1.2.1