Confusion about Libtool archive (*.la) files in -dev' packages
Hi,
http://www.debian.org/Bugs/db/37/37338.html
I am currently working on editing in the policy amendments,
and I find this amendment quite confusing. Could the rpincipals
involved in this clarify exactly where the .la files are supposed to
go? Are they meant for the -dev packages, or the main ones (there was
a comment about needing the .la files at run time)? Could someone
also provide a rationale for this? I am not quite upto speed with
libtool (having never used it -- I program cross platform for non-GNU
systems ;-()), and I would prefer if I could have the exact wording
people want to be included in the policy manual.
manoj
This is what I was able to glean from the bug report itself:
<p>
An ever increasing number of packages are using libtool to
do their linking. The latest GNU libtools (>= 1.3a) can take
advantage of installed libtool archive files (`*.la'). The
main advantage of libtool's .la files is that it allows
libtool to store and subsequently access metadata with
respect to the libraries it builds. libtool will search for
those files, which contain a lot of useful information about
a library (e.g. dependency libraries for static
linking). Also, they're essential for programs using
libltdl.
</p>
<p>
Certainly libtool is fully capable of linking against shared
libraries which don't have .la files, but being a mere shell
script it can add considerably to the build time of a
libtool using package if that shellscript has to derive all
this infomation from first principles for each library every
time it is linked. With the advent of libtool-1.4 (and to a
lesser extent libtool-1.3), the .la files will also store
information about inter-library dependencies which cannot
necessarily be derived after the .la file is deleted.
</p>
<p>
Packages that use libtool to create shared libraries must
include the <em>.la</em> files in the <em>-dev</em>
packages. This is a good idea in general, and espescially
for static linking issues.
</p>
--
"You can measure a programmer's perspective by noting his attitude on
the continuing viability of Fortran." Alan Perlis
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/%7Esrivasta/>
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
Reply to: