On Wed, Sep 10, 2014 at 4:03 AM, Felix Kauselmann wrote: > I'll try. YACReaders source code uses a bunch of code and header files from > p7zip to build it's internal logic (wrapper, whatever) to access 7z.so and > Rar29.so. 7z.so and Rar29.so are then loaded dynamically at runtime. As I > wrote before, there are no official library headers for these .so files, so > this approach is not as uncommon as one would think.
Interesting approach, it left me wondering why the normal command-line interface wasn't good enough though. You will note that Rar29.so is available in p7zip-rar, which is non-free. It might be a good idea to use unar (not unrar) instead of p7zip since it supports RAR files using DFSG-free code and also supports 7z and other files. In case you need metadata about files in an archive before extracting an archive, it has lsar -json, which obviously outputs machine-readable metadata about the files in JSON format. http://unarchiver.c3.cx/commandline https://packages.debian.org/sid/unar > From a packagers perspective it's a bit problematic. While the pragmatic > approach of repackaging the upstream tarball and adding the needed source > files works, I'd rather not do that. It's sort of a case of convenience > code, the code already exists in Debian and from a security point of view it > would be good if the version of that code would match the version of p7zip > present in Debian. I'm glad to hear you don't want to add embedded code copies, it isn't often that upstream developers have that opinion. -- bye, pabs https://wiki.debian.org/PaulWise -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: https://lists.debian.org/caktje6gcrrerokt4w0yd5w-v7xqv8ufxuyk1s3xstxttvah...@mail.gmail.com
