lib/path-concat.c redefines mempcpy before including <string.h>, causing a parse error in the header file (GCC's context reporting isn't great here): path-concat.c:25:29: error: expected identifier or '(' before 'void' 25 | # define mempcpy(D, S, N) ((void *) ((char *) memcpy (D, S, N) + (N))) | ^~~~ path-concat.c:25:37: error: expected ')' before '(' token 25 | # define mempcpy(D, S, N) ((void *) ((char *) memcpy (D, S, N) + (N))) | ^ Overall, the package is in a very bad state. There are also many implicit function declarations during the build, like this example later in the same file: path-concat.c: In function 'path_concat': path-concat.c:81:23: warning: implicit declaration of function 'malloc' [-Wimplicit-function-declaration] 81 | p_concat = (char *) malloc (dir_len + base_len + 2); | ^~~~~~ These will be hard build failures with future compilers, and also tend to cause crashes on 64-bit architectures (if pointers are truncated to int). The gperf configure version test does not work because it cannot cope with multi-line output from gperf --version. I'm not sure if it is worth fixing this package because it still does not support UTF-8. Reproducible: Always
Hi Florian, thank you for reporting the issue! There is a new a2ps version with new upstream https://bugzilla.redhat.com/show_bug.cgi?id=2176254 , I'll try to update to the newest version and we will see how many issue will be fixed.
Any progress on the update?
I'm in the process of rebasing the package to the latest version - https://bugzilla.redhat.com/show_bug.cgi?id=2176254 - it has many downstream patches, so in case you are in hurry, please send a PR with rebase and I'll review it.
Fixed in rawhide https://koji.fedoraproject.org/koji/taskinfo?taskID=104013671