11 Jan 2025 |
mjm | idk, i think fork_point is more intuitive to me | 16:25:22 |
| oak changed their profile picture. | 16:46:02 |
mjm | merge base has always confused me as a term | 16:25:52 |
| oak changed their profile picture. | 16:46:59 |
| aprl changed their display name from aprl📍38c3 📞 2774 to aprl. | 17:54:19 |
Pouya Abbassi | Hi all! I'm trying to update my system and I get an error from one of the packages I'm using for a few months. The cnijfilter4 Nothing in the package itself or its source code has changed in a few years, but this time I'm getting this error:
> cnijnpr.c: In function 'get_printer_devid':
> cnijnpr.c:588:12: error: type of 'fdusb' defaults to 'int' [8;;https://gcc.gnu.org/onlinedocs/gcc-14.2.0/gcc/Warning-Options.html#index-Wimplicit-int-Wimplicit-int8;;]
> 588 | static int get_printer_devid(fdusb)
> | ^~~~~~~~~~~~~~~~~
It happens while running this line of the package. Is there a way to suppress or fix it?
| 21:06:32 |
K900 | GCC upgrade likely | 21:08:10 |
K900 | Ideally patch it to be (int fdusb) | 21:08:21 |
| pandapip1 joined the room. | 21:17:07 |
emily | check debian/fedora/arch for distro patches | 21:45:31 |
emily | oh this is some awful non-free printer blob | 21:45:38 |
emily | n/m do whatever you want :p | 21:45:41 |
Pouya Abbassi | K900: I'm getting new errors like
warning: ignoring return value of 'read' declared with attribute 'warn_unused_result'
Isn't there a way to suppress them all, or at least tell gcc to NOT treat warnings as errors for this package?
| 22:00:47 |
K900 | It should not do that by default | 22:01:01 |
K900 | So presumably it's hitting some warnings that are being treated as errors | 22:01:15 |
emily | you want to search for "error:" | 22:18:38 |
emily | have you considered using Internet Printing Protocol or whatever it is btw? | 22:19:19 |
emily | these ancient proprietary drivers are just going to bitrot more and more over time | 22:19:40 |
Pouya Abbassi |
have you considered using Internet Printing Protocol or whatever it is btw? printer does not have wifi and is connected directly to my laptop. I guess I'll just remove the driver and use my windows pc to print stuff :|
| 22:22:55 |
Pouya Abbassi | *
have you considered using Internet Printing Protocol or whatever it is btw?
printer does not have wifi and is connected directly to my laptop. I guess I'll just remove the driver and use my windows pc to print stuff :|
| 22:23:28 |
emily | you can just set -Wno-error | 22:35:57 |
emily | for the errors that appear | 22:35:58 |
emily | (again search error: in the compile log) | 22:36:03 |
emily | ideally send a PR | 22:36:13 |
Pouya Abbassi |
you can just set -Wno-error
how do I set this?
I fixed two issues. Now I'm getting the third one and I'm not a C programmer so they're a bit cryptic.
getipc.c:62:72: error: passing argument 3 of 'accept' from incompatible pointer type [-Wincompatible-pointer-types]
62 | while( (client_fd = accept(server_fd, (struct sockaddr *)&sun, &len)) >= 0 )
| ^~~~
| 22:39:24 |
emily | -Wno-error=incompatible-pointer-types | 22:39:42 |
emily | it's not really worth patching because it's an unfree source dump that will never be updated | 22:40:00 |
emily | just env.NIX_CFLAGS_COMPILE = "-Wno-error=a -Wno-error=b …"; until it compiles | 22:40:12 |
Pouya Abbassi | Thanks. I found a file in the project that has CFLAGS definition. I have added -Wno-error to it 🤞 I would add yet another patch to the patches that are availabele in the nixpkgs for this package. That should help others. If I can fix it at all. | 22:43:01 |
| kuruczgy joined the room. | 22:59:44 |