​:boosts_ok_gay:​
Can anyone help with this error while compiling nix from source? I'm trying to compile Nix, the package manager and language, inside Termux. Despite every needed library being there, it still gives me this error and i couldn't find out why.
EDIT: do NOT suggest me to use nix-on-droid. it is slow because it's bound by proot.

FAILED: [code=1] src/libutil/libnixutil.so.p/file-descriptor.cc.o 
c++ -Isrc/libutil/libnixutil.so.p -Isrc/libutil -I../src/libutil -Isrc/libutil/include -I../src/libutil/include -I../src/libutil/widecharwidth -Isrc/libutil/linux/include -I../src/libutil/linux/include -Isrc/libutil/unix -I../src/libutil/unix -Isrc/libutil/unix/include -I../src/libutil/unix/include -I/data/data/com.termux/files/usr/include -I/usr/local/include -fdiagnostics-color=always -D_LIBCPP_HARDENING_MODE=_LIBCPP_HARDENING_MODE_FAST -fpch-instantiate-templates -include-pch src/libutil/libnixutil.so.p/precompiled-headers.hh.pch -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -std=c++23 -Wdeprecated-copy -Werror=suggest-override -Werror=switch -Werror=switch-enum -Werror=undef -Werror=unused-result -Werror=sign-compare -Wignored-qualifiers -Wimplicit-fallthrough -Wno-deprecated-declarations -fpch-instantiate-templates -fPIC -DBOOST_CONTEXT_DYN_LINK=1 -DBOOST_COROUTINES_DYN_LINK=1 -DBOOST_IOSTREAMS_DYN_LINK=1 -DBOOST_ALL_NO_LIB -pthread -MD -MQ src/libutil/libnixutil.so.p/file-descriptor.cc.o -MF src/libutil/libnixutil.so.p/file-descriptor.cc.o.d -o src/libutil/libnixutil.so.p/file-descriptor.cc.o -c ../src/libutil/file-descriptor.cc
../src/libutil/file-descriptor.cc:116:11: error: no member named 'sync_file_range' in the global namespace
  116 |         ::sync_file_range(fd, 0, 0, SYNC_FILE_RANGE_WRITE);
      |           ^~~~~~~~~~~~~~~
../src/libutil/file-descriptor.cc:116:37: error: use of undeclared identifier 'SYNC_FILE_RANGE_WRITE'
  116 |         ::sync_file_range(fd, 0, 0, SYNC_FILE_RANGE_WRITE);
      |                                     ^~~~~~~~~~~~~~~~~~~~~

#InfCloud hasn't been updated for a decade, yikes. But apparently it's still the *only* available statically servable (i.e. no database / no PHP / whatever) #CalDAV #CardDAV web client out there. 😱

The code is also... interesting. There's so much hard-coding going on, but maybe that's how you do web dev 😅 I made a #nix package you can override with some useful options such as increasing the animation speed and allowing vCard 4.0 support.

https://gitlab.com/nobodyinperson/yannix/-/blob/28052b214673c2b7b225a47bac13feb1637ac925/packages/infcloud.nix

Setting Up Cuirass Locally via @federicoschonborn https://lobste.rs/s/4zmz17 #lisp #nix #testing
https://carlo.zancanaro.id.au/posts/setting-up-cuirass-locally.html

I think what irks me the most about the understaffed darwin (= macOS) maintainers situation in #nixpkgs is not only that it forces pure Linux users to look out for darwin quirks and thus invest a significant amount of time in darwin maintenance, but also that a significant amount of the business value of DetSys, Flox and similar #Nix-based developer tools companies comes from this work. As a 50% darwin user (and maintainer) I'm grateful but still feel remorse about the unfairness.

disko-zfs: Declaratively Managing ZFS Datasets https://lobste.rs/s/twefpn #nix #release
https://redalder.org/blog/disko-zfs-declaratively-managing-zfs-datasets.html

https://borretti.me/article/some-data-should-be-code

100% agree with this. I dislike static config files so, so much. Either they're too limited, or they end up awkwardly embedding ad-hoc bits of a programming language.

On a related note, one of my favorite things about #nix is that it lets me generate config files with a programming language.

I resist the urge to boast about #nix superiority when someone brags about creating a highly non-deterministic, fragile script for #ansible #docker or #podman .

Even when writing private code, I strive for reproducibility with Nix.

I oppose the unreliability of other solutions.

In the #talos and Nix communities, we aim to make it work natively everywhere.

Is that wrong?

It is disheartening to me that presenting an objectively superior solution can often be perceived as confrontational.

I noticed that the aarch64 git annex standalone build test suite (emulated on NixOS under x86_64) is *far* slower than the the nixpkgs-provided one:

> nix run nixpkgs#legacyPackages.aarch64-linux.git-annex test
# ... takes ~8min

> nix run gitlab:nobodyinperson/yannix#packages.aarch64-linux.git-annex-standalone test
# ... takes ~35min

#nix #gitAnnex #NixOS

oh wait nvm...
it's not the master branch's fault...
sigh...

​:boosts_ok_gay:​
can someone help me compile nix? i got to the point of running meson compile -C build -j 8 but it eventually gives this error:

../src/libutil/file-descriptor.cc:116:11: error: no member named 'sync_file_range' in the global namespace
  116 |         ::sync_file_range(fd, 0, 0, SYNC_FILE_RANGE_WRITE);
      |           ^~~~~~~~~~~~~~~
../src/libutil/file-descriptor.cc:116:37: error: use of undeclared identifier 'SYNC_FILE_RANGE_WRITE'
  116 |         ::sync_file_range(fd, 0, 0, SYNC_FILE_RANGE_WRITE);
      |                                     ^~~~~~~~~~~~~~~~~~~~~
2 errors generated.

Pleasew someone help we're loosing ourt flipping mind ​:neocat_cry:​

how and why do these two #nix expression result in the same output?

nix-repl> :b fetchFromGitHub { owner = "lucee"; repo = "lucee-dockerfiles"; rev = "0b34c46e8c1385d7c4014ee6c154476a9995189d"; sha256 = "sha256-tzR30emegBC27E4XImgQrOBTYOGacaXmdvCZrvjLhsg="; }

This derivation produced the following outputs:
  out -> /nix/store/v5cb6cg9vhlixiaw3frmwcym4p2nilv5-source

nix-repl> :b fetchFromGitHub { owner = "MasaCMS"; repo = "MasaCMS"; rev = "44815bfe904a6470d977de4b776bea0cf913a13c"; sha256 = "sha256-tzR30emegBC27E4XImgQrOBTYOGacaXmdvCZrvjLhsg="; }

This derivation produced the following outputs:
  out -> /nix/store/v5cb6cg9vhlixiaw3frmwcym4p2nilv5-source

@tris

Did I start actually doing that because my #nixos install had no clipboard manager? Yes. Yes, I did.

Did that took a fuck ton more time than to paste stuff, and was it annoying as hell? Yes it did, and yes it was.

Did I learn a shit ton about #nix and #linux in general, though? Absolutely, Yes again.

I’m now much more comfortable with the core utils than I ever was so far despite *kinda knowing* how it works for years.

Linux / nix also helps to teach you what’s ticking.

How I manage my Guix System configs via @federicoschonborn https://lobste.rs/s/ytqpkb #nix
https://www.terracrypt.net/posts/guix-config.html

Riffing on a new flake parts project, Le Grille. Similar to Shell Game, but it also tries to keep the treefmt config in sync. Adding a language should add not just a devshell, but a treefmt config using the same tool versions. Also takes care of some other boilerplate, like deadnix+statix+nixfmt and checking headers with REUSE.

https://codeberg.org/rossabaker/le-grille

#Nix #FlakeParts #Devshell #Treefmt

The option `programs.yazi.shellWrapperName` default has changed from `yy` to `y` to align with the recommendation from upstream.
#NixOS #Nix #HomeManager

That ended up being as easy as adding vulkan-loader to my #nix development shell (no macOS-specific code paths needed in my Rust code), thanks to the work done by Randy in https://github.com/reckenrode/nixpkgs/commits/kosmickrisp/ and of course thanks to the awesome work done by the #KosmicKrisp developers at #Mesa

okay pop quiz: if you specify a #nix #flake input like so:

{
  url = "https://example.com/whatever";
  flake = false;
}

what type does it have?

Modularize Your NixOS Modules Pt. 2: Dendritic host configuration

https://feddit.org/post/25817921

Following up on my post last week, I have now finished Part 2 of my write-up on dendritic #nix modules, which covers a very simple #NixOS host. Check it out at https://pc-hass.de/blog/dendritic-machines/! Since then I have also finished up Part 1 that I posted last week in an almost-complete state. Let me know what you think and if you're interested in the topics mentioned at the end!

Nixtamal 1.0.0 released

https://fed.brid.gy/r/https://nixtamal.toast.al/changelog/

It seems like #agenix won't set the password for the runtime just from a file, for the #selfhosted webapps which re run as a systemd service , it has happened to #freshrss trust home user ( myself) but it works fine for gitlab runner
Why?
#nix is it some race condition analogous async callback which don't go well for pure Haskell without alib