On Tue, Sep 30, 2014 at 3:13 AM, Sander van der Burg <[email protected]> wrote:
> I've used the reengineered npm2nix on a private project for the company I > work for, as well as some other utilities and it seems to work fine for me. > > Hopefully, you can also try npm2nix on your projects to find out whether > there any additional issues. :) > > Furthermore, I'm still working a blog post to rationalize all the stuff, > which should give you better insights in the problem, details and the > choices I made. > > Let me know what you think! > Hi Sander, I've just tried out this version and I'm finding it pleasant improvement over the old way. The generated code is clean and simple, and I like the minimal dependency it has on nixpkgs. However, there's one thing I haven't been able to get working: local path dependencies. I have several node apps that rely on a collection of node modules. These modules may, in turn rely on other modules. In addition, both the apps and the modules depend on modules from npm. My current setup involves generating nix expressions using npm2nix (Sheay's version) to get the NPM dependencies, then hand-editing the default.nix files for each app and module, to pull in the additional dependencies that don't come from npm. As of NPM 2.0, dependencies can be specified as paths on the local file system, and I'd like to use that feature to generate nix expressions for both local and npm dependencies using npm2nix. It almost works, but not quite. AFAICT, the nix expressions are generated just fine, but then when the derivation is being created, the relative paths are resolved against the current directory of the of the process, rather than against the directory that contains the package.json file. That causes the derivation to fail with ENOENT. All in all, looks like big improvement! Colin
_______________________________________________ nix-dev mailing list [email protected] http://lists.science.uu.nl/mailman/listinfo/nix-dev
