Age | Commit message (Collapse) | Author |
|
|
|
Auth was broken without the xsrf thing disabled. Anyway, its best to have
well-defined live and test settings. I will re-enable xsrf when I figure out the
js snippet I need.
|
|
|
|
Using lib.pipe for readability. Moved analyze into the public functions so I can
debug things easier by calling bild.analyze from the nix repl.
|
|
There is a bug in nixpkgs where lib.strings.splitString overflows on long lines.
builtins.split performs better, but needs the extra filter for some reason.
https://github.com/NixOS/nixpkgs/issues/68951
|
|
Boy that was harder than I thought, but the code is pretty clean now. I still
need to fixup the error handling, and maybe move all the extra args into an app
monad, but I think otherwise I have it in a good place.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Every key is just a new line in the $USER.pub file. This is not automatically
reflected to gitolite, which uses a separate config, so I'll need to come up
with a way to replace gitolite someday.
|
|
|
|
|
|
|
|
|
|
|
|
Now bild knows how to determine between modules that require ghcjs and ghc. It
also knows what *not* to build, meaning it won't try to build non-buildable nix
targets, for example (unfortunately this is just hardcoded for now), but it also
won't build scm or py targets that I haven't implemented yet. It just silently
fails, which is fine, because it means I can do `bild **/*` and everything just
works.
Of course, if I want to build scm code then I will have to implement that, but
that's not a priority right now.
|
|
|
|
|
|
I had to refactor Biz/Bild/Rules.nix. I also had to checkin my patched
hoogle.nix file, but I also upstreamed the patch to nixpkgs-dev so it shouldn't
stick around for too long.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Still calibrating my use of hlint.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
I decided to finish the Devalloc frontend by just removing it. This way I can
have everything in one file, the code is much simpler, and I can scale up as
needed.
Anyway, this works fine with Turbolinks and that's pretty cool, so you get the
same basic feel as with any other SPA.
|
|
After hours of trying to get a page abstraction working in a Miso app (both
Devalloc and Hero) I had a revelation: Miso is an SPA - *single* page
application - framework, and so trying to put multiple pages into it is
fundamentally at odds with the rest of the architecture.
Of course this is also a problem with Elm's pattern of nesting models in order
to create a page abstraction. They can pull it off because they don't also try
to do isomorphic rendering.
In hindsight this should be obvious... if I actually do want some kind of
page-like abstraction or separation, then I need a much more complex server that
can embed different Miso apps based on some logic. But this is more like
multi-tenancy, or something.
Also I'm starting to think that I don't want Devalloc to be an SPA anyway, so
I'll try an experimental branch where I rip it out and just use Servant, Lucid,
and Turbolinks.
|
|
|
|
|