Haskell Code Explorer

You'll need to index a package before serving it.

A typical indexing workflow is as follows, take sqlite-simple as an example:

cabal unpack sqlite-simple
cd sqlite-simple-x.y.z.w
cabal configure --enable-testsuites --enable-benchmarks
cabal build all
haskell-code-indexer -p .

After this you can do

haskell-code-server -p /path/to/package1 -p /path/to/package2 ...

and the app will be available at localhost:3000.

For the readme of the original version, see README.md.original in the project tree, also available at https://g.ypei.me/haskell-code-explorer.git/tree/README.md.original.

The indexer of this version looks for all build targets, including testsuites and benchmarks, in the .cabal file, therefore you'll need to build them all before indexing, otherwise the indexer will complain about missing files.

The original haskell-code-explorer is written by Alexey Kiryushin (alexey.a.kiryushin@gmail.com), hosted at github.

The port to 9.2.2 is done by Yuchen Pei (id@ypei.org).

The original work is under the expat license, while the changes by Yuchen Pei are covered by GNU AGPLv3+ and you may find the license text in a file named COPYING.agpl3 in the project tree. As a combination, this work as a whole is licensed under GNU AGPLv3+ (why?).