Geek Answers Handbook

What are the best practices for managing Cabal related packages?

I am working on a data stream optimization library written in Haskell. Now it seems likely that the library needs to be divided into two parts:

  • A kernel with minimal build dependencies; name him hoopl-core.

  • Full snippet, name it hoopl, which may have additional package dependencies such as prettyprinter, QuickCheck, etc.

The idea is that the Glasgow Haskell compiler will depend only on it hoopl-core, so it won't be too difficult to load the compiler. Other compilers will get extra goodies at hoopl. Package hooplwill depend on hoopl-core.

Debian package tools can create multiple packages from the same source tree. Unfortunately, Cabal has not yet reached that level of difficulty . But there should be other libraries or application developers who have similar problems (for example, one package for the main library, another for the command line interface, another for the GUI interface).

What are the current guidelines for creating and managing multiple Haskell related packages using Cabal?

+5
source share
2 answers

I would put the two packages in separate subdirectories and have Makefilewith something like this:

.PHONY: all hoopl hoop-core
all : hoopl

hoopl : hoopl-core
       cd hoopl && cabal build && cabal register --inplace

hoopl-core
       cd hoopl-core && cabal build && cabal register --inplace

, , hoopl-core (--inplace), hoopl. Makefile.

, GHC, ;-) . , GHC , ...

+3

.

, , .

+2

More articles:


All Articles