On 28/09/16 17:06, Joachim Durchholz wrote:
Am 28.09.2016 um 08:29 schrieb Tony Morris:
There is nothing of merit in Python libraries to be learned.
That's almost true, but not 100%. E.g. does Haskell have doctests? I.e. you can write example code in the API-level docs, and there is tooling that can extract them, run them, and report whether the examples still work.
I've been doing that for years, with the exception that doing so in Haskell is far superior than in Python for reasons too long to list. https://hackage.haskell.org/package/doctest
There's also the old motto of "nothing is completely useless, it can still serve as a bad example".
Please don't ruin Haskell to the point of Python.
The Python stdlib is a collection of things people needed. So if you want a list of batteries that Haskell might be missing, Python's stdlib is a good shopping list.
Where is the useful bit? I have only heard of it, never actually seen it.
You don't want to copy the library API structure, but that danger is negligible. Python is even more imperative than C++ or Java, it's dynamically typed, and with these concept differences, what's a good library design in Python that leverages all the things that Python is good at is almost automatically neither desirable nor even possible in Haskell. _______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.