Curious about library inclusion

[Paul Moore]

Probably low. There would need to be a clear justification as to why
having the library in the stdlib (and hence tied to Python's release
schedule for updates/bug fixes etc - which is typically a really
severe limitation for a newish library) would be better than having it
available as a 3rd party library on PyPI. It would also mean that it
would only be available in Python 3.9+, unless a parallel backport
module was maintained on PyPI, making the argument for inclusion in
the stdlib even weaker.

Basically, someone needs to explain the benefits of having *this*
library in the stdlib, and demonstrate that they are compelling.
(Generic arguments like "being in the stdlib means no need for a 3rd
party dependency" don't count because they would apply to any library
on PyPI equally...) That's a pretty high bar to reach. Not impossible,
by any means, but it needs a lot more than "this is a neat library".

As another measure, look at various other libraries on PyPI and ask
yourself why *this* library needs to be in the stdlib more than those
others. The answer to that question would be a good start for an
argument to include the library.

Paul

On Thu, 10 Oct 2019 at 11:37, Antoon Pardon <antoon.pardon at vub.be> wrote:
>
> That seems to have been thoruoghly garbled so I start again.
>
> I was wondering how likely it would be that piped iterators like shown
> in http://code.activestate.com/recipes/580625-collection-pipeline-in-python/
> would make it into a future python version/ Once I started using them
> (and included some more) I found I really liked using them. For instance
> I used to write the following a lot: for line in some_file: line =
> line.strip() lst = line.split(':') do_stuff(lst) Now I seem to drift
> into writing: for lst in some_file | Strip | Split(':'): do_stuff(lst)
> where Strip and Split are defined as follows: Strip =
> Map(methodcaller('strip')) def Split(st): return
> Map(methodcaller('split', st)) I also found that Apply can be used as a
> decorator, that easily turns a generator into a piped version. So what
> are the odds? -- Antoon.
>
> --
> https://mail.python.org/mailman/listinfo/python-list