Re: [PATCH] gnu: Add python-xlrd.

[Mathieu Lirzin <mthl@openmailbox.org>]

Ben Woodcroft <b.woodcroft@uq.edu.au> writes:

> From cb470509de810ea93ac4e576bb0347225b68ad98 Mon Sep 17 00:00:00 2001
> From: Ben Woodcroft <donttrustben@gmail.com>
> Date: Sat, 19 Sep 2015 23:35:18 +1000
> Subject: [PATCH] gnu: Add python-xlrd.
>
> * gnu/packages/python.scm (python-xlrd, python2-xlrd): New variables.
> ---
>  gnu/packages/python.scm | 37 +++++++++++++++++++++++++++++++++++++
>  1 file changed, 37 insertions(+)

[...]

> +    (arguments
> +     `(#:phases
> +       (modify-phases %standard-phases
> +         ;; current test in setup.py does not work as of 0.9.4,

“Current”

> +         ;; so use nose to run tests instead for now.
> +         (replace 'check (lambda _ (zero? (system* "nosetests")))))))
> +    (native-inputs
> +     `(("python-setuptools" ,python2-setuptools)
> +       ("python-nose" ,python2-nose)))

I don't know python packaging but when I look at other packages it
seems that these are usually used as “inputs” not
“native-inputs”.

I think it should be “python-*” instead of “python-2”.  Am I wrong?

> +    (home-page "http://www.python-excel.org/")
> +    (synopsis
> +     "Library for extracting data from Microsoft Excel (tm) files")
> +    (description
> +     "Extract data from Excel spreadsheets (.xls and .xlsx, versions 2.0
> +onwards) on any platform.  It is pure Python (2.6, 2.7, 3.2+), has support for
> +Excel dates and is Unicode-aware.")

What about something like this?

        (synopsis "Library for extracting data from Excel files")
        (description "This packages provides a library to extract data from
    spreadsheets using Microsoft Excel® proprietary file formats @samp{.xls} and
    @samp{.xlsx} (versions 2.0 onwards).  It has support for Excel dates and is
    Unicode-aware.  It is not intended as an end-user tool.")

> +    (license bsd-3)))
> +(define-public python2-xlrd
> +  (package-with-python2 python-xlrd))

--
Mathieu Lirzin