[PYTHON DOC-SIG] Re: Inclusion of contributed modules
Andrew Kuchling
amk@magnet.com
Thu, 7 Nov 1996 12:48:53 -0500 (EST)
Guido van Rossum wrote (in comp.lang.python):
> A good start would probably to have a standard infrastructure to build
> third-party extensions. How about having them inside a subdirectory
> Extensions in the Python toplevel directory, if they are relatively
IMHO, all non-standard modules should live in such an
Extensions/ subdirectory; I've tried to do this with the sizable
extensions that I write. ILU's really a rare special case, since it's
much more than simply a Python module.
Another important point to consider is documentation for
extension modules; if possible, lib*.tex files should be automatically
included in the library reference. I hate the fact that many modules
come with READMEs
The following script looks through the directories in sys.path
for anything that can be imported, cross-references them with the
files in Doc/, and produces TeX code for a customized library
reference. The only thing needed is to have it walk through
subdirectories of Extensions/, find any lib*.tex files, and generate
appropriate TeX code. (Perhaps it shouldn't list the standard modules
at all; then adding a new module won't require reprinting the entire
Library Reference, but only a much smaller custom modules document.)
Andrew Kuchling
amk@magnet.com
# Generate custlib.tex, which is a site-specific library document.
# Phase I: list all the things that can be imported
import glob, os, sys, string
modules={}
for modname in sys.builtin_module_names:
modules[modname]=modname
for dir in sys.path:
# Look for *.py files
filelist=glob.glob(os.path.join(dir, '*.py'))
for file in filelist:
path, file = os.path.split(file)
base, ext=os.path.splitext(file)
modules[string.lower(base)]=base
# Look for shared library files
filelist=(glob.glob(os.path.join(dir, '*.so')) +
glob.glob(os.path.join(dir, '*.sl')) +
glob.glob(os.path.join(dir, '*.o')) )
for file in filelist:
path, file = os.path.split(file)
base, ext=os.path.splitext(file)
if base[-6:]=='module': base=base[:-6]
modules[string.lower(base)]=base
# Minor oddity: the types module is documented in libtypes2.tex
if modules.has_key('types'):
del modules['types'] ; modules['types2']=None
# Phase II: find all documentation files (lib*.tex)
# and eliminate modules that don't have one.
docs={}
filelist=glob.glob('lib*.tex')
for file in filelist:
modname=file[3:-4]
docs[modname]=modname
mlist=modules.keys()
mlist=filter(lambda x, docs=docs: docs.has_key(x), mlist)
mlist.sort()
mlist=map(lambda x, docs=docs: docs[x], mlist)
modules=mlist
# Phase III: write custlib.tex
# Write the boilerplate
# XXX should be fancied up.
print """\documentstyle[twoside,11pt,myformat]{report}
\\title{Python Library Reference}
\\input{boilerplate}
\\makeindex % tell \\index to actually write the .idx file
\\begin{document}
\\pagenumbering{roman}
\\maketitle
\\input{copyright}
\\begin{abstract}
\\noindent This is a customized version of the Python Library Reference.
\\end{abstract}
\\pagebreak
{\\parskip = 0mm \\tableofcontents}
\\pagebreak\\pagenumbering{arabic}"""
for modname in mlist:
print "\\input{lib%s}" % (modname,)
# Write the end
print """\\input{custlib.ind} % Index
\\end{document}"""
=================
DOC-SIG - SIG for the Python Documentation Project
send messages to: doc-sig@python.org
administrivia to: doc-sig-request@python.org
=================