view Tools/scripts/nm2def.py @ 73323:dad02a080bbc

Improve byte-compilation in packaging to be independent of -O or -B. The code I fixed to comply with PEP 3147 still had one bug: When run under python -O, some paths for pyc files would be pyo, because I called imp.cache_from_source without explicit debug_override argument in some places, and under -O that would return .pyo (this is well explained in the imp docs). Now all code (util.byte_compile, build_py, install_lib) can create .pyo files according to options given by users, without interference from the calling Python’s own optimize mode. On a related topic, I also removed the code that prevented byte compilation under python -B. The rationale is that packaging gives control over the creation of pyc files to the user with its own explicit option, and the behavior should not be changed if the calling Python happens to run with -B for whatever reason. I will argue that this is a bug fix and ask to be allowed to backport this change to distutils. Finally, I moved one nugget of information about the --compile and --optimize options from the source into the doc. It clears up a misunderstanding that I (and maybe other people) had.
author Éric Araujo <merwok@netwok.org>
date Thu, 03 Nov 2011 05:08:28 +0100
parents fcfe324663a6
children
line wrap: on
line source

#! /usr/bin/env python3
"""nm2def.py

Helpers to extract symbols from Unix libs and auto-generate
Windows definition files from them. Depends on nm(1). Tested
on Linux and Solaris only (-p option to nm is for Solaris only).

By Marc-Andre Lemburg, Aug 1998.

Additional notes: the output of nm is supposed to look like this:

acceler.o:
000001fd T PyGrammar_AddAccelerators
         U PyGrammar_FindDFA
00000237 T PyGrammar_RemoveAccelerators
         U _IO_stderr_
         U exit
         U fprintf
         U free
         U malloc
         U printf

grammar1.o:
00000000 T PyGrammar_FindDFA
00000034 T PyGrammar_LabelRepr
         U _PyParser_TokenNames
         U abort
         U printf
         U sprintf

...

Even if this isn't the default output of your nm, there is generally an
option to produce this format (since it is the original v7 Unix format).

"""
import os, sys

PYTHONLIB = 'libpython'+sys.version[:3]+'.a'
PC_PYTHONLIB = 'Python'+sys.version[0]+sys.version[2]+'.dll'
NM = 'nm -p -g %s'                      # For Linux, use "nm -g %s"

def symbols(lib=PYTHONLIB,types=('T','C','D')):

    lines = os.popen(NM % lib).readlines()
    lines = [s.strip() for s in lines]
    symbols = {}
    for line in lines:
        if len(line) == 0 or ':' in line:
            continue
        items = line.split()
        if len(items) != 3:
            continue
        address, type, name = items
        if type not in types:
            continue
        symbols[name] = address,type
    return symbols

def export_list(symbols):

    data = []
    code = []
    for name,(addr,type) in symbols.items():
        if type in ('C','D'):
            data.append('\t'+name)
        else:
            code.append('\t'+name)
    data.sort()
    data.append('')
    code.sort()
    return ' DATA\n'.join(data)+'\n'+'\n'.join(code)

# Definition file template
DEF_TEMPLATE = """\
EXPORTS
%s
"""

# Special symbols that have to be included even though they don't
# pass the filter
SPECIALS = (
    )

def filter_Python(symbols,specials=SPECIALS):

    for name in list(symbols.keys()):
        if name[:2] == 'Py' or name[:3] == '_Py':
            pass
        elif name not in specials:
            del symbols[name]

def main():

    s = symbols(PYTHONLIB)
    filter_Python(s)
    exports = export_list(s)
    f = sys.stdout # open('PC/python_nt.def','w')
    f.write(DEF_TEMPLATE % (exports))
    f.close()

if __name__ == '__main__':
    main()