Source code for bibolamazi.filters.util.auxfile

# -*- coding: utf-8 -*-
################################################################################
#                                                                              #
#   This file is part of the Bibolamazi Project.                               #
#   Copyright (C) 2013 by Philippe Faist                                       #
#   philippe.faist@bluewin.ch                                                  #
#                                                                              #
#   Bibolamazi is free software: you can redistribute it and/or modify         #
#   it under the terms of the GNU General Public License as published by       #
#   the Free Software Foundation, either version 3 of the License, or          #
#   (at your option) any later version.                                        #
#                                                                              #
#   Bibolamazi is distributed in the hope that it will be useful,              #
#   but WITHOUT ANY WARRANTY; without even the implied warranty of             #
#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the              #
#   GNU General Public License for more details.                               #
#                                                                              #
#   You should have received a copy of the GNU General Public License          #
#   along with Bibolamazi.  If not, see <http://www.gnu.org/licenses/>.        #
#                                                                              #
################################################################################

"""
Utilities (actually for now, utility) to parse .aux files from LaTeX documents.
"""


import os
import os.path
import re
import logging
logger = logging.getLogger(__name__)

from bibolamazi.core.bibfilter import BibFilterError #, BibFilter


rx_bibolamazibib_suffix = re.compile(r'(\.bibolamazi)?\.bib$', flags=re.IGNORECASE)

def get_action_jobname(jobname, bibolamazifile):
    """
    If `jobname` is non-None and nonempty, then return `jobname`. Otherwise,
    find the basename of the bibolamazifile, and return that.

    .. versionadded:: 4.3
        Added function :py:func:`get_action_jobname()`.
    """
    if jobname:
        return jobname
    # use bibolamazi file base name
    jobname = rx_bibolamazibib_suffix.sub('', os.path.basename(bibolamazifile.fname()))
                     
    logger.debug("Using jobname \"%s\" guessed from bibolamazi file name", jobname)
    return jobname



# BibTeX uses "\citation{}", BibLatex uses "\abx@aux@cite{}" in aux file... support both
rx_citation_aux_macro_pat = r'(?:\\citation|\\abx@aux@cite)'

def get_all_auxfile_citations(jobname, bibolamazifile, filtername, search_dirs=None,
                              callback=None, return_set=True):
    r"""
    Get a list of bibtex keys that a specific LaTeX document cites, by
    inspecting its .aux file.

    Look for the file ``<jobname>.aux`` in the current directory, or in the
    search directories `search_dirs` if given. Parse that file for commands of
    the type ``\citation{..}``, and collect all the arguments of such
    commands. These commands are generated by calls to the ``\cite{}`` command
    in the LaTeX document.

    Return a python set (unless `return_set=False`) with the list of all bibtex
    keys that the latex document cites.

    Note: latex/pdflatex must have run at least once on the document already.

    Arguments:

      - `jobname`: the base name of the TEX file; the AUX file that is searched
        for is "<jobname>.aux".  The `jobname` is expected to be non-empty; see
        :py:func:`get_action_jobname()` for help on that.

      - `bibolamazifile`: The bibolamazifile relative to which we are analyzing
        citations.  This is used to determine in which directory(ies) to search
        for the AUX file.

      - `filtername`: The name of the filter that is calling this function.
        Used for error messages and logs.

      - `search_dirs`: list of directories in which to search for the AUX file.
        These may be absolute paths or relative paths; the latter are
        interpreted as being relative to the bibolamazifile's location.

      - `callback`: A python callable can be specified in this argument.  It
        will be called for each occurrence of a citation in the document, with
        the citation key as single argument.

      - `return_set`: If True (the default), then this function returns a python
        `set` with all the citation keys encountered.  Set this to False if
        you're going to ignore the return value of this function.
    """

    logger.debug("%s: Retrieving citations from job name %r", filtername, jobname)

    citations_list = set()

    if (search_dirs is None):
        search_dirs = ['.', '_cleanlatexfiles']

    allaux = None
    for maybeauxfile in (os.path.join(bibolamazifile.fdir(), searchdir, jobname+'.aux')
                         for searchdir in search_dirs):
        try:
            with open(maybeauxfile, 'r') as auxf:
                logger.debug("%s: Reading auxfile %r", filtername, maybeauxfile)
                allaux = auxf.read()
        except IOError:
            pass

    if not allaux:
        raise BibFilterError(
            filtername,
            ("Can't analyze citations [filter {filtername}]: can't find file \"{jobname}.aux\". "
             "Please run latex on your document before running bibolamazi.").format(
                 filtername=filtername,
                 jobname=jobname
             )
        )

    #
    # parse allaux for \citation{...}
    #
    
    for citation in re.finditer(rx_citation_aux_macro_pat + r'\s*\{(?P<citekey>[^\}]+)\}', allaux):
        for citekey in (x.strip() for x in citation.group('citekey').split(',')):
            if return_set:
                citations_list.add(citekey)
            if callback is not None:
                callback(citekey)

    if return_set:
        return citations_list

    return