# -*- coding: iso-8859-1 -*- """MoinMoin - Bibtex Parser Configuration Options: bibtex_bib2html_cmd -- path and command name of bib2xhtml, needs to be set if bib2xhtml is not in PATH bibtex_bst_path -- path to bib2xhtml bst-style files, needs to be set if these are not found by bibtex automatically Notes: This parser depends on bibtex and bib2xhtml, see http://en.wikipedia.org/wiki/Bibtex and http://www.spinellis.gr/sw/textproc/bib2xhtml/ for more information. TODO: - attachement links $Id: bibtex.py,v 1.1.1.1 2007-06-07 20:36:15 gber Exp $ Copyright 2007 by Guido Berhoerster Licensed under the GNU GPL, see COPYING for details. $Log: bibtex.py,v $ Revision 1.1.1.1 2007-06-07 20:36:15 gber initial import into CVS """ import os import tempfile import subprocess import codecs import re from MoinMoin import wikiutil Dependencies = [] # subprocess.check_call and CalledProcessError are only available in # Python 2.5, so recreate them here class CalledProcessError(Exception): """ This exception is raised when a process run by check_call() returns a non-zero exit status. The exit status will be stored in the returncode attribute. """ def __init__(self, returncode, cmd): self.returncode = returncode self.cmd = cmd def __str__(self): return "Command '%s' returned non-zero exit status %d" \ % (self.cmd, self.returncode) class BibtexRenderer: """Abstracted bibtex renderer Arguments: bibtex -- string containing bibtex markup to be rendered request -- request object Keyword arguments: citations -- list with keys of bibtex entries to be rendered (default: empty) abstracts -- boolean, show abstracts or not (default: False) chronological -- boolean or the string "reversed", chronological or reversed chronological sorting (default: False) style -- string "empty", "plain", "alpha", "named", "unsort", "unsortlist", determining the style to use (default: None) Notes: Raises OSError if bib2xhtml is not found or CalledProcessError if bib2xhtml returns wit a non-zero exit status. """ def __init__(self, bibtex, request, citations=[], abstracts=False,\ label=False, chronological=False, style=None): cfg = request.cfg self.bib2html_cmd = "bib2xhtml" self.bst_path = None # retrieve configuration try: self.bib2html_cmd = cfg.bibtex_bib2html except AttributeError: pass try: self.bst_path = cfg.bibtex_bst_path except AttributeError: pass # the original bibtex implementation is not 8-bit clean, replace # non-ASCII characters with "?" self.bibtex = bibtex.encode("ascii", "replace") self.args = [self.bib2html_cmd, "-u", "-dMoinMoin"] if citations: cit_list = [] cit_list.append(u"") cit_list.append(u"") cit_list.append(u"") self.citations = u"\n".join(cit_list) # also encode as ASCII self.citations = self.citations.encode("ascii", "replace") self.args.append("-i") else: self.citations = None if abstracts: self.args.append("-a") if label: self.args.append("-k") if chronological and chronological == "reversed": self.args.extend(["-c", "-r"]) elif chronological: self.args.append("-c") if style in ("empty", "plain", "alpha", "named", "unsort", "unsortlist"): self.args.extend(["-s", style]) def render(self): """Render the bibtex markup (if requested, only cited entries) and return HTML output in a string. """ output = [] # create temporary files for Bibtex input, HTML output, and logging bibfd, bibfile = tempfile.mkstemp(".bib") xhtmlfd, xhtmlfile = tempfile.mkstemp(".xhtml") #logfd, logfile = tempfile.mkstemp(".log") self.args.extend([bibfile, xhtmlfile]) # write Bibtex input to temporary file f = open(bibfile, "w") f.write(self.bibtex) f.close() if self.citations: # write citations to temporary output file f = open(xhtmlfile, "w") f.write(self.citations) f.close() # execute bib2xhtml converter subprocess on forementionened # temporary files, bib2xhtml creates its temporary files in the # current working directory, so set it to a reasonable location, set # the set the BSTINPUTS environment variable if required in order to # help Bibtex finds the needed .bst files if self.bst_path: bstinputs = {"BSTINPUTS": self.bst_path} else: bstinputs = None try: retcode = subprocess.call(self.args, env=bstinputs, cwd=tempfile.gettempdir(), stdout=open(os.devnull, "w"), stderr=open(os.devnull, "w")) #retcode = subprocess.call(self.args, # env=bstinputs, cwd=tempfile.gettempdir(), # stdout=open(os.devnull, "w"), stderr=open(logfile, "w")) if retcode: raise CalledProcessError(retcode, "".join(self.args)) except OSError, error: # bib2xhtml not found or not executable os.remove(bibfile) os.remove(xhtmlfile) raise except CalledProcessError, error: # non-zero exit status os.remove(bibfile) os.remove(xhtmlfile) #os.remove(logfile) raise os.remove(bibfile) #os.remove(logfile) name_pattern = re.compile('', re.UNICODE) href_pattern = re.compile('', re.UNICODE) inside_dl = False # read the output (encoded as utf-8) back in f = codecs.open(xhtmlfile, "r", encoding="utf-8") for line in f.readlines(): if line.startswith(u'\n': # ...and the end break if inside_dl: # use a ref:-prefix for anchor links in order to avoid # interference with other anchors and replace the name- with # the id-attribute (it would be more appropriate to fix this in # the bst-file) line = name_pattern.sub(ur'', line) line = href_pattern.sub(ur'', line) output.append(line) f.close() os.remove(xhtmlfile) output = "".join(output) return output class Parser: """Parse Bibtex markup.""" extensions = ['.bib'] Dependencies = [] def __init__(self, raw, request, **kw): self.request = request self.show = True self.abstracts = False self.label = True self.chronological = False self.style = "named" self.raw = raw self.request = request self._ = request.getText # parse format arguments attrs, msg = wikiutil.parseAttributes(self.request, kw.get( 'format_args','')) if not msg: if attrs.get('show','"on"')[1:-1].lower() not in ('on', 'true', 'yes'): self.show = False if attrs.get('abstracts','"off"')[1:-1].lower() in ('on', 'true', 'yes'): self.abstracts = True if attrs.get('label','"on"')[1:-1].lower() not in ('on', 'true', 'yes'): self.label = False if attrs.get('chronological','"off"')[1:-1].lower() in ('on', 'true', 'yes'): self.chronological = True elif attrs.get('chronological','"off"')[1:-1].lower() in ( 'reverse', 'reversed'): self.chronological = "reversed" if attrs.get('style','"empty"')[1:-1].lower() in ('empty', 'plain', 'alpha', 'named', 'unsort', 'unsortlist'): self.style = attrs.get('style','')[1:-1].lower() def format(self, formatter): """Format Bibtex markup as HTML and write output.""" if not self.show: return _ = self._ bib_renderer = BibtexRenderer(self.raw, self.request, abstracts=self.abstracts, label=self.label, chronological=self.chronological, style=self.style) try: output = bib_renderer.render() except OSError, error: self.request.write(formatter.sysmsg(1)) self.request.write(formatter.escapedText( _("Error: %(bib2html_cmd)s could not be found.") % {"bib2html_cmd": self.bib2html_cmd})) self.request.write(formatter.sysmsg(0)) return except CalledProcessError, error: msg = (_("An error occured while executing %(bib2html_cmd)s:" % {"bib2html_cmd": self.bib2html_cmd})) self.request.write(formatter.sysmsg(1)) self.request.write(formatter.escapedText(msg)) self.request.write(formatter.sysmsg(0)) return # write through the rawHTML formatter if possible, otherwise write # just the Bibtex markup as preformatted text self.request.write(formatter.div(1, css_class="bibliography")) self.request.write(formatter.heading(1, 1)) self.request.write(formatter.escapedText(_("Bibliography"))) self.request.write(formatter.heading(0, 1)) try: self.request.write(formatter.rawHTML(output)) except: self.request.write(formatter.preformatted(1)) self.request.write(formatter.text(self.raw.expandtabs())) self.request.write(formatter.preformatted(0))