예제 #1
0
 def validate(self):
     '''
     Do a syntax check on the format string. Doing a semantic check
     (verifying that the fields exist) is not useful in the presence of
     custom fields, because they may or may not exist.
     '''
     tmpl = preprocess_template(self.opt_template.text())
     try:
         t = validation_formatter.validate(tmpl)
         if t.find(validation_formatter._validation_string) < 0:
             return question_dialog(self, _('Constant template'),
                 _('The template contains no {fields}, so all '
                   'books will have the same name. Is this OK?'))
     except Exception as err:
         error_dialog(self, _('Invalid template'),
                 '<p>'+_('The template %s is invalid:')%tmpl +
                 '<br>'+str(err), show=True)
         return False
     return True
예제 #2
0
    def run(self, path_to_output, opts, db, notification=DummyReporter()):
        from calibre.utils.date import isoformat
        from calibre.utils.html2text import html2text
        from calibre.utils.bibtex import BibTeX
        from calibre.library.save_to_disk import preprocess_template
        from calibre.utils.date import now as nowf
        from calibre.utils.logging import default_log as log

        library_name = os.path.basename(db.library_path)

        def create_bibtex_entry(entry,
                                fields,
                                mode,
                                template_citation,
                                bibtexdict,
                                db,
                                citation_bibtex=True,
                                calibre_files=True):

            #Bibtex doesn't like UTF-8 but keep unicode until writing
            #Define starting chain or if book valid strict and not book return a Fail string

            bibtex_entry = []
            if mode != "misc" and check_entry_book_valid(entry):
                bibtex_entry.append(u'@book{')
            elif mode != "book":
                bibtex_entry.append(u'@misc{')
            else:
                #case strict book
                return ''

            if citation_bibtex:
                # Citation tag
                bibtex_entry.append(
                    make_bibtex_citation(entry, template_citation, bibtexdict))
                bibtex_entry = [u' '.join(bibtex_entry)]

            for field in fields:
                if field.startswith('#'):
                    item = db.get_field(entry['id'], field, index_is_id=True)
                    if isinstance(item, (bool, float, int)):
                        item = repr(item)
                elif field == 'title_sort':
                    item = entry['sort']
                elif field == 'library_name':
                    item = library_name
                else:
                    item = entry[field]

                #check if the field should be included (none or empty)
                if item is None:
                    continue
                try:
                    if len(item) == 0:
                        continue
                except TypeError:
                    pass

                if field == 'authors':
                    bibtex_entry.append(u'author = "%s"' %
                                        bibtexdict.bibtex_author_format(item))

                elif field == 'id':
                    bibtex_entry.append(u'calibreid = "%s"' % int(item))

                elif field == 'rating':
                    bibtex_entry.append(u'rating = "%s"' % int(item))

                elif field == 'size':
                    bibtex_entry.append(u'%s = "%s octets"' %
                                        (field, int(item)))

                elif field == 'tags':
                    #A list to flatten
                    bibtex_entry.append(
                        u'tags = "%s"' %
                        bibtexdict.utf8ToBibtex(u', '.join(item)))

                elif field == 'comments':
                    #\n removal
                    item = item.replace(u'\r\n', u' ')
                    item = item.replace(u'\n', u' ')
                    # unmatched brace removal (users should use \leftbrace or \rightbrace for single braces)
                    item = bibtexdict.stripUnmatchedSyntax(item, u'{', u'}')
                    #html to text
                    try:
                        item = html2text(item)
                    except:
                        log.warn("Failed to convert comments to text")
                    bibtex_entry.append(u'note = "%s"' %
                                        bibtexdict.utf8ToBibtex(item))

                elif field == 'isbn':
                    # Could be 9, 10 or 13 digits
                    bibtex_entry.append(u'isbn = "%s"' % format_isbn(item))

                elif field == 'formats':
                    #Add file path if format is selected
                    formats = [
                        format.rpartition('.')[2].lower() for format in item
                    ]
                    bibtex_entry.append(u'formats = "%s"' %
                                        u', '.join(formats))
                    if calibre_files:
                        files = [u':%s:%s' % (format, format.rpartition('.')[2].upper())\
                            for format in item]
                        bibtex_entry.append(u'file = "%s"' % u', '.join(files))

                elif field == 'series_index':
                    bibtex_entry.append(u'volume = "%s"' % int(item))

                elif field == 'timestamp':
                    bibtex_entry.append(u'timestamp = "%s"' %
                                        isoformat(item).partition('T')[0])

                elif field == 'pubdate':
                    bibtex_entry.append(u'year = "%s"' % item.year)
                    bibtex_entry.append(
                        u'month = "%s"' %
                        bibtexdict.utf8ToBibtex(strftime("%b", item)))

                elif field.startswith('#') and isinstance(item, basestring):
                    bibtex_entry.append(
                        u'custom_%s = "%s"' %
                        (field[1:], bibtexdict.utf8ToBibtex(item)))

                elif isinstance(item, basestring):
                    # elif field in ['title', 'publisher', 'cover', 'uuid', 'ondevice',
                    # 'author_sort', 'series', 'title_sort'] :
                    bibtex_entry.append(u'%s = "%s"' %
                                        (field, bibtexdict.utf8ToBibtex(item)))

            bibtex_entry = u',\n    '.join(bibtex_entry)
            bibtex_entry += u' }\n\n'

            return bibtex_entry

        def check_entry_book_valid(entry):
            #Check that the required fields are ok for a book entry
            for field in ['title', 'authors', 'publisher']:
                if entry[field] is None or len(entry[field]) == 0:
                    return False
            if entry['pubdate'] is None:
                return False
            else:
                return True

        def make_bibtex_citation(entry, template_citation, bibtexclass):

            #define a function to replace the template entry by its value
            def tpl_replace(objtplname):

                tpl_field = re.sub(u'[\{\}]', u'', objtplname.group())

                if tpl_field in TEMPLATE_ALLOWED_FIELDS:
                    if tpl_field in ['pubdate', 'timestamp']:
                        tpl_field = isoformat(
                            entry[tpl_field]).partition('T')[0]
                    elif tpl_field in ['tags', 'authors']:
                        tpl_field = entry[tpl_field][0]
                    elif tpl_field in ['id', 'series_index']:
                        tpl_field = str(entry[tpl_field])
                    else:
                        tpl_field = entry[tpl_field]
                    return tpl_field
                else:
                    return u''

            if len(template_citation) > 0:
                tpl_citation = bibtexclass.utf8ToBibtex(
                    bibtexclass.ValidateCitationKey(
                        re.sub(u'\{[^{}]*\}', tpl_replace, template_citation)))

                if len(tpl_citation) > 0:
                    return tpl_citation

            if len(entry["isbn"]) > 0:
                template_citation = u'%s' % re.sub(u'[\D]', u'', entry["isbn"])

            else:
                template_citation = u'%s' % str(entry["id"])

            return bibtexclass.ValidateCitationKey(template_citation)

        self.fmt = path_to_output.rpartition('.')[2]
        self.notification = notification

        # Combobox options
        bibfile_enc = ['utf8', 'cp1252', 'ascii']
        bibfile_enctag = ['strict', 'replace', 'ignore', 'backslashreplace']
        bib_entry = ['mixed', 'misc', 'book']

        # Needed beacause CLI return str vs int by widget
        try:
            bibfile_enc = bibfile_enc[opts.bibfile_enc]
            bibfile_enctag = bibfile_enctag[opts.bibfile_enctag]
            bib_entry = bib_entry[opts.bib_entry]
        except:
            if opts.bibfile_enc in bibfile_enc:
                bibfile_enc = opts.bibfile_enc
            else:
                log.warn("Incorrect --choose-encoding flag, revert to default")
                bibfile_enc = bibfile_enc[0]
            if opts.bibfile_enctag in bibfile_enctag:
                bibfile_enctag = opts.bibfile_enctag
            else:
                log.warn(
                    "Incorrect --choose-encoding-configuration flag, revert to default"
                )
                bibfile_enctag = bibfile_enctag[0]
            if opts.bib_entry in bib_entry:
                bib_entry = opts.bib_entry
            else:
                log.warn("Incorrect --entry-type flag, revert to default")
                bib_entry = bib_entry[0]

        if opts.verbose:
            opts_dict = vars(opts)
            log("%s(): Generating %s" % (self.name, self.fmt))
            if opts.connected_device['is_device_connected']:
                log(" connected_device: %s" % opts.connected_device['name'])
            if opts_dict['search_text']:
                log(" --search='%s'" % opts_dict['search_text'])

            if opts_dict['ids']:
                log(" Book count: %d" % len(opts_dict['ids']))
                if opts_dict['search_text']:
                    log(" (--search ignored when a subset of the database is specified)"
                        )

            if opts_dict['fields']:
                if opts_dict['fields'] == 'all':
                    log(" Fields: %s" % ', '.join(FIELDS[1:]))
                else:
                    log(" Fields: %s" % opts_dict['fields'])

            log(" Output file will be encoded in %s with %s flag" %
                (bibfile_enc, bibfile_enctag))

            log(" BibTeX entry type is %s with a citation like '%s' flag" %
                (bib_entry, opts_dict['bib_cit']))

        # If a list of ids are provided, don't use search_text
        if opts.ids:
            opts.search_text = None

        data = self.search_sort_db(db, opts)

        if not len(data):
            log.error(
                "\nNo matching database entries for search criteria '%s'" %
                opts.search_text)

        # Get the requested output fields as a list
        fields = self.get_output_fields(db, opts)

        if not len(data):
            log.error(
                "\nNo matching database entries for search criteria '%s'" %
                opts.search_text)

        #Initialize BibTeX class
        bibtexc = BibTeX()

        #Entries writing after Bibtex formating (or not)
        if bibfile_enc != 'ascii':
            bibtexc.ascii_bibtex = False
        else:
            bibtexc.ascii_bibtex = True

        #Check citation choice and go to default in case of bad CLI
        if isinstance(opts.impcit, (StringType, UnicodeType)):
            if opts.impcit == 'False':
                citation_bibtex = False
            elif opts.impcit == 'True':
                citation_bibtex = True
            else:
                log.warn("Incorrect --create-citation, revert to default")
                citation_bibtex = True
        else:
            citation_bibtex = opts.impcit

        #Check add file entry and go to default in case of bad CLI
        if isinstance(opts.addfiles, (StringType, UnicodeType)):
            if opts.addfiles == 'False':
                addfiles_bibtex = False
            elif opts.addfiles == 'True':
                addfiles_bibtex = True
            else:
                log.warn("Incorrect --add-files-path, revert to default")
                addfiles_bibtex = True
        else:
            addfiles_bibtex = opts.addfiles

        #Preprocess for error and light correction
        template_citation = preprocess_template(opts.bib_cit)

        #Open output and write entries
        with codecs.open(path_to_output, 'w', bibfile_enc, bibfile_enctag)\
            as outfile:
            #File header
            nb_entries = len(data)

            #check in book strict if all is ok else throw a warning into log
            if bib_entry == 'book':
                nb_books = len(filter(check_entry_book_valid, data))
                if nb_books < nb_entries:
                    log.warn("Only %d entries in %d are book compatible" %
                             (nb_books, nb_entries))
                    nb_entries = nb_books

            # If connected device, add 'On Device' values to data
            if opts.connected_device[
                    'is_device_connected'] and 'ondevice' in fields:
                for entry in data:
                    entry[
                        'ondevice'] = db.catalog_plugin_on_device_temp_mapping[
                            entry['id']]['ondevice']

            outfile.write(
                u'%%%Calibre catalog\n%%%{0} entries in catalog\n\n'.format(
                    nb_entries))
            outfile.write(
                u'@preamble{"This catalog of %d entries was generated by calibre on %s"}\n\n'
                % (nb_entries, nowf().strftime("%A, %d. %B %Y %H:%M").decode(
                    preferred_encoding)))

            for entry in data:
                outfile.write(
                    create_bibtex_entry(entry, fields, bib_entry,
                                        template_citation, bibtexc, db,
                                        citation_bibtex, addfiles_bibtex))
예제 #3
0
파일: bibtex.py 프로젝트: AEliu/calibre
    def run(self, path_to_output, opts, db, notification=DummyReporter()):
        from calibre.utils.date import isoformat
        from calibre.utils.html2text import html2text
        from calibre.utils.bibtex import BibTeX
        from calibre.library.save_to_disk import preprocess_template
        from calibre.utils.date import now as nowf
        from calibre.utils.logging import default_log as log

        library_name = os.path.basename(db.library_path)

        def create_bibtex_entry(entry, fields, mode, template_citation,
                                    bibtexdict, db, citation_bibtex=True, calibre_files=True):

            #Bibtex doesn't like UTF-8 but keep unicode until writing
            #Define starting chain or if book valid strict and not book return a Fail string

            bibtex_entry = []
            if mode != "misc" and check_entry_book_valid(entry) :
                bibtex_entry.append(u'@book{')
            elif mode != "book" :
                bibtex_entry.append(u'@misc{')
            else :
                #case strict book
                return ''

            if citation_bibtex :
                # Citation tag
                bibtex_entry.append(make_bibtex_citation(entry, template_citation,
                    bibtexdict))
                bibtex_entry = [u' '.join(bibtex_entry)]

            for field in fields:
                if field.startswith('#'):
                    item = db.get_field(entry['id'],field,index_is_id=True)
                    if isinstance(item, (bool, float, int)):
                        item = repr(item)
                elif field == 'title_sort':
                    item = entry['sort']
                elif field == 'library_name':
                    item = library_name
                else:
                    item = entry[field]

                #check if the field should be included (none or empty)
                if item is None:
                    continue
                try:
                    if len(item) == 0 :
                        continue
                except TypeError:
                    pass

                if field == 'authors' :
                    bibtex_entry.append(u'author = "%s"' % bibtexdict.bibtex_author_format(item))

                elif field == 'id' :
                    bibtex_entry.append(u'calibreid = "%s"' % int(item))

                elif field == 'rating' :
                    bibtex_entry.append(u'rating = "%s"' % int(item))

                elif field == 'size' :
                    bibtex_entry.append(u'%s = "%s octets"' % (field, int(item)))

                elif field == 'tags' :
                    #A list to flatten
                    bibtex_entry.append(u'tags = "%s"' % bibtexdict.utf8ToBibtex(u', '.join(item)))

                elif field == 'comments' :
                    #\n removal
                    item = item.replace(u'\r\n',u' ')
                    item = item.replace(u'\n',u' ')
                    # unmatched brace removal (users should use \leftbrace or \rightbrace for single braces)
                    item = bibtexdict.stripUnmatchedSyntax(item, u'{', u'}')
                    #html to text
                    try:
                        item = html2text(item)
                    except:
                        log.warn("Failed to convert comments to text")
                    bibtex_entry.append(u'note = "%s"' % bibtexdict.utf8ToBibtex(item))

                elif field == 'isbn' :
                    # Could be 9, 10 or 13 digits
                    bibtex_entry.append(u'isbn = "%s"' % format_isbn(item))

                elif field == 'formats' :
                    #Add file path if format is selected
                    formats = [format.rpartition('.')[2].lower() for format in item]
                    bibtex_entry.append(u'formats = "%s"' % u', '.join(formats))
                    if calibre_files:
                        files = [u':%s:%s' % (format, format.rpartition('.')[2].upper())\
                            for format in item]
                        bibtex_entry.append(u'file = "%s"' % u', '.join(files))

                elif field == 'series_index' :
                    bibtex_entry.append(u'volume = "%s"' % int(item))

                elif field == 'timestamp' :
                    bibtex_entry.append(u'timestamp = "%s"' % isoformat(item).partition('T')[0])

                elif field == 'pubdate' :
                    bibtex_entry.append(u'year = "%s"' % item.year)
                    bibtex_entry.append(u'month = "%s"' % bibtexdict.utf8ToBibtex(strftime("%b", item)))

                elif field.startswith('#') and isinstance(item, basestring):
                    bibtex_entry.append(u'custom_%s = "%s"' % (field[1:],
                        bibtexdict.utf8ToBibtex(item)))

                elif isinstance(item, basestring):
                    # elif field in ['title', 'publisher', 'cover', 'uuid', 'ondevice',
                        # 'author_sort', 'series', 'title_sort'] :
                    bibtex_entry.append(u'%s = "%s"' % (field, bibtexdict.utf8ToBibtex(item)))

            bibtex_entry = u',\n    '.join(bibtex_entry)
            bibtex_entry += u' }\n\n'

            return bibtex_entry

        def check_entry_book_valid(entry):
            #Check that the required fields are ok for a book entry
            for field in ['title', 'authors', 'publisher'] :
                if entry[field] is None or len(entry[field]) == 0 :
                    return False
            if entry['pubdate'] is None :
                return False
            else :
                return True

        def make_bibtex_citation(entry, template_citation, bibtexclass):

            #define a function to replace the template entry by its value
            def tpl_replace(objtplname) :

                tpl_field = re.sub(u'[\{\}]', u'', objtplname.group())

                if tpl_field in TEMPLATE_ALLOWED_FIELDS :
                    if tpl_field in ['pubdate', 'timestamp'] :
                        tpl_field = isoformat(entry[tpl_field]).partition('T')[0]
                    elif tpl_field in ['tags', 'authors'] :
                        tpl_field =entry[tpl_field][0]
                    elif tpl_field in ['id', 'series_index'] :
                        tpl_field = str(entry[tpl_field])
                    else :
                        tpl_field = entry[tpl_field]
                    return tpl_field
                else:
                    return u''

            if len(template_citation) >0 :
                tpl_citation = bibtexclass.utf8ToBibtex(
                    bibtexclass.ValidateCitationKey(re.sub(u'\{[^{}]*\}',
                        tpl_replace, template_citation)))

                if len(tpl_citation) >0 :
                    return tpl_citation

            if len(entry["isbn"]) > 0 :
                template_citation = u'%s' % re.sub(u'[\D]',u'', entry["isbn"])

            else :
                template_citation = u'%s' % str(entry["id"])

            return bibtexclass.ValidateCitationKey(template_citation)

        self.fmt = path_to_output.rpartition('.')[2]
        self.notification = notification

        # Combobox options
        bibfile_enc = ['utf8', 'cp1252', 'ascii']
        bibfile_enctag = ['strict', 'replace', 'ignore', 'backslashreplace']
        bib_entry = ['mixed', 'misc', 'book']

        # Needed beacause CLI return str vs int by widget
        try:
            bibfile_enc = bibfile_enc[opts.bibfile_enc]
            bibfile_enctag = bibfile_enctag[opts.bibfile_enctag]
            bib_entry = bib_entry[opts.bib_entry]
        except:
            if opts.bibfile_enc in bibfile_enc :
                bibfile_enc = opts.bibfile_enc
            else :
                log.warn("Incorrect --choose-encoding flag, revert to default")
                bibfile_enc = bibfile_enc[0]
            if opts.bibfile_enctag in bibfile_enctag :
                bibfile_enctag = opts.bibfile_enctag
            else :
                log.warn("Incorrect --choose-encoding-configuration flag, revert to default")
                bibfile_enctag = bibfile_enctag[0]
            if opts.bib_entry in bib_entry :
                bib_entry = opts.bib_entry
            else :
                log.warn("Incorrect --entry-type flag, revert to default")
                bib_entry = bib_entry[0]

        if opts.verbose:
            opts_dict = vars(opts)
            log("%s(): Generating %s" % (self.name,self.fmt))
            if opts.connected_device['is_device_connected']:
                log(" connected_device: %s" % opts.connected_device['name'])
            if opts_dict['search_text']:
                log(" --search='%s'" % opts_dict['search_text'])

            if opts_dict['ids']:
                log(" Book count: %d" % len(opts_dict['ids']))
                if opts_dict['search_text']:
                    log(" (--search ignored when a subset of the database is specified)")

            if opts_dict['fields']:
                if opts_dict['fields'] == 'all':
                    log(" Fields: %s" % ', '.join(FIELDS[1:]))
                else:
                    log(" Fields: %s" % opts_dict['fields'])

            log(" Output file will be encoded in %s with %s flag" % (bibfile_enc, bibfile_enctag))

            log(" BibTeX entry type is %s with a citation like '%s' flag" % (bib_entry, opts_dict['bib_cit']))

        # If a list of ids are provided, don't use search_text
        if opts.ids:
            opts.search_text = None

        data = self.search_sort_db(db, opts)

        if not len(data):
            log.error("\nNo matching database entries for search criteria '%s'" % opts.search_text)

        # Get the requested output fields as a list
        fields = self.get_output_fields(db, opts)

        if not len(data):
            log.error("\nNo matching database entries for search criteria '%s'" % opts.search_text)

        #Initialize BibTeX class
        bibtexc = BibTeX()

        #Entries writing after Bibtex formating (or not)
        if bibfile_enc != 'ascii' :
            bibtexc.ascii_bibtex = False
        else :
            bibtexc.ascii_bibtex = True

        #Check citation choice and go to default in case of bad CLI
        if isinstance(opts.impcit, (StringType, UnicodeType)) :
            if opts.impcit == 'False' :
                citation_bibtex= False
            elif opts.impcit == 'True' :
                citation_bibtex= True
            else :
                log.warn("Incorrect --create-citation, revert to default")
                citation_bibtex= True
        else :
            citation_bibtex= opts.impcit

        #Check add file entry and go to default in case of bad CLI
        if isinstance(opts.addfiles, (StringType, UnicodeType)) :
            if opts.addfiles == 'False' :
                addfiles_bibtex = False
            elif opts.addfiles == 'True' :
                addfiles_bibtex = True
            else :
                log.warn("Incorrect --add-files-path, revert to default")
                addfiles_bibtex= True
        else :
            addfiles_bibtex = opts.addfiles

        #Preprocess for error and light correction
        template_citation = preprocess_template(opts.bib_cit)

        #Open output and write entries
        with codecs.open(path_to_output, 'w', bibfile_enc, bibfile_enctag)\
            as outfile:
            #File header
            nb_entries = len(data)

            #check in book strict if all is ok else throw a warning into log
            if bib_entry == 'book' :
                nb_books = len(filter(check_entry_book_valid, data))
                if nb_books < nb_entries :
                    log.warn("Only %d entries in %d are book compatible" % (nb_books, nb_entries))
                    nb_entries = nb_books

            # If connected device, add 'On Device' values to data
            if opts.connected_device['is_device_connected'] and 'ondevice' in fields:
                for entry in data:
                    entry['ondevice'] = db.catalog_plugin_on_device_temp_mapping[entry['id']]['ondevice']

            outfile.write(u'%%%Calibre catalog\n%%%{0} entries in catalog\n\n'.format(nb_entries))
            outfile.write(u'@preamble{"This catalog of %d entries was generated by calibre on %s"}\n\n'
                % (nb_entries, nowf().strftime("%A, %d. %B %Y %H:%M").decode(preferred_encoding)))

            for entry in data:
                outfile.write(create_bibtex_entry(entry, fields, bib_entry, template_citation,
                    bibtexc, db, citation_bibtex, addfiles_bibtex))