Пример #1
0
 def __init__(self, headercoef=2.5):
     """
     Constructor.
     @param headercoef: lower border of elemet's visibility to be handled as header
     """
     self.generalizer = _RRSPropertyGeneralizer()
     self.ee = EntityExtractor()
     self.headercoef = headercoef
     self.bibtexparser = BibTeXParser()
     self.crawler = Crawler()
     self.mime_handler = MIMEhandler()
     self.crawler.set_handler(FileDownloader)
Пример #2
0
 def __init__(self, xmlcompatibility='db09'):
     self.seqwrapper = HTMLSequenceWrapper(childcoef=7.0,
                                           headercoef=3.0,
                                           mintextlen=30)
     self.citaextractor = CitationEntityExtractor(
         ALL, xmlcompatibility=xmlcompatibility)
     self.ee = EntityExtractor()
     self.mime = MIMEhandler()
     self.crawler = Crawler()
     self.bibtex = BibTeXParser()
     self.xmlcompatibility = xmlcompatibility
     self._xmlvalid = int(xmlcompatibility.lstrip('db'))
     self._publ_list = []
__modulename__ = "bibtexparser"
__author__ = "Matus Kontra"
__email__ = "*****@*****.**"
__date__ = "$24-Mai-2010 18:40:28$"

import re
from itertools import imap
from unicodedata import *
import datetime

from lxml import etree

from rrslib.db.model import *
from rrslib.extractors.entityextractor import EntityExtractor

_ee = EntityExtractor()


#Active State Code Recipes
#Recipe 66061: Assign and test
class DataHolder:
    def __init__(self, value=None):
        self.value = value

    def set(self, value):
        self.value = value
        return value

    def get(self):
        return self.value