collins cobuild 6th xml file formatter

gistfile1.py

# coding: utf8

import re

class Formatter(dict):
  def __init__(self, input_file):
    dict.__init__(self)
    xml_source = open(input_file, 'r')
    for line in xml_source.xreadlines():
      line = line.strip()
      if line.startswith('<?xml'): continue
      elif line.startswith('<dictionary'): continue
      elif line.startswith('</dictionary'): continue
      elif line.startswith('<headword'): # 开始词汇
        self.current_word = None
        self.current_str = ''
        self.num = 0
      elif line.startswith('</headword'): # 结束词汇
        # 格式化词汇 to implement
        self.current_str += '\n'
        #print '=' * 10, self.current_word, '=' * 10
        #print self.current_str
        self[self.current_word] = self.current_str
      elif line.startswith('<'):
        cmd = line[1:5]
        if cmd in [
          'FREQ',
          'BXTH',
          'BLOC',
          'BTSY',
          'XRCA',
          'XRCA',
          'BXWL',
          'BWWL',
          'BWEG',
          'BXWP',
          'BWCP',
          'BOXU',
          'BXUT',
          'PXWW',
          'PXPD',
          'CAT0',
          'BOXA',
          'BOXB',
          'XRC0',
        ]: continue

        if 'value=' in line:
          value = line[14:-4]
        else:
          value = ''
        formatter = getattr(self, 'format_' + cmd)
        self.current_str += formatter(value)

  @staticmethod
  def nojunk(s):
    s = s.replace('&lt;b&gt;', '')
    s = s.replace('&lt;/b&gt;', '')
    s = s.replace('&lt;x&gt;', '')
    s = s.replace('&lt;/x&gt;', '')
    s = s.replace('&lt;i&gt;', '')
    s = s.replace('&lt;/i&gt;', '')
    s = s.replace('&amp;', '&')
    return s

  @staticmethod
  def format_word(w):
    w = w.replace('+', '')
    w = w.replace('^', '')
    return w

  # 词汇
  def format_HWME(self, v):
    v = self.nojunk(v)
    v = self.format_word(v)
    self.current_word = v
    return v
  def format_HWSE(self, v):
    v = self.nojunk(v)
    v = self.format_word(v)
    self.current_word = v
    return v
  def format_HWAE(self, v):
    v = self.nojunk(v)
    v = self.format_word(v)
    self.current_word = v
    return v

  # also ...
  def format_HWAS(self, v):
    return '\nalso %s' % v
    
  # 发音
  def format_PRON(self, v):
    return '\n\n' + self.nojunk(v) + ' '

  # 变体
  def format_HDIF(self, v):
    return '(' + v + ') '
  def format_RNON(self, v):
    return '(' + v + ') '
  def format_HDIA(self, v):
    return '(' + v + ') '

  # 编号
  def format_DNUM(self, v):
    self.num += 1
    return '\n\n' + str(self.num) + ' '

  # 词性
  def format_POSP(self, v):
    return '[%s] ' % v

  # 解释
  def format_DEFN(self, v):
    return '\n' + self.nojunk(v)
  def format_DEFX(self, v):
    return '\n' + self.nojunk(v)
  def format_POSC(self, v):
    return '\n' + self.nojunk(v)

  # 例句
  def format_EGPH(self, v):
    return '\n=> ' + self.nojunk(v)

  # or ...
  def format_HWAF(self, v):
    return '\nor ' + v

  # 用法提示, in addition to ...
  def format_BOXX(self, v):
    return '\n-> ' + self.nojunk(v)

  # see ...
  def format_XRSE(self, v):
    return '\n-> see ' + v

  # BRIN or AM
  def format_LBRN(self, v):
    return ' (' + v + ')'

  # 用法提示
  def format_BOXR(self, v):
    return '\n-> ' + self.nojunk(v)

  # 跟在词性后面的语法用法
  def format_GRAM(self, v):
    return ' [' + self.nojunk(v) + ']'

  def format_HDGR(self, v):
    return ' [' + self.nojunk(v) + ']'

  # 情感
  def format_BOXE(self, v):
    return ' (' + v + ')'

  # see also ..
  def format_XRSA(self, v):
    return '\n-> see also ' + v
  def format_XRPX(self, v):
    return '\n-> see also ' + v

  # 反义词
  def format_BTAN(self, v):
    return '\n<> ' + v

  # 性质
  def format_LBRR(self, v):
    return ' (' + v + ')'
  def format_LBSF(self, v):
    return ' (' + v + ')'

  # 短语
  def format_PHVA(self, v):
    return '\n\n** ' + v + ' '
  def format_PHVB(self, v):
    return '\n\n** ' + v + ' '
  def format_XRLS(self, v):
    return '\n** ' + self.nojunk(v) + ' '

  # 提示
  def format_BOXP(self, v):
    return '\n-> ' + v
  def format_BOXL(self, v):
    return '\n-> ' + v
  def format_XRXR(self, v):
    return '\n-> ' + v

  # 消歧
  def format_MNXP(self, v):
    return '\n-> ' + v

  # trademark
  def format_LBTM(self, v):
    return ' (' + v + ')'

outfile = open('collins.src', 'w')
for c in range(ord('a'), ord('z') + 1):
  filename = 'xmls/%s.xml' % chr(c)
  print filename
  f = Formatter(filename)
  for word in f:
    d = f[word].replace('\n', r'\n')
    outfile.write('%s\t%s\n' % (word, d))

outfile.close()