4
#===============================================================================
6
#===============================================================================
9
#4.0.264 - version update
17
logger = logging.getLogger('.'.join([os.path.splitext(os.path.basename(sys.argv[0]))[0],'manager','filtergraph',__name__]))
35
def open(self,parent,cfg):
36
self.encoding = 'utf8' if 'utf8' in cfg['encoding'].lower().replace('-','') else cfg['encoding']
37
self.inputfile = cfg['inputfile'].replace('%(rootfolder)s',self.p.rootfolder) if cfg['inputfile'] else self.inputfile
38
self.outputfile = cfg['outputfile'].replace('%(rootfolder)s',self.p.rootfolder) if cfg['outputfile'] else self.outputfile
39
if (self.inputfile and not self.outputfile) or (not self.inputfile and self.outputfile):
40
self.errors.append([__name__,'invalid','[%s] inputfile=%s without outputfile= value'%(__name__,cfg['inputfile'])])
41
logger.warn('%s\t%s',*self.errors[-1][1:])
42
self.regex = re._compile("<.+?\s*?.*?>", re.IGNORECASE | re.UNICODE)
46
skipclose = not self.inputfile
47
if self.inputfile: return
49
self.p.cfoutput[k]['tempbuff'] = [self.regex.sub('', line) for line in self.p.cfoutput[k]['tempbuff']]
59
if not os.path.exists(self.inputfile):
60
self.errors.append([__name__,'missing','[%s] %s'%(__name__,self.inputfile)])
61
logger.error('%s\t%s',*self.errors[-1][1:])
66
os.makedirs(os.path.dirname(self.outputfile))
69
logger.exception('%s\t%s, %s, %s',*['failed',e.errno,e.strerror,e.filename,])
72
# open input and output files
74
if out == self.inputfile:
76
fd,out = tempfile.mkstemp(suffix='.tmp', prefix='~', dir=self.p.tempdir)
79
o = codecs.open(out,'w',self.encoding)
80
i = codecs.open(self.inputfile,'r',self.encoding)
82
raise RuntimeError('Failed to open [%s] input/output files'%(__name__))
84
sys.stderr.write('[%s] %s\n Please wait'%(__name__,self.outputfile))
87
# loop writes output line-by-line
89
o.write('%s\n'%(self.regex.sub('', line.rstrip('\r\n'))))
91
if not cnt%5000: sys.stderr.write('.')
92
sys.stderr.write('\n')
93
# close input and output files
97
if not out == self.outputfile:
99
shutil.move(out,self.outputfile)
101
except KeyboardInterrupt:
103
raise KeyboardInterrupt()
105
if not os.path.exists(self.outputfile):
106
self.errors.append([__name__,'missing','[%s] %s'%(__name__,self.outputfile)])
107
logger.error('%s\t%s',*self.errors[-1][1:])
110
'''Command prompt help.'''
111
return "\n%s\n\tUsage:\n\tfrom %s import filter\n"%(
112
os.path.basename(sys.argv[0]),
113
os.path.splitext(os.path.basename(sys.argv[0]))[0]
116
licensetxt=u'''CorpusFiltergraph™ v4.0
117
Copyright © 2010-2012 Precision Translation Tools Co., Ltd.
119
This program is free software: you can redistribute it and/or modify
120
it under the terms of the GNU Lesser General Public License as published by
121
the Free Software Foundation, either version 3 of the License, or
122
(at your option) any later version.
124
This program is distributed in the hope that it will be useful,
125
but WITHOUT ANY WARRANTY; without even the implied warranty of
126
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
127
GNU Lesser General Public License for more details.
129
You should have received a copy of the GNU Lesser General Public License
130
along with this program. If not, see http://www.gnu.org/licenses/.
132
For more information, please contact Precision Translation Tools Co., Ltd.
133
at: http://www.precisiontranslationtools.com'''
135
if __name__ == "__main__":
138
sys.stdout.write(usage().encode('utf8')+'\n')