def execute(indexobj, filename):
return filter.execfilter("pdftotext -enc UTF-8 %s -", filename)
def execute(indexobj, filename):
# using -q switch to get quiet operation (no messages, no errors),
# because poppler-utils pdftotext on Debian/Etch otherwise generates
# lots of output on stderr (e.g. 10MB stderr output) and that causes
# problems in current execfilter implementation.
return execfilter("pdftotext -q -enc UTF-8 %s -", filename)
def execute(indexobj, filename):
data = execfilter("catppt -dutf-8 %s", filename)
return data
def execute(indexobj, filename):
return filter.execfilter("antiword %s", filename)
def execute(indexobj, filename):
data = execfilter("catppt -dutf-8 %s", filename)
return data
def execute(indexobj, filename):
data = execfilter("xls2csv %s", filename)
# xls2csv uses comma as field separator and "field content",
# we strip both to not confuse the indexer
data = data.replace(u',', u' ').replace(u'"', u' ')
return data
def execute(indexobj, filename):
cmd = "antiword %s"
if os.name == 'posix':
cmd = "HOME=/tmp " + cmd # no HOME makes antiword complain (on Linux)
return execfilter(cmd, filename)
def execute(indexobj, filename):
return filter.execfilter("catdoc %s", filename)
def execute(indexobj, filename):
data = execfilter("xls2csv %s", filename)
# xls2csv uses comma as field separator and "field content",
# we strip both to not confuse the indexer
data = data.replace(u',', u' ').replace(u'"', u' ')
return data
def execute(indexobj, filename):
return execfilter("catdoc %s", filename)
