def process(dict_names, gram_fname, xml, encoding):
""""""
gram_file = open(gram_fname, "r")
gram = gram_file.read()
gram_file.close()
lexicon = parse_corpus(dict_names, grammar=gram, encoding=encoding, errors="replace")
mkr_counts, nonblank_mkr_counts = count_mkrs(lexicon)
analysis = analyse_dict(lexicon)
if xml:
indent(lexicon)
out_file = open(xml, "w")
out_file.write(ET.tostring(lexicon, encoding="UTF-8"))
out_file.close()
print "analysing files\n%s\n" % "\n".join(dict_names)
if xml:
print 'XML lexicon output in file "%s"\n' % xml
print "====chunk grammar===="
print gram
print "\n"
max_positions = 30
for structure, patt_dict in analysis.items():
print "\n\n===%s===: total= %d" % (structure, pattern_count(patt_dict))
for pattern, positions in sorted(patt_dict.items(), key=lambda t: (-len(t[1]), t[0])):
if len(positions) <= max_positions:
pos_str = "Entries: %s" % ", ".join(positions)
else:
pos_str = "Too many entries to list."
print "\t%5d: %s %s" % (len(positions), ":".join(pattern), pos_str)
print "\n\n"
print "mkr\tcount\tnonblank"
for mkr in mkr_counts:
print "%s\t%5d\t%5d" % (mkr, mkr_counts.get(mkr, 0), nonblank_mkr_counts.get(mkr, 0))
def process(dict_names, gram_fname, xml, encoding):
""""""
gram_file = open(gram_fname, 'r')
gram = gram_file.read()
gram_file.close()
lexicon = parse_corpus(dict_names,
grammar=gram,
encoding=encoding,
errors='replace')
mkr_counts, nonblank_mkr_counts = count_mkrs(lexicon)
analysis = analyse_dict(lexicon)
if xml:
indent(lexicon)
out_file = open(xml, "w")
out_file.write(ET.tostring(lexicon, encoding='UTF-8'))
out_file.close()
print 'analysing files\n%s\n' % '\n'.join(dict_names)
if xml:
print 'XML lexicon output in file "%s"\n' % xml
print '====chunk grammar===='
print gram
print '\n'
max_positions = 30
for structure, patt_dict in analysis.items():
print '\n\n===%s===: total= %d' % (structure, pattern_count(patt_dict))
for pattern, positions in sorted(patt_dict.items(),
key=lambda t: (-len(t[1]), t[0])):
if len(positions) <= max_positions:
pos_str = 'Entries: %s' % ', '.join(positions)
else:
pos_str = 'Too many entries to list.'
print "\t%5d: %s %s" % (len(positions), ':'.join(pattern),
pos_str)
print "\n\n"
print 'mkr\tcount\tnonblank'
for mkr in mkr_counts:
print '%s\t%5d\t%5d' % (mkr, mkr_counts.get(
mkr, 0), nonblank_mkr_counts.get(mkr, 0))
#!/usr/bin/env python
"""
Build the corpus package index. Usage:
build_pkg_index.py <path-to-packages> <base-url> <output-file>
"""
xml_header = """<?xml version="1.0"?>
<?xml-stylesheet href="index.xsl" type="text/xsl"?>
"""
import sys
from nltk.downloader import build_index
from nltk.etree import ElementTree
if len(sys.argv) != 4:
print "Usage: "
print "build_pkg_index.py <path-to-packages> <base-url> <output-file>"
sys.exit(-1)
ROOT, BASE_URL, OUT = sys.argv[1:]
index = build_index(ROOT, BASE_URL)
s = ElementTree.tostring(index)
out = open(OUT, 'w')
out.write(xml_header)
out.write(s)
out.close()
#!/usr/bin/env python
"""
Build the corpus package index. Usage:
build_pkg_index.py <path-to-packages> <base-url> <output-file>
"""
xml_header = """<?xml version="1.0"?>
<?xml-stylesheet href="index.xsl" type="text/xsl"?>
"""
import sys
from nltk.downloader import build_index
from nltk.etree import ElementTree
if len(sys.argv) != 4:
print "Usage: "
print "build_pkg_index.py <path-to-packages> <base-url> <output-file>"
sys.exit(-1)
ROOT, BASE_URL, OUT = sys.argv[1:]
index = build_index(ROOT, BASE_URL)
s = ElementTree.tostring(index)
out = open(OUT, 'w')
out.write(xml_header)
out.write(s)
out.close()
