def main(argv):
import optparse
parser = optparse.OptionParser(usage='usage: %prog [options]',
version='%prog 0.1',
description='make tocs')
parser.add_option('--human',
action='store_true',
default=False,
help='print some human-readable stuff')
global opts
opts, args = parser.parse_args(argv)
doc = ''
callback = None
if len(args) == 4:
(item_id, doc, path, callback) = args
elif len(args) == 3:
(item_id, doc, path) = args
else:
(book_id, ) = args
path = book_id
book_id = args[0]
iabook = Book(book_id, doc, path)
global scandata_ns
scandata_ns = iabook.get_scandata_ns()
if djvu:
pages = iabook.get_pages_as_djvu()
else:
pages = iabook.get_pages_as_abbyy()
pages = filter(pages)
pages = annotate(pages)
def clear_page(page):
page.clear()
windowed_pages = windowed_iterator(pages, 5, clear_page)
pages = analyze(windowed_pages)
toc_result = make_toc.make_toc(iabook, pages, hardcode_toc_pages,
hardcode_nottoc_pages)
toc_result['contents_leafnos'] = iabook.get_contents_indices()
toc_result['readable'] = print_readable(toc_result['qdtoc'])
if opts.human:
for r in ('readable', 'comments', 'isok'):
print r + ':'
print toc_result[r]
print
else:
if callback is not None:
print '%s(' % callback
print json.dumps(toc_result)
if callback is not None:
print ')'
else:
print json.dumps(toc_result, indent=4)
def main(argv):
import optparse
parser = optparse.OptionParser(usage='usage: %prog [options]',
version='%prog 0.1',
description='make tocs')
parser.add_option('--human',
action='store_true',
default=False,
help='print some human-readable stuff')
global opts
opts, args = parser.parse_args(argv)
doc = ''
callback = None
if len(args) == 4:
(item_id, doc, path, callback) = args
elif len(args) == 3:
(item_id, doc, path) = args
else:
(book_id,) = args
path = book_id
book_id = args[0]
iabook = Book(book_id, doc, path)
global scandata_ns
scandata_ns = iabook.get_scandata_ns()
if djvu:
pages = iabook.get_pages_as_djvu()
else:
pages = iabook.get_pages_as_abbyy()
pages = filter(pages)
pages = annotate(pages)
def clear_page(page):
page.clear()
windowed_pages = windowed_iterator(pages, 5, clear_page)
pages = analyze(windowed_pages)
toc_result = make_toc.make_toc(iabook, pages, hardcode_toc_pages, hardcode_nottoc_pages)
toc_result['contents_leafnos'] = iabook.get_contents_indices()
toc_result['readable'] = print_readable(toc_result['qdtoc'])
if opts.human:
for r in ('readable', 'comments', 'isok'):
print r + ':'
print toc_result[r]
print
else:
if callback is not None:
print '%s(' % callback
print json.dumps(toc_result)
if callback is not None:
print ')'
else:
print json.dumps(toc_result, indent=4)
def local_monotonic_p(seq):
w = windowed_iterator(seq, 1)
acceptable = 1
for p in w:
if p is 0:
yield False
diffs = 0
expected = 0
for n in w.neighbors(1):
if n is not 0:
diffs += abs(p - n)
expected += 1
if expected > 0 and diffs <= expected + acceptable:
yield True
else:
yield False
def local_monotonic_p(seq):
w = windowed_iterator(seq, 1)
acceptable = 1
for p in w:
if p is 0:
yield False
diffs = 0
expected = 0
for n in w.neighbors(1):
if n is not 0:
diffs += abs(p - n)
expected += 1
if expected > 0 and diffs <= expected + acceptable:
yield True
else:
yield False
def main(args):
path = args[0]
pageNum = int(args[1])
callback = args[2]
if not re.match('^/\d{1,2}/items/.+_djvu.xml$', path):
sys.exit(-1)
if ('ttsNextPageCB' != callback):
callback = 'ttsStartCB'
f = open(path)
context = etree.iterparse(f, tag='OBJECT')
def drop_event(iter):
for event, page in iter:
yield page
pages = drop_event(context)
def clear_page(page):
page.clear()
pages = windowed_iterator(pages, windowsize, clear_page)
for i, page in enumerate(pages):
if i == pageNum:
break
hfs = guess_hfs(page, pages)
lines = page.findall('.//LINE')
#print 'got %s .//lines' % len(lines)
textBlocks = []
block = ''
rects = []
numWords = 0
for line in lines:
# skip headers/footers
if line in hfs:
continue
top = sys.maxint
left = sys.maxint
right = -1
bottom = -1
numWordsInLine = 0
words = line.findall('.//WORD')
#print 'at start of line, rects ='
#print rects
for word in words:
numWordsInLine += 1
text = word.text
#print 'got text ' + text
coords = word.get('coords').split(',') #l,b,r,t
coords = map(int, coords)
if int(coords[0]) < left:
left = coords[0]
if coords[1] > bottom:
bottom = coords[1]
if coords[2] > right:
right = coords[2]
if coords[3] < top:
top = coords[3]
block += word.text + ' '
numWords += 1
if text.endswith('.') and (numWords > minWordsInBlock):
#print 'end of block with numWords=%d' % numWords
#print 'block = ' + block
rects.append([left, bottom, right, top])
#textBlocks.append(block.strip())
rects.insert(0, block.strip())
textBlocks.append(rects)
block = ''
rects = []
numWords = 0
numWordsInLine = 0
top = sys.maxint
left = sys.maxint
right = -1
bottom = -1
#end of line
if numWordsInLine > 0:
rects.append([left, bottom, right, top])
if numWords > maxWordsInBlock:
#textBlocks.append(block.strip())
rects.insert(0, block.strip())
textBlocks.append(rects)
block = ''
numWords = 0
rects = []
#print 'at end of line, rects ='
#print rects
if '' != block:
#textBlocks.append(block.strip())
rects.insert(0, block.strip())
textBlocks.append(rects)
print 'br.%s(%s);' % (callback, json.dumps(textBlocks))
def main(argv):
import optparse
parser = optparse.OptionParser(usage='usage: %prog [options]',
version='%prog 0.1',
description='make tocs')
parser.add_option('--in_deriver',
action='store_true',
default=False,
help='match deriver-style args')
parser.add_option('--simpletoc',
action='store_true',
default=False,
help='do a simple toc analysis')
parser.add_option('--human',
action='store_true',
default=False,
help='print some human-readable stuff')
global opts
opts, args = parser.parse_args(argv)
if opts.in_deriver:
(dvju_xml_path, scandata_path) = args
iabook = DeriverBook(dvju_xml_path, scandata_path)
else:
doc = ''
callback = None
if len(args) == 4:
(item_id, doc, path, callback) = args
elif len(args) == 3:
(item_id, doc, path) = args
else:
(book_id, ) = args
doc = path = book_id
book_id = args[0]
iabook = Book(book_id, doc, path)
global scandata_ns
scandata_ns = iabook.get_scandata_ns()
if djvu:
pages = iabook.get_pages_as_djvu()
else:
pages = iabook.get_pages_as_abbyy()
pages = filter(pages)
pages = annotate(pages)
def clear_page(page):
page.clear()
windowed_pages = windowed_iterator(pages, 5, clear_page)
pages = analyze(windowed_pages)
if opts.simpletoc:
toc_result = make_toc.simple_make_toc(iabook, pages)
else:
toc_result = make_toc.make_toc(iabook, pages, hardcode_toc_pages,
hardcode_nottoc_pages)
# toc_result['readable'] = print_readable(toc_result['qdtoc'])
if opts.simpletoc:
xml = toc_to_xml.make_xml(toc_result['qdtoc_tuples'], version)
print etree.tostring(xml, pretty_print=True), # pretty_print adds nl
elif opts.human:
for r in ('readable', 'comments', 'isok'):
print r + ':'
print toc_result[r]
print
else:
if callback is not None:
print '%s(' % callback
print json.dumps(toc_result)
# print_one_per_line(qdtoc)
if callback is not None:
print ')'
else:
print json.dumps(toc_result, indent=4)
def main(args):
path = args[0]
pageNum = int(args[1])
callback = args[2]
if not re.match("^/\d{1,2}/items/.+_djvu.xml$", path):
sys.exit(-1)
if "ttsNextPageCB" != callback:
callback = "ttsStartCB"
f = open(path)
context = etree.iterparse(f, tag="OBJECT")
def drop_event(iter):
for event, page in iter:
yield page
pages = drop_event(context)
def clear_page(page):
page.clear()
pages = windowed_iterator(pages, windowsize, clear_page)
for i, page in enumerate(pages):
if i == pageNum:
break
hfs = guess_hfs(page, pages)
lines = page.findall(".//LINE")
# print 'got %s .//lines' % len(lines)
textBlocks = []
block = ""
rects = []
numWords = 0
for line in lines:
# skip headers/footers
if line in hfs:
continue
top = sys.maxint
left = sys.maxint
right = -1
bottom = -1
numWordsInLine = 0
words = line.findall(".//WORD")
# print 'at start of line, rects ='
# print rects
for word in words:
numWordsInLine += 1
text = word.text
# print 'got text ' + text
coords = word.get("coords").split(",") # l,b,r,t
coords = map(int, coords)
if int(coords[0]) < left:
left = coords[0]
if coords[1] > bottom:
bottom = coords[1]
if coords[2] > right:
right = coords[2]
if coords[3] < top:
top = coords[3]
block += word.text + " "
numWords += 1
if text.endswith(".") and (numWords > minWordsInBlock):
# print 'end of block with numWords=%d' % numWords
# print 'block = ' + block
rects.append([left, bottom, right, top])
# textBlocks.append(block.strip())
rects.insert(0, block.strip())
textBlocks.append(rects)
block = ""
rects = []
numWords = 0
numWordsInLine = 0
top = sys.maxint
left = sys.maxint
right = -1
bottom = -1
# end of line
if numWordsInLine > 0:
rects.append([left, bottom, right, top])
if numWords > maxWordsInBlock:
# textBlocks.append(block.strip())
rects.insert(0, block.strip())
textBlocks.append(rects)
block = ""
numWords = 0
rects = []
# print 'at end of line, rects ='
# print rects
if "" != block:
# textBlocks.append(block.strip())
rects.insert(0, block.strip())
textBlocks.append(rects)
print "br.%s(%s);" % (callback, json.dumps(textBlocks))
def main(argv):
import optparse
parser = optparse.OptionParser(usage='usage: %prog [options]',
version='%prog 0.1',
description='make tocs')
parser.add_option('--in_deriver',
action='store_true',
default=False,
help='match deriver-style args')
parser.add_option('--simpletoc',
action='store_true',
default=False,
help='do a simple toc analysis')
parser.add_option('--human',
action='store_true',
default=False,
help='print some human-readable stuff')
global opts
opts, args = parser.parse_args(argv)
if opts.in_deriver:
(dvju_xml_path, scandata_path) = args
iabook = DeriverBook(dvju_xml_path, scandata_path)
else:
doc = ''
callback = None
if len(args) == 4:
(item_id, doc, path, callback) = args
elif len(args) == 3:
(item_id, doc, path) = args
else:
(book_id,) = args
doc = path = book_id
book_id = args[0]
iabook = Book(book_id, doc, path)
global scandata_ns
scandata_ns = iabook.get_scandata_ns()
if djvu:
pages = iabook.get_pages_as_djvu()
else:
pages = iabook.get_pages_as_abbyy()
pages = filter(pages)
pages = annotate(pages)
def clear_page(page):
page.clear()
windowed_pages = windowed_iterator(pages, 5, clear_page)
pages = analyze(windowed_pages)
if opts.simpletoc:
toc_result = make_toc.simple_make_toc(iabook, pages)
else:
toc_result = make_toc.make_toc(iabook, pages, hardcode_toc_pages, hardcode_nottoc_pages)
# toc_result['readable'] = print_readable(toc_result['qdtoc'])
if opts.simpletoc:
xml = toc_to_xml.make_xml(toc_result['qdtoc_tuples'], version)
print etree.tostring(xml, pretty_print=True), # pretty_print adds nl
elif opts.human:
for r in ('readable', 'comments', 'isok'):
print r + ':'
print toc_result[r]
print
else:
if callback is not None:
print '%s(' % callback
print json.dumps(toc_result)
# print_one_per_line(qdtoc)
if callback is not None:
print ')'
else:
print json.dumps(toc_result, indent=4)