def check_segments():
segments = []
infile = codecs.open(filename, 'r', 'utf-8')
headers = TagTester(u'@30', infile, u'@30מצוה ([\u05d0-\u05ea"]{1,5})').grab_each_header()
tester = TagTester(u'@44', infile, u'@44\(([\u05d0-\u05ea]{1,2})\)')
while not tester.eof:
segments.append(tester.grab_each_header(u'@30מצוה ([\u05d0-\u05ea"]{1,5})', 1))
infile.close()
for sec_number, section in enumerate(segments):
index = 1
for title in section:
title = title.replace(u'"', u'')
count = util.getGematria(title)
if count != index:
print headers[sec_number-1]
print util.numToHeb(index)
index = count
index += 1
def check_segments():
segments = []
infile = codecs.open(filename, 'r', 'utf-8')
headers = TagTester(u'@30', infile,
u'@30מצוה ([\u05d0-\u05ea"]{1,5})').grab_each_header()
tester = TagTester(u'@44', infile, u'@44\(([\u05d0-\u05ea]{1,2})\)')
while not tester.eof:
segments.append(
tester.grab_each_header(u'@30מצוה ([\u05d0-\u05ea"]{1,5})', 1))
infile.close()
for sec_number, section in enumerate(segments):
index = 1
for title in section:
title = title.replace(u'"', u'')
count = util.getGematria(title)
if count != index:
print headers[sec_number - 1]
print util.numToHeb(index)
index = count
index += 1
def check_chapters():
with codecs.open('Minchat_Chinuch.txt', 'r', 'utf-8') as chinuch:
test = TagTester(u'@30', chinuch, u'@30מצוה ([\u05d0-\u05ea"]{1,5})')
index = 1
for header in test.grab_each_header(capture_group=1):
header = header.replace(u'"', u'')
count = util.getGematria(header)
if count != index:
print util.numToHeb(index)
index = count
index += 1
def check_chapters():
with codecs.open('Minchat_Chinuch.txt', 'r', 'utf-8') as chinuch:
test = TagTester(u'@30', chinuch, u'@30מצוה ([\u05d0-\u05ea"]{1,5})')
index = 1
for header in test.grab_each_header(capture_group=1):
header = header.replace(u'"', u'')
count = util.getGematria(header)
if count != index:
print util.numToHeb(index)
index = count
index += 1
def raph_alignment_report(ja_smk, letter_ja):
csv_lst = []
lst_raph = []
smk_siman = 0
smk_pages = map_semak_page_siman(ja_smk, to_print=False)
for seg in traverse_ja(ja_smk):
for raph_l_in_smk in re.finditer(u'@55([\u05d0-\u05ea]{1,3})', seg['data']):
lst_raph.append((raph_l_in_smk.group(1),
seg['data'][raph_l_in_smk.span()[0] - 20: raph_l_in_smk.span()[1] + 20],
(seg['indices'][0] + 1)))
raph_11 = []
for raph in traverse_ja(letter_ja):
raph_11.append(raph) # re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1))
page = 21
prob = 0
for raph, smk_l in zip(raph_11, lst_raph):
print re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1), smk_l[0], numToHeb(smk_l[2])
csv_dict = {u'smk letter': smk_l[0], u'raph letter': re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1),
u'smk words': smk_l[1], u'raph line': raph['data'], u'siman': numToHeb(smk_l[2]), u'aprx page in scan': smk_pages[numToHeb(smk_l[2])]}
if re.search(u'@77', smk_l[1]):
page += 1
if re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1) != smk_l[0]:
prob += 1
print "*"
csv_dict['problem'] = True
# break
csv_lst.append(csv_dict)
print 'prob', prob
print 'done'
toCSV(u'testcsvreport', csv_lst, [u'smk letter', u'raph letter', u'smk words',
u'raph line', u'siman', u'aprx page in scan', u'problem'])
return csv_lst
def hagahot_alignment(ja_smk, ja_raph, ja_hagahot):
ja_smk = JaggedArray(ja_smk)
ja_raph = JaggedArray(ja_raph)
ja_hagahot = JaggedArray(ja_hagahot)
# for i, seg_smk, j, seg_raph in zip(enumerate(ja_smk.array()), enumerate(ja_raph.array())):
dict_lst = []
dict = {u'siman': [], u'smk': [], u'raph': []}
for i, seg in enumerate(zip(ja_smk.array(), ja_raph.array())):
# print numToHeb(i+1)
dict['siman'] = numToHeb(i + 1)
for i, smk_line in enumerate(seg[0]):
hag_lett = re.findall(ur'@88\((?P<gim>[\u05d0-\u05ea]{1,3})\)',
smk_line)
if hag_lett:
dict['smk'].extend([(hag_l, i + 1) for hag_l in hag_lett])
# print [getGematria(lett) for lett in hag_lett]
# print 'RAPH'
for i, raph_line in enumerate(seg[1]):
hag_lett = re.findall(ur'@88\((?P<gim>[\u05d0-\u05ea]{1,3})\)',
raph_line)
if hag_lett:
dict['raph'].extend([(hag_l, i + 1) for hag_l in hag_lett])
# print [getGematria(lett) for lett in hag_lett]
dict_lst.append(dict)
dict = {u'siman': [], u'smk': [], u'raph': []}
return dict_lst
def chapter_seven(number):
hebrew_letter = util.numToHeb(number)
chapter = SchemaNode()
chapter.add_title('Chapter {}'.format(number), "en", primary=True)
chapter.add_title(u'{} {}'.format(u'סימן', hebrew_letter), "he", primary=True)
chapter.key = 'Chapter {}'.format(number)
chapter.append(create_intro_nodes())
chapter.append(create_shorash_node())
return chapter
def chapter_nine(number):
hebrew_letter = util.numToHeb(number)
chapter = JaggedArrayNode()
chapter.add_title('Chapter {}'.format(number), "en", primary=True)
chapter.add_title(u'{} {}'.format(u'סימן', hebrew_letter), "he", primary=True)
chapter.key = 'Chapter {}'.format(number)
chapter.depth = 2
chapter.addressTypes = ["Integer", "Integer"]
chapter.sectionNames = ["Section", "Mitzvah"]
return chapter
def generate_URLs(books):
urls = []
opening = u"""https://he.wikisource.org/wiki/מלבי"ם_על_"""
for book_title in books:
book = library.get_index(book_title)
heTitle = book.get_title('he')
for perek_n, perek in enumerate(book.all_section_refs()):
url = u"{}{}_{}".format(opening, heTitle, numToHeb(perek_n+1))
urls.append((url, book_title, perek_n+1))
return urls
def output(self, filename=u'temp_result.txt'):
full_text = []
for com_siman, base_siman in zip(self.commentary_simanim, self.base_simanim):
if not self.almost_equals(com_siman['total'], base_siman['total_refs']):
print "Divergence in siman {}".format(base_siman['num'])
full_text.append(u'@12{}\n'.format(numToHeb(base_siman['num'])))
full_text.extend(self.source_lines[com_siman['start']:com_siman['end']+1])
with codecs.open(filename, 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
def regular_chapter_nodes(number):
hebrew_letter = util.numToHeb(number)
chapter = JaggedArrayNode()
chapter.add_title('Chapter {}'.format(number), "en", primary=True)
chapter.add_title(u'{} {}'.format(u'סימן',hebrew_letter), "he", primary=True)
chapter.key = 'Chapter {}'.format(number)
chapter.depth = 1
chapter.addressTypes = ["Integer"]
chapter.sectionNames = ["Comment"]
return chapter
def output(self, filename=u'temp_result.txt'):
full_text = []
for com_siman, base_siman in zip(self.commentary_simanim, self.base_simanim):
if not self.almost_equals(com_siman['total'], base_siman['total_refs']):
print "Divergence in siman {}. {} in base and {} in commentary".format\
(base_siman['num'], base_siman['total_refs'], com_siman['total'])
full_text.append(u'@12{}\n'.format(numToHeb(base_siman['num'])))
full_text.extend(self.source_lines[com_siman['start']:com_siman['end']+1])
with codecs.open(filename, 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
def insert_chapter_marker(filename, safe_mode=False):
with codecs.open(filename, 'r', 'utf-8') as infile:
lines = infile.readlines()
count = 0
new_lines = []
for line in lines:
if re.search(u'^@22\u05d0( |$)', line) is not None:
count += 1
new_lines.append(u'@00\u05e4\u05e8\u05e7 {}\n{}'.format(numToHeb(count), line))
else:
new_lines.append(line)
if safe_mode:
filename += '.tmp'
with codecs.open(filename, 'w', 'utf-8') as outfile:
outfile.writelines(new_lines)
def insert_chapter_marker(filename, safe_mode=False):
with codecs.open(filename, 'r', 'utf-8') as infile:
lines = infile.readlines()
count = 0
new_lines = []
for line in lines:
if re.search(u'^@22\u05d0( |$)', line) is not None:
count += 1
new_lines.append(u'@00\u05e4\u05e8\u05e7 {}\n{}'.format(
numToHeb(count), line))
else:
new_lines.append(line)
if safe_mode:
filename += '.tmp'
with codecs.open(filename, 'w', 'utf-8') as outfile:
outfile.writelines(new_lines)
def fix_file(filepath, start_siman, test_mode=False):
output_list = []
with codecs.open(filepath, 'r', 'utf-8') as fp:
lines = fp.readlines()
counter = 0
for line in lines:
match = re.match(u'^@11([\u05d0-\u05ea]{1,3})$', line)
if match and getGematria(match.group(1)) == 1:
output_list.append(u'@00{}\n'.format(
numToHeb(counter + start_siman)))
counter += 1
output_list.append(line)
if test_mode:
filepath = re.sub(ur'\.txt$', u'_test.txt', filepath)
with codecs.open(filepath, 'w', 'utf-8') as fp:
fp.writelines(output_list)
def map_semak_page_siman(smk_ja, to_print=True):
'''
create a dictionary from key: siman value: page(s) that the siman is on
:param smk_ja: smk ja parsed according to simanim @22
:return: dictionary. keys: siman (he letter), value: list of pages the siman spans over. (pages according to scan -
starts on p. 21)
'''
siman_page = OrderedDict()
page_count = 21
start_page = False
lst_seg = {'data': '', 'indices': []}
for seg in traverse_ja(smk_ja):
for i, page in enumerate(re.finditer(u'@77', seg['data'])):
page_count += 1
try:
siman_page[numToHeb(seg['indices'][0] + 1)].append(page_count)
except KeyError:
if not start_page:
siman_page[numToHeb(seg['indices'][0] +
1)] = [page_count - 1, page_count]
start_page = False
else:
siman_page[numToHeb(seg['indices'][0] + 1)] = [page_count]
if re.search(u'@77 ?$', lst_seg['data']):
start_page = True
siman_page[numToHeb(lst_seg['indices'][0] +
1)].remove(page_count)
if not list(re.finditer(u'@77', seg['data'])):
try:
siman_page[numToHeb(seg['indices'][0] + 1)]
except KeyError:
siman_page[numToHeb(seg['indices'][0] + 1)] = [page_count]
if re.search(u'@77 ?$', lst_seg['data']):
start_page = True
try:
siman_page[numToHeb(lst_seg['indices'][0] +
1)].remove(page_count)
except ValueError:
pass
lst_seg = seg
if to_print:
for k in siman_page.keys():
print k, siman_page[k]
return siman_page
def map_semak_page_siman(smk_ja, to_print = True):
'''
create a dictionary from key: siman value: page(s) that the siman is on
:param smk_ja: smk ja parsed according to simanim @22
:return: dictionary. keys: siman (he letter), value: list of pages the siman spans over. (pages according to scan -
starts on p. 21)
'''
siman_page = OrderedDict()
page_count = 21
start_page = False
lst_seg = {'data': '', 'indices': []}
for seg in traverse_ja(smk_ja):
for i, page in enumerate(re.finditer(u'@77', seg['data'])):
page_count += 1
try:
siman_page[numToHeb(seg['indices'][0]+1)].append(page_count)
except KeyError:
if not start_page:
siman_page[numToHeb(seg['indices'][0] + 1)] = [page_count - 1, page_count]
start_page = False
else:
siman_page[numToHeb(seg['indices'][0]+1)] = [page_count]
if re.search(u'@77 ?$', lst_seg['data']):
start_page = True
siman_page[numToHeb(lst_seg['indices'][0] + 1)].remove(page_count)
if not list(re.finditer(u'@77', seg['data'])):
try:
siman_page[numToHeb(seg['indices'][0]+1)]
except KeyError:
siman_page[numToHeb(seg['indices'][0] + 1)] = [page_count]
if re.search(u'@77 ?$', lst_seg['data']):
start_page = True
try:
siman_page[numToHeb(lst_seg['indices'][0] + 1)].remove(page_count)
except ValueError:
pass
lst_seg = seg
if to_print:
for k in siman_page.keys():
print k, siman_page[k]
return siman_page
def hagahot_alignment(ja_smk, ja_raph, ja_hagahot):
ja_smk = JaggedArray(ja_smk)
ja_raph = JaggedArray(ja_raph)
ja_hagahot = JaggedArray(ja_hagahot)
# for i, seg_smk, j, seg_raph in zip(enumerate(ja_smk.array()), enumerate(ja_raph.array())):
dict_lst = []
dict = {u'siman':[], u'smk':[], u'raph':[]}
for i, seg in enumerate(zip(ja_smk.array(), ja_raph.array())):
# print numToHeb(i+1)
dict['siman'] = numToHeb(i+1)
for i, smk_line in enumerate(seg[0]):
hag_lett = re.findall(ur'@88\((?P<gim>[\u05d0-\u05ea]{1,3})\)', smk_line)
if hag_lett:
dict['smk'].extend([(hag_l, i+1) for hag_l in hag_lett])
# print [getGematria(lett) for lett in hag_lett]
# print 'RAPH'
for i, raph_line in enumerate(seg[1]):
hag_lett = re.findall(ur'@88\((?P<gim>[\u05d0-\u05ea]{1,3})\)', raph_line)
if hag_lett:
dict['raph'].extend([(hag_l, i+1) for hag_l in hag_lett])
# print [getGematria(lett) for lett in hag_lett]
dict_lst.append(dict)
dict = {u'siman': [], u'smk': [], u'raph': []}
return dict_lst
def raph_alignment_report(ja_smk, letter_ja):
csv_lst = []
lst_raph = []
smk_siman = 0
smk_pages = map_semak_page_siman(ja_smk, to_print=False)
for seg in traverse_ja(ja_smk):
for raph_l_in_smk in re.finditer(u'@55([\u05d0-\u05ea]{1,3})', seg['data']):
lst_raph.append((raph_l_in_smk.group(1),
seg['data'][raph_l_in_smk.span()[0] - 20: raph_l_in_smk.span()[1] + 20],
(seg['indices'][0] + 1)))
raph_11 = []
for raph in traverse_ja(letter_ja):
raph_11.append(raph) # re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1))
page = 21
prob = 0
i = 0
for raph, smk_l in zip(letter_ja, lst_raph): # zip(raph_11, lst_raph):
# print re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1), smk_l[0], numToHeb(smk_l[2])
csv_dict = {u'smk letter': smk_l[0], u'raph': raph[i], u'siman': numToHeb(smk_l[2]), u'aprx page in scan': smk_pages[numToHeb(smk_l[2])]}
# u'raph letter': re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1), u'raph line': raph['data']
# u'smk words': smk_l[1],
i += 0
if re.search(u'@77', smk_l[1]):
page += 1
# if re.search(u'@11([\u05d0-\u05ea]{1,3})', raph['data']).group(1) != smk_l[0]:
# prob += 1
# print "*"
# csv_dict['problem'] = True
# # break
csv_lst.append(csv_dict)
print 'prob', prob
print 'done'
toCSV(u'testcsvreport', csv_lst, [u'smk letter', u'raph',
u'siman', u'aprx page in scan']) #, u'problem', u'smk words',u'raph line',
return csv_lst
def test_num_to_heb():
assert util.numToHeb(16) == 'טז'
assert util.numToHeb(962) == 'תתקסב'
def mark_simanim(volume_number):
def transition(index_a, index_b):
return (index_b - index_a) % 22 != 1 and index_b == 1
def terminate():
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
full_text.extend(lines[line_num:])
with codecs.open('temp_result.txt', 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
def get_next_siman(siman_list):
siman, total_refs = None, 0
while total_refs == 0:
siman = siman_list.next()
total_refs = len(siman.locate_references(u'@44'))
return siman, total_refs
with codecs.open(filenames['part_{}'.format(volume_number)], 'r', 'utf-8') as infile:
lines = infile.readlines()
volume = Root('../../Orach_Chaim.xml').get_base_text().get_volume(1)
simanim = iter(volume.get_child())
current_siman, expected_refs = get_next_siman(simanim)
full_text, current_siman_text = [],[]
count = 0
seif_markers = (None, None)
for line_num, line in enumerate(lines):
match = re.search(u'^@11([\u05d0-\u05ea])', line)
if match:
count += 1
seif_markers = (seif_markers[1], he_ord(match.group(1)))
if count - expected_refs == 1:
if match.group(1) == u'א':
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
current_siman_text = []
count = 1
try:
current_siman, expected_refs = get_next_siman(simanim)
except StopIteration:
print "Ran out of Simanim"
terminate()
return
else:
print "Siman {}: Completed refs before transition occurred".format(current_siman.num)
terminate()
return
elif None not in seif_markers and transition(*seif_markers):
print "Siman {}: Transition occurred before completing refs".format(current_siman.num)
terminate()
return
current_siman_text.append(line)
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
with codecs.open('temp_result.txt', 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
def terminate():
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
full_text.extend(lines[line_num:])
with codecs.open('temp_result.txt', 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
sys.path.insert(0, p)
from local_settings import *
sys.path.insert(0, SEFARIA_PROJECT_PATH)
os.environ['DJANGO_SETTINGS_MODULE'] = "local_settings"
from data_utilities.util import numToHeb
reload(sys)
sys.setdefaultencoding("utf-8")
def wikiGet(url, title):
try:
opener = urllib2.build_opener()
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
page = opener.open(url)
print "got", title
with open("./pages/{}".format(title), "w") as file:
file.write(page.read())
except:
print "page doesn't exist", title
for siman in range(1, 697): #696 simanim in O.C
title = "Biur_Halacha." + str(siman)
wikiGet(
u"https://he.wikisource.org/w/index.php?title=ביאור_הלכה_על_אורח_חיים_%s&printable=yes"
% (numToHeb(siman)), title)
def repl(m):
siman = getGematria(m.group(1))
if siman >= increment_start:
siman -= 1
return u'@00{}'.format(numToHeb(siman))
def mark_simanim(volume_number):
def transition(index_a, index_b):
return (index_b - index_a) % 22 != 1 and index_b == 1
def terminate():
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
full_text.extend(lines[line_num:])
with codecs.open('temp_result.txt', 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
def get_next_siman(siman_list):
siman, total_refs = None, 0
while total_refs == 0:
siman = siman_list.next()
total_refs = len(siman.locate_references(u'@44'))
return siman, total_refs
with codecs.open(filenames['part_{}'.format(volume_number)], 'r',
'utf-8') as infile:
lines = infile.readlines()
volume = Root('../../Orach_Chaim.xml').get_base_text().get_volume(1)
simanim = iter(volume.get_child())
current_siman, expected_refs = get_next_siman(simanim)
full_text, current_siman_text = [], []
count = 0
seif_markers = (None, None)
for line_num, line in enumerate(lines):
match = re.search(u'^@11([\u05d0-\u05ea])', line)
if match:
count += 1
seif_markers = (seif_markers[1], he_ord(match.group(1)))
if count - expected_refs == 1:
if match.group(1) == u'א':
full_text.append(u'@12{}\n'.format(
numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
current_siman_text = []
count = 1
try:
current_siman, expected_refs = get_next_siman(simanim)
except StopIteration:
print "Ran out of Simanim"
terminate()
return
else:
print "Siman {}: Completed refs before transition occurred".format(
current_siman.num)
terminate()
return
elif None not in seif_markers and transition(*seif_markers):
print "Siman {}: Transition occurred before completing refs".format(
current_siman.num)
terminate()
return
current_siman_text.append(line)
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
with codecs.open('temp_result.txt', 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
def terminate():
full_text.append(u'@12{}\n'.format(numToHeb(current_siman.num)))
full_text.extend(current_siman_text)
full_text.extend(lines[line_num:])
with codecs.open('temp_result.txt', 'w', 'utf-8') as outfile:
outfile.writelines(full_text)
