def do_book_comparison(book_index_xml):
#check number of chapters.
#compare to the parsed Leningrad itself
#compare to the current sefaria versions.
canonical_name = book_index_xml.find('./names/name').text
print canonical_name
diff_file = open("results/%s_wlc_koren.html" % canonical_name, 'wb+')
length_results = open("results/length_comparison.txt", 'ab+')
wlc_chapter_count = int(book_index_xml.find(
'./cs').text) #listed length of the leningrad chapters
with open("preprocess_json/%s.json" % canonical_name, 'r') as filep:
wlc_text = json.load(filep)['text']
wlc_real_chapter_count = len(
wlc_text) #physical length of the parsed leningrad chapters
sefaria_book = Helper.getKnownTexts(canonical_name)
sefria_chapter_count = sefaria_book['lengths'][0]
sefaria_text = Helper.api_get_text(
"%s 1-%s" % (sefaria_book['title'], sefria_chapter_count), 'he',
"Tanach with Ta'amei Hamikra")['he']
if sefria_chapter_count == 1:
sefaria_text = [sefaria_text]
sefaria_real_chapter_count = len(sefaria_text)
if not all_same([
wlc_chapter_count, wlc_real_chapter_count, sefria_chapter_count,
sefaria_real_chapter_count
]):
ch_res_str = "%s: Leningrad has %s chapters listed and %s chapters in text. Sefaria version has %s chapters listed and %s chapters in text\n" % (
canonical_name.encode('utf-8'), wlc_chapter_count,
wlc_real_chapter_count, sefria_chapter_count,
sefaria_real_chapter_count)
length_results.write(ch_res_str)
for chapter in book_index_xml.findall('c'):
ch_num = int(chapter.get('n'))
wlc_verse_count = int(chapter.find('vs').text)
wlc_real_verse_count = len(wlc_text[ch_num - 1])
sefaria_real_verse_count = len(sefaria_text[ch_num - 1])
if not all_same(
[wlc_verse_count, wlc_real_verse_count, sefaria_real_verse_count]):
v_res_str = "%s:%s Leningrad has %s verses listed and %s verses in text. Sefaria version has %s verses\n" % (
canonical_name.encode('utf-8'), ch_num, wlc_verse_count,
wlc_real_verse_count, sefaria_real_verse_count)
length_results.write(v_res_str)
html_diff = difflib.HtmlDiff().make_file(
flatten_text(make_consonantal_text(wlc_text)),
flatten_text(make_consonantal_text(sefaria_text)), 'Leningrad Codex',
'Sefaria/Koren')
html_diff = html_diff.replace('charset=ISO-8859-1', 'charset=utf-8')
diff_file.write(html_diff.encode('utf-8'))
length_results.close()
diff_file.close()
def run_post_to_api(sub_directory=None):
directory = "preprocess_json/%s" % sub_directory if sub_directory else "preprocess_json"
wlc_index_xml = ET.parse("source/TanachIndex.xml") # this lists num of chapters and verses for all books in the WLC
books_xml_r = wlc_index_xml.getroot().find("tanach")
for book in books_xml_r.findall("book"):
canonical_name = book.find("./names/name").text
print canonical_name
with open("%s/%s.json" % (directory, canonical_name), "r") as filep:
file_text = filep.read()
sefaria_book = Helper.getKnownTexts(canonical_name)
Helper.postText(sefaria_book["title"], file_text, False)
def run_post_to_api(sub_directory=None):
directory = "preprocess_json/%s" % sub_directory if sub_directory else "preprocess_json"
wlc_index_xml = ET.parse(
'source/TanachIndex.xml'
) #this lists num of chapters and verses for all books in the WLC
books_xml_r = wlc_index_xml.getroot().find('tanach')
for book in books_xml_r.findall('book'):
canonical_name = book.find('./names/name').text
print canonical_name
with open("%s/%s.json" % (directory, canonical_name), 'r') as filep:
file_text = filep.read()
sefaria_book = Helper.getKnownTexts(canonical_name)
Helper.postText(sefaria_book['title'], file_text, False)
def do_book_comparison(book_index_xml):
# check number of chapters.
# compare to the parsed Leningrad itself
# compare to the current sefaria versions.
canonical_name = book_index_xml.find("./names/name").text
print canonical_name
diff_file = open("results/%s_wlc_koren.html" % canonical_name, "wb+")
length_results = open("results/length_comparison.txt", "ab+")
wlc_chapter_count = int(book_index_xml.find("./cs").text) # listed length of the leningrad chapters
with open("preprocess_json/%s.json" % canonical_name, "r") as filep:
wlc_text = json.load(filep)["text"]
wlc_real_chapter_count = len(wlc_text) # physical length of the parsed leningrad chapters
sefaria_book = Helper.getKnownTexts(canonical_name)
sefria_chapter_count = sefaria_book["lengths"][0]
sefaria_text = Helper.api_get_text(
"%s 1-%s" % (sefaria_book["title"], sefria_chapter_count), "he", "Tanach with Ta'amei Hamikra"
)["he"]
if sefria_chapter_count == 1:
sefaria_text = [sefaria_text]
sefaria_real_chapter_count = len(sefaria_text)
if not all_same([wlc_chapter_count, wlc_real_chapter_count, sefria_chapter_count, sefaria_real_chapter_count]):
ch_res_str = (
"%s: Leningrad has %s chapters listed and %s chapters in text. Sefaria version has %s chapters listed and %s chapters in text\n"
% (
canonical_name.encode("utf-8"),
wlc_chapter_count,
wlc_real_chapter_count,
sefria_chapter_count,
sefaria_real_chapter_count,
)
)
length_results.write(ch_res_str)
for chapter in book_index_xml.findall("c"):
ch_num = int(chapter.get("n"))
wlc_verse_count = int(chapter.find("vs").text)
wlc_real_verse_count = len(wlc_text[ch_num - 1])
sefaria_real_verse_count = len(sefaria_text[ch_num - 1])
if not all_same([wlc_verse_count, wlc_real_verse_count, sefaria_real_verse_count]):
v_res_str = (
"%s:%s Leningrad has %s verses listed and %s verses in text. Sefaria version has %s verses\n"
% (
canonical_name.encode("utf-8"),
ch_num,
wlc_verse_count,
wlc_real_verse_count,
sefaria_real_verse_count,
)
)
length_results.write(v_res_str)
html_diff = difflib.HtmlDiff().make_file(
flatten_text(make_consonantal_text(wlc_text)),
flatten_text(make_consonantal_text(sefaria_text)),
"Leningrad Codex",
"Sefaria/Koren",
)
html_diff = html_diff.replace("charset=ISO-8859-1", "charset=utf-8")
diff_file.write(html_diff.encode("utf-8"))
length_results.close()
diff_file.close()