def dump_text(base_dir, do_transliteration=False):
unit_info_file = os.path.join(os.path.dirname(text_data.__file__), "shatapatha.json")
titus_url = "http://titus.uni-frankfurt.de/texte/etcs/ind/aind/ved/yvw/sbm/sbm.htm"
for kaanda_index in text_data.get_subunit_list(json_file=unit_info_file, unit_path_list=[]):
sarga_list = text_data.get_subunit_list(json_file=unit_info_file, unit_path_list=[kaanda_index])
for sarga_index in sarga_list:
logging.info("kaanDa %d adhyaaya %d", kaanda_index, sarga_index)
outfile_path = os.path.join(base_dir, "%02d" % (kaanda_index), "%02d" % sarga_index + ".md")
if os.path.exists(outfile_path):
logging.info("Skipping " + outfile_path)
continue
titus.navigate_to_part(base_page_url=titus_url, level_3_id=kaanda_index, level_4_id=sarga_index)
sentences = titus.get_text()
lines = ["\n"]
for sentence in sentences:
sentence = roman.RomanScheme.simplify_accent_notation(sentence)
sentence = sentence.replace("/", ".")
if not sentence.endswith("."):
sentence = sentence + ".."
if do_transliteration:
if kaanda_index == 12:
sentence = sanscript.transliterate(sentence, sanscript.IAST, sanscript.DEVANAGARI)
else:
sentence = sanscript.transliterate(sentence, sanscript.TITUS, sanscript.DEVANAGARI)
sentence = roman.RomanScheme.to_shatapatha_svara(sentence)
lines.append(sentence + " \n")
os.makedirs(name=os.path.dirname(outfile_path), exist_ok=True)
with open(outfile_path, "w") as outfile:
outfile.writelines(lines)
def test_devanaagarii_equivalence():
"""Test all synonmous transliterations."""
logging.info(
sanscript.transliterate("rAmo gUDhaM vaktI~Ngitaj~naH kShetre",
sanscript.ITRANS, sanscript.DEVANAGARI))
assert sanscript.transliterate("rAmo gUDhaM vaktI~Ngitaj~naH kShetre", sanscript.ITRANS, sanscript.DEVANAGARI) == \
sanscript.transliterate("raamo guuDhaM vaktii~NgitaGYaH xetre", sanscript.ITRANS, sanscript.DEVANAGARI)
def get_headers(self): return [ self.gana_varga_naama, NIGHANTU_NAAMA + " - _" + self.index, transliterate(self.gana_varga_naama, DEVANAGARI, ITRANS), transliterate(NIGHANTU_NAAMA + " - _" + self.index, DEVANAGARI, ITRANS) ]
def get_file_path(out_dir, title_iast, author_iast="", catalog_number=""):
title_optitrans = sanscript.transliterate(data=title_iast, _from=sanscript.IAST, _to=sanscript.OPTITRANS)
author_optitrans = sanscript.transliterate(data=author_iast, _from=sanscript.IAST, _to=sanscript.OPTITRANS)
file_path = "%s_%s_%s.md" % (title_optitrans, author_optitrans, catalog_number.strip())
file_path = file_helper.clean_file_path(file_path=file_path)
file_path = os.path.join(out_dir, file_path)
return file_path
def get_item(id, dir_path):
import urllib.parse
dashaka_id = "नारायणीयम्/दशकम्_%s" % sanscript.transliterate(
str(id), sanscript.SLP1, sanscript.DEVANAGARI)
logging.info(dashaka_id)
item_url = "https://sa.wikisource.org/wiki/" + urllib.parse.quote(
dashaka_id)
logging.info(item_url)
browser.get(item_url)
text = browser.find_element_by_css_selector("div.poem").text
text = text.replace("cअ", "च").replace("cइ", "चि").replace(
"cई", "ची").replace("cउ", "चु").replace("cऊ", "चू").replace(
"cऋ", "चृ").replace("cॠ", "चॄ").replace("cऌ", "चॢ").replace(
"cॡ", "चॣ").replace("cए", "चे").replace("cऐ", "चै").replace(
"cओ", "चो").replace("cऔ",
"चौ").replace("c",
"च्").replace("ळ", "ल")
shlokas = text.split("\n\n")
outfile_path = os.path.join(dir_path, "%03d.md" % id)
os.makedirs(name=os.path.dirname(outfile_path), exist_ok=True)
with open(outfile_path, "w") as outfile:
for shloka_id in range(1, len(shlokas) + 1):
outfile.write(
"<div class=\"audioEmbed\" caption=\"सीतालक्ष्मी-वाचनम्\" src=\"https://sanskritdocuments.org/sites/completenarayaneeyam/SoundFiles/%03d/%03d_%02d.mp3\"></div> \n"
% (id, id, shloka_id))
outfile.writelines(shlokas[shloka_id - 1].replace("\n", " \n") +
"\n\n")
md_file = md_helper.MdFile(file_path=outfile_path)
md_file.set_title(sanscript.transliterate("%03d" % id, sanscript.SLP1,
sanscript.DEVANAGARI),
dry_run=False)
def fix_lazy_anusvaara(self, data_in, omit_sam=False, omit_yrl=False, ignore_padaanta=False):
from indic_transliteration import sanscript
if ignore_padaanta:
return self.fix_lazy_anusvaara_except_padaantas(data_in=data_in, omit_sam=omit_sam, omit_yrl=omit_yrl)
data_out = sanscript.transliterate(data=data_in, _from=self.name, _to=sanscript.DEVANAGARI)
data_out = sanscript.SCHEMES[sanscript.DEVANAGARI].fix_lazy_anusvaara(data_in=data_out, omit_sam=omit_sam, omit_yrl=omit_yrl)
return sanscript.transliterate(data=data_out, _from=sanscript.DEVANAGARI, _to=self.name)
def get_storage_name(text,
source_script=None,
max_length=50,
maybe_use_dravidian_variant=True,
mixed_languages_in_titles=True):
from indic_transliteration import detect
if source_script is None:
source_script = detect.detect(text=text)
text_optitrans = regex.sub("/ *", "__", text)
if source_script in roman.ALL_SCHEME_IDS:
if source_script in roman.CAPITALIZABLE_SCHEME_IDS:
if mixed_languages_in_titles:
text_optitrans = sanscript.SCHEMES[
sanscript.IAST].mark_off_non_indic_in_line(text_optitrans)
text_optitrans = sanscript.transliterate(
text_optitrans,
source_script,
sanscript.OPTITRANS,
suspend_on=set('<'),
suspend_off=set('>'),
maybe_use_dravidian_variant=maybe_use_dravidian_variant)
else:
text_optitrans = sanscript.transliterate(
text_optitrans,
source_script,
sanscript.OPTITRANS,
maybe_use_dravidian_variant=maybe_use_dravidian_variant)
storage_name = clean_file_path(text_optitrans)
if max_length is not None:
storage_name = storage_name[:max_length]
return storage_name
def extractFirstPage(im, coords):
"""
im: Image object
coords: List of size-4 tuples. Order: Box-1 containing mandal name, pin code etc.
Box-2 polling station name
Box-3 polling station address
Box-4 net electors male
Box-5 net electors female
Box-6 net electors third gender
Box-7 total electors
"""
im.seek(0)
resAll = dealWithFirstPage(im, coords)
res2 = resAll[0].split("\n")
resNoEmp = [i for i in res2 if not i == '']
main_town = (resNoEmp[0].split("ः")[1].strip() if len(
resNoEmp[0].split(":")) == 1 else resNoEmp[0].split(":")[1]).strip()
police_station = (resNoEmp[5].split("ः")[1].strip() if len(
resNoEmp[5].split(":")) == 1 else resNoEmp[5].split(":")[1]).strip()
pin_code = (resNoEmp[8].split("ः")[1].strip() if len(
resNoEmp[8].split(":")) == 1 else resNoEmp[8].split(":")[1]).strip()
main_town_eng = transliterate(main_town, sanscript.DEVANAGARI,
sanscript.ITRANS)
police_station_eng = transliterate(police_station, sanscript.DEVANAGARI,
sanscript.ITRANS)
polling_station_name = transliterate(resAll[1], sanscript.DEVANAGARI,
sanscript.ITRANS)
polling_station_address = transliterate(resAll[2], sanscript.DEVANAGARI,
sanscript.ITRANS)
return [
main_town_eng, police_station_eng, pin_code, polling_station_name,
polling_station_address, resAll[3], resAll[4], resAll[5], resAll[6],
main_town, police_station, resAll[1], resAll[2]
]
def get_headers(self): return [ self.varga_naama, NIGHANTU_NAAMA + ' _' + transliterate(self.index, ITRANS, DEVANAGARI), transliterate(self.varga_naama, DEVANAGARI, ITRANS), transliterate(NIGHANTU_NAAMA, DEVANAGARI, ITRANS) + ' _' + self.index, 'abdm _' + self.index ]
def transliterate(text, toDevanagari=False):
if toDevanagari:
text = sanscript.transliterate(text, sanscript.DEVANAGARI,
sanscript.IAST)
else:
text = sanscript.transliterate(text, sanscript.IAST,
sanscript.DEVANAGARI)
return text
def get_headers(self):
if not self.shloka_sankhya:
return []
shloka_sankhya_it = transliterate(self.shloka_sankhya, DEVANAGARI, ITRANS)
return [
NIGHANTU_NAAMA + ' ' + self.shloka_sankhya,
transliterate(NIGHANTU_NAAMA, DEVANAGARI, ITRANS) + ' ' + shloka_sankhya_it,
'abdm {}'.format(shloka_sankhya_it)
]
def get_standard_form(self, data):
"""Roman schemes define multiple representations of the same devanAgarI character. This method gets a library-standard representation.
data : a text in the given scheme.
"""
if self.synonym_map is None:
return data
from indic_transliteration import sanscript
return sanscript.transliterate(data=sanscript.transliterate(_from=self.name, _to=sanscript.DEVANAGARI, data=data), _from=sanscript.DEVANAGARI, _to=self.name)
def test_devanaagarii_equivalence(self):
"""Test all synonmous transliterations."""
print S.transliterate("rAmo gUDhaM vaktI~Ngitaj~naH kShetre", S.ITRANS,
S.DEVANAGARI),
self.assertEqual(
S.transliterate("rAmo gUDhaM vaktI~Ngitaj~naH kShetre", S.ITRANS,
S.DEVANAGARI),
S.transliterate("raamo guuDhaM vaktii~NgitaGYaH xetre", S.ITRANS,
S.DEVANAGARI))
def repl(matchobject):
data = transliterate(matchobject.group(0), 'devanagari', 'slp1')
data = re.sub('M([kKgG])', 'N\g<1>', data)
data = re.sub('M([cCjJ])', 'Y\g<1>', data)
data = re.sub('M([wWqQ])', 'R\g<1>', data)
data = re.sub('M([tTdD])', 'n\g<1>', data)
data = re.sub('M([pPbB])', 'm\g<1>', data)
data = transliterate(data, 'slp1', 'devanagari')
return data
def separate_rks(dry_run=False):
dest_dir_Rks = "/home/vvasuki/vvasuki-git/vedAH/static/atharva/shaunakam/rUDha-saMhitA/mUlam/"
suukta_paths = glob.glob("/home/vvasuki/vvasuki-git/vedAH/content/atharva/shaunakam/rUDha-saMhitA_alt/*/*.md",
recursive=True)
for suukta_path in suukta_paths:
md_file = MdFile(file_path=suukta_path)
[metadata, md] = md_file.read_md_file()
lines = md.split("\n")
meta_lines = list(itertools.takewhile(lambda line: "॒" not in line and "॑" not in line, lines))
lines = list(itertools.dropwhile(lambda line: "॒" not in line and "॑" not in line, lines))
lines = [line for line in lines if line != ""]
rk_id = 0
chapter_id = suukta_path.split("/")[-2]
suukta_id = metadata["title"].split()[0]
suukta_id_roman = sanscript.transliterate(suukta_id, sanscript.DEVANAGARI, sanscript.IAST)
suukta_title = " ".join(metadata["title"].split()[1:]).replace("।", "").strip()
dest_path_suukta = os.path.join(dest_dir_suuktas, chapter_id, suukta_id_roman + ".md")
rk_map = {}
while(len(lines) > 0):
lines_rk = list(itertools.takewhile(lambda line: "॥" not in line, lines))
lines_rk.append(lines[len(lines_rk)])
if len(lines) == len(lines_rk):
lines = []
else:
lines = lines[len(lines_rk):]
rk_id = rk_id + 1
rk_md = "\n".join(lines_rk)
rk_id_str = sanscript.transliterate("%02d" % rk_id, sanscript.IAST, sanscript.DEVANAGARI)
from doc_curation import text_data
title_Rk = text_data.get_rk_title(rk_id=rk_id_str, rk_text=rk_md)
dest_path_Rk = os.path.join(dest_dir_Rks, chapter_id, suukta_id_roman, sanscript.transliterate(rk_id_str, sanscript.DEVANAGARI, sanscript.IAST) + ".md")
md_file_Rk = MdFile(file_path=dest_path_Rk)
md_file_Rk.dump_to_file(metadata={"title": title_Rk}, content=rk_md, dry_run=dry_run)
md_file_Rk.set_filename_from_title(transliteration_source=sanscript.DEVANAGARI, dry_run=dry_run)
rk_map[rk_id_str] = md_file_Rk.file_path
suukta_md = ""
for rk_id in sorted(rk_map.keys()):
dest_path_Rk = rk_map[rk_id]
suukta_md = suukta_md + """
<div class="js_include" url="%s" newLevelForH1="2" includeTitle="false"> </div>
""" % dest_path_Rk.replace("/home/vvasuki/vvasuki-git", "").replace("static/", "")
import textwrap
suukta_md = """
## परिचयः
%s
## पाठः
%s
""" % ("\n ".join(meta_lines), suukta_md)
md_file_suukta = MdFile(file_path=dest_path_suukta)
md_file_suukta.dump_to_file(metadata={"title": "%s %s" % (suukta_id, suukta_title)}, content=textwrap.dedent(suukta_md), dry_run=dry_run)
md_file_suukta.set_filename_from_title(transliteration_source=sanscript.DEVANAGARI, dry_run=dry_run)
def tessBox(box_lst):
"""
box_lst : Input format : A list of lists, with each inner list being of the format:
0: Main box
1: voterIDBox
2: boxNumber
3: Pagenumber
"""
rows = []
st = time.time()
indicesWithErrors = []
resLst = picEater(box_lst)
for i, rex in enumerate(resLst):
op = rex[0].communicate()[0].decode('utf-8')
voter_id = rex[1].communicate()[0].decode('utf-8')
voter_id = dealWithID(voter_id)
boxNumber = rex[2].communicate()[0].decode('utf-8')
print("box number :" + str(i))
try:
res = op.split("\n")
res = [i for i in res if not i == '']
name_regional = (res[0].split("ः")[1] if len(res[0].split(":"))
== 1 else res[0].split(":")[1]).strip()
husband_or_father_regional = (res[1].split("ः")[1] if len(
res[0].split(":")) == 1 else res[1].split(":")[1]).strip()
house_number = ''
if (len(res[2].split(":")) > 1):
house_number = res[2].split(":")[1].strip()
elif (len(res[2].split("ः")) > 1):
house_number = res[2].split("ः")[1].strip()
house_number = dealWithHouses(house_number)
has_husband = 1 if (
res[1].split("ः")[0] if len(res[0].split(":")) == 1 else
res[1].split(":")[0]).split(" ")[0].strip() == "पति" else 0
age = re.search('[0-9]+', res[len(res) - 1]).group(0)
if (int(age) < 18):
raise ValueError()
intermediate = res[len(res) - 1].split(" ")
gender = 'F' if re.match('म', intermediate[
len(intermediate) -
1]) != None else 'M' #3 for other states, 4 for UK.
opName = transliterate(name_regional, sanscript.DEVANAGARI,
sanscript.ITRANS)
opHusband_or_father = transliterate(husband_or_father_regional,
sanscript.DEVANAGARI,
sanscript.ITRANS)
rows.append([
opName, age, gender, opHusband_or_father, has_husband,
house_number, voter_id, boxNumber, box_lst[i][3],
name_regional, husband_or_father_regional
])
except:
indicesWithErrors.append(i)
print("Had Errors.")
return rows, indicesWithErrors
def fix_lazy_anusvaara(self, data_in):
from indic_transliteration import sanscript
data_itrans = sanscript.transliterate(data=data_in,
_from=self.name,
_to=sanscript.ITRANS)
itrans_fixed = sanscript.SCHEMES[sanscript.ITRANS].fix_lazy_anusvaara(
data_in=data_itrans)
return sanscript.transliterate(data=itrans_fixed,
_from=sanscript.ITRANS,
_to=self.name)
def equal_dvng(self, w1, w2):
ma = "म्"
ma = ma.decode('utf-8')
if not (w1.endswith(ma) or w2.endswith(ma)):
return w1 == w2
w1 = sanscript.transliterate(w1, sanscript.DEVANAGARI, sanscript.SLP1)
w2 = sanscript.transliterate(w2, sanscript.DEVANAGARI, sanscript.SLP1)
w1 = re.sub(r'[mM]$', 'm', w1)
w2 = re.sub(r'[mM]$', 'm', w2)
return w1 == w2
def convert(text, intran, outtran):
"""Convert a text from intran to outtran transliteration."""
result = ''
if intran == outtran:
result = text
elif sys.version_info[0] < 3:
result = sanscript.transliterate(text, intran,
outtran).replace(u'|', u'.')
else:
result = sanscript.transliterate(text, intran,
outtran).replace('|', '.')
return result
def test_from_devanagari(test_case):
dev_string = test_case["dev_string"]
script = test_case["script"]
expected_text = test_case["text"]
# logging.debug("Converting %s, expecting %s in %s" % (dev_string, expected_text, script))
if script in SCRIPT_NAME_MAP.keys():
script = SCRIPT_NAME_MAP[script]
if script in "dev" or (script not in sanscript.SCHEMES.keys()):
logging.debug("Skipping over script - " + script)
return
result = sanscript.transliterate(dev_string, sanscript.DEVANAGARI, script)
result_dev = sanscript.transliterate(result, script, sanscript.DEVANAGARI)
assert expected_text == result or dev_string == result_dev, "Failed to convert to " + script + " from devanAgarI: got " + result + " instead of " + expected_text
def back_transliterate_word(self, word: str, predicted_lang_id=None):
word = word.lower()
if not self.en_dict.check(word) and predicted_lang_id != 1:
word = unicodedata.normalize('NFKC', word)
word = transliterate(word, sanscript.KOLKATA, sanscript.DEVANAGARI)
else:
transliterated_word = unicodedata.normalize('NFKC', word)
transliterated_word = transliterate(transliterated_word,
sanscript.KOLKATA,
sanscript.DEVANAGARI)
if unicodedata.normalize('NFKC',
transliterated_word) in self.hindi_words:
word = transliterated_word
return unicodedata.normalize('NFKC', word).upper()
def handle_data(self, data):
# print("Encountered some data :", self.current_tag, ': ', data)
final_data = data
if self.current_tag in ['l', 'pc']: # as of not not using this info
return
if self.current_tag in ['key1', 'key2'] and self.key_fromLang != self.key_toLang:
final_data = transliterate(
data, self.key_fromLang, self.key_toLang)
if self.current_tag == 's' and self.fromLang != self.toLang:
# sanskrit word
final_data = transliterate(
data, self.fromLang, self.toLang)
self.mark_down = self.mark_down + final_data
def get_headers(self): dhaatu_headers = [ self.dhaatu_naama, NIGHANTU_NAAMA + ' _' + transliterate(self.index, ITRANS, DEVANAGARI), transliterate(self.dhaatu_naama, DEVANAGARI, ITRANS), transliterate(NIGHANTU_NAAMA, DEVANAGARI, ITRANS) + ' _' + self.index, 'abdm _' + self.index ] for shloka in self.shlokas: shloka_headers = shloka.get_headers() if shloka_headers is None: continue dhaatu_headers.extend(shloka_headers) return dhaatu_headers
def unused_convert_sanskrit(text, inTran, outTran):
"""Return transliterated adjusted text."""
text1 = ''
counter = 0
# Remove '<srs/>'
text = text.replace('<srs/>', '')
# Change the s tag to span.
for i in re.split('<s>([^<]*)</s>', text):
if counter % 2 == 0:
text1 += i
else:
text1 += '<span class="s">' + sanscript.transliterate(
i, 'slp1', outTran) + '</span>'
counter += 1
# PE nesting of LB tag
text1 = text1.replace('<div n="1"/>', 'emsp;<div n="1"></div>')
text1 = text1.replace('<div n="2"/>', 'emsp;emsp;<div n="2"></div>')
text1 = text1.replace('<div n="3"/>', 'emsp;emsp;emsp;<div n="3"></div>')
text1 = text1.replace('<div n="4"/>',
'emsp;emsp;emsp;emsp;<div n="4"></div>')
text1 = text1.replace('<div n="5"/>',
'emsp;emsp;emsp;emsp;emsp;<div n="5"></div>')
text1 = re.sub('<div n="([^"]*)"/>', '<div n="\g<1>"></div>', text1)
text1 = text1.replace('<lb/>', '<br />')
# AP90 compounds and meanings break
text1 = text1.replace('<b>--', '<br /><b>--')
text1 = text1.replace('<span class="s">--', '<br /><span class="s">--')
# — breaks
text1 = text1.replace('— ', '<br />— ')
return text1
def dump_deep_text(url_text_id,
url_leaf_id_padding,
dir_path,
unit_info_file,
get_collapsible_content=False,
dry_run=False):
unit_data = text_data.get_subunit_data(unit_info_file, [])
for subunit_path in text_data.get_subunit_path_list(
json_file=unit_info_file, unit_path_list=[]):
relative_dir_path = "/".join(["%02d" % x for x in subunit_path[:-1]])
outfile_path = os.path.join(dir_path, relative_dir_path,
"%03d.md" % subunit_path[-1])
import urllib
item_url = "https://sa.wikisource.org/wiki/%s/%s" % (
urllib.parse.quote(url_text_id),
get_wiki_path(subunit_path=subunit_path,
unit_data=unit_data,
url_id_padding=url_leaf_id_padding))
title = sanscript.transliterate("%03d" % subunit_path[-1],
sanscript.SLP1, sanscript.DEVANAGARI)
logging.info("Getting %s to %s with title %s", item_url, outfile_path,
title)
if not dry_run:
dump_item(title=title,
outfile_path=outfile_path,
item_url=item_url,
get_collapsible_content=get_collapsible_content)
def devanagari(self, strict_io=True):
""" Return devanagari transcoding of self
"""
s = self.thing
if not strict_io:
s = normalization.denormalize(s)
return sanscript.transliterate(s, SLP1, DEVANAGARI)
def create_slp(code):
"""Create SLP file for a given dictionary code.
code is to be selected from dictcode.json.
"""
# ENSK -> ekaksharanamamala_sadhukalashagani
fullName = utils.code_to_dict(code)
# ekaksharanamamala, sadhukalashagani
bookName, author = fullName.split('_')
# Read the .txt file
filein = os.path.join('..', fullName, 'orig', bookName + '.txt')
fin = codecs.open(filein, 'r', 'utf-8')
data = fin.read()
fin.close()
# Convert the data to SLP1.
data = sanscript.transliterate(data, 'devanagari', 'slp1')
# Output directory
directory = os.path.join('..', fullName, 'slp')
# Create if the directory does not exist.
if not os.path.exists(directory):
os.mkdir(directory)
fileout = os.path.join(directory, bookName + '.txt')
# Create output file and save the SLP data in it.
fout = codecs.open(fileout, 'w', 'utf-8')
fout.write(data)
fout.close()
def hyperlink(bl_file_path, all_words: list):
if sys.platform != 'linux':
print(
'hyperlinking currently requires sed, which may exist only on unix compilant platforms'
)
sys.exit(1)
print('\nhyperlinking.... may take a minute')
considered_words = filter(lambda w: len(w) > 3 and '"' not in w, all_words)
sorted_words = sorted(considered_words, key=lambda w: len(w), reverse=True)
batch_size = 100
no_batches = math.ceil(len(sorted_words) / batch_size)
for bno in range(no_batches):
words = sorted_words[batch_size * bno:batch_size * (bno + 1)]
sed_command = 'sed -i \'' + '; '.join([
r'7~3s#\([,;\. -]\)\({w}\)\([,;-\. -]\)#\1<a href="{s}">{s}</a>\3#ig'
.format(w=w,
s=sanscript.transliterate(
w.lower(), scheme_map=HWS_XLITERATE_SCHEME_MAPS[0]))
for w in words
]) + '\' "{}"'.format(bl_file_path)
# print(sed_command)
os.system(sed_command)
def test_to_devanagari(test_case):
logging.debug(str(test_case))
dev_string = test_case["dev_string"]
script = test_case["script"]
text = test_case["text"]
result = sanscript.transliterate(text, script, sanscript.DEVANAGARI)
assert result == dev_string, "Failed to convert " + script + " to devanAgarI: got " + result + " instead of " + dev_string
def adjustlines(lines):
newlines = [
] # start with an empty list, in which the new lines will be put
# adjust each line in a python 'for loop'
for line in lines:
# we know, from the way read_lines was constructed, that each element
# in lines represents a line of the text file used as input ('filein')
# As such, it is a python 'str' (for 'string'). There is a buildin way to
# split strings into a list
x1 = line.split(":")
newline1 = x1[0]
x2 = line.split(":")
newline2 = x1[1]
x3 = newline2.split()
newline3 = x3[0]
newline3a = re.sub(r"([-~*‘’])|(\[a\])", "", newline3)
iastrev = remove_accent(newline3a)
newline4 = sanscript.transliterate(newline1, 'slp1', 'iast')
# We want to add the new line to our list of new lines.
# 'append' is the way to do that
newlines.append('%s' % "-" * 15)
newlines.append('orig = %s' % line)
newlines.append('slp1 = %s' % newline1)
newlines.append('rest = %s' % newline2)
newlines.append('iast = %s' % newline3a)
newlines.append('iastrev = %s' % iastrev)
newlines.append('slp-iast = %s' % newline4)
# we're done with the for loop, so we go back one level of indentation
# We need to return the newlines object that this function computed
return newlines
def test_syllables(test_case):
logging.debug(str(test_case))
pattern_lines = identify.to_pattern_lines(test_case["verse"].split("\n"))
id_result = identify.identifier.IdentifyFromPatternLines(pattern_lines)
assert 'exact' in id_result, id_result
exact_matches = [sanscript.transliterate(metre.lower(), _from=sanscript.IAST, _to=sanscript.DEVANAGARI) for metre in id_result['exact'].keys()]
assert exact_matches == test_case["exactMatches"], id_result
