def alif_count_verse(quran_dict, sura, verse, assumptions):
mod = "strong"
try:
trans = transString(quran_dict[sura][verse]["arabic"])
# print(sura, verse, trans)
total = sum([1 for letter in trans if letter in assumptions])
letters = "".join(
[letter for letter in trans if letter in assumptions])
return {
"verse":
Markup(quran_dict[sura][verse]["arabic"].replace(
"ا", "<" + mod + ">ا</" + mod + ">")),
"count":
total,
"tgv":
calc_val(letters, "tgv")
}
except:
{
"verse":
Markup(
"<b>Error</b> in function 'alif_count_sura': Invalid sura and verse number "
),
"count":
"Error in alif_count_verse function"
}
def index():
global verse_obj, alif_count, search_term
alif_count = ""
# test=['a', 'b']
if request.method == "GET":
# search_term = '_undefined_' if len(search_term) < 1 else search_term
return render_template("main_page.html",
comments=Word.query.all(),
verses=verse_obj,
alif_count=alif_count,
search_term=search_term)
else:
execute_this = request.form["submit"]
logging.error(request.form)
if execute_this == "Clear":
n = Word.query.delete()
db.session.commit()
print(n)
elif execute_this == "Calculate TGV":
word = request.form["contents"]
addition = calc_val(transString(word), "tgv")
string = "TGV of " + word + " is: " + str(addition)
comment = Word(content=string)
db.session.add(comment)
db.session.commit()
elif execute_this == "Count alif":
req = request.form["contents"].strip()
req = list(map(int, re.split("\D+", req)))
if len(req) == 1:
sura = req[0]
alif_count = alif_count_sura(quran_dict, sura, ["A"])
else:
try:
sura, verse = req[0], req[1]
alif_count = alif_count_verse(quran_dict, sura, verse,
["A"])
except:
alif_count = "Invalid sura-verse input"
elif execute_this == "Get verse":
req = request.form["contents"].strip()
if re.match("\d+\D+\d+\D*", req):
verse_obj_pre = query_verses_number(req, db)
search_term = 'N/A'
else:
search_term = remove_diacritics(req) if detect_arabic(
req) else req
verse_obj_pre = query_verses_text(req, db)
verse_obj = []
for verse_dict in verse_obj_pre:
temp = verse_dict.copy()
temp['ar'] = remove_diacritics(verse_dict['ar'])
verse_obj.append(temp)
return redirect(url_for('index'))
def tgv_matching():
global verse_obj, alif_count
alif_count = ""
# test=['a', 'b']
if request.method == "GET":
return render_template("tgv_matching.html",
comments=Word.query.all(),
verses=verse_obj,
alif_count=alif_count,
search_package={})
else:
execute_this = request.form["submit"]
logging.error(request.form)
if execute_this == "Clear":
n = Word.query.delete()
db.session.commit()
print(n)
elif execute_this == "Calculate TGV":
word = request.form["contents"]
addition = calc_val(transString(word), "tgv")
string = "TGV of " + word + " is: " + str(addition)
comment = Word(content=string)
db.session.add(comment)
db.session.commit()
elif execute_this == "Find TGV pairs":
req = request.form["contents"].strip()
if re.fullmatch('\d*', req):
tgv = int(req)
pairs = [ngram for ngram in all_ngrams if ngram['tgv'] == tgv]
pairs_dict = build_tgv_match_dict(pairs)
search_package = {'term': tgv, 'type': 'number'}
else:
word = req
tgv = calc_val(transString(word), "tgv")
pairs = [ngram for ngram in all_ngrams if ngram['tgv'] == tgv]
pairs_dict = build_tgv_match_dict(pairs)
search_package = {'term': word, 'type': 'word'}
return render_template("tgv_matching.html",
comments=Word.query.all(),
pairs=pairs_dict,
search_package=search_package,
tgv=tgv)
return redirect(url_for('tgv_matching'))
def get_ngrams_verse(sura_nbr, verse_nbr, verse):
words = nltk.word_tokenize(verse)
my_bigrams = [' '.join(x) for x in nltk.bigrams(words)]
my_trigrams = [' '.join(x) for x in nltk.trigrams(words)]
all_grams = words + my_bigrams + my_trigrams
return [{
"gram": w,
"tgv": calc_val(transString(w), type_v="tgv"),
"sura_nbr": sura_nbr,
"verse_nbr": verse_nbr
} for w in all_grams]
def verse2dict(verse):
'''function used to read in Arabic Quran text. Use regular expression
pattern matching to extract sura and verse number, and Arabic and English text
from Quran text file'''
rd = {}
rd["nSura"] = re.search("\d{1,3}\|", verse).group(0)[0:-1]
rd["nVerse"] = re.search("\|\d{1,3}\|", verse).group(0)[1:-1]
rd["ar"] = verse.split("|")[2]
rd["eng"] = verse.split("|")[3]
rd["translit"] = transString(rd["ar"])
return rd
def query_verses_text(req, db):
if req=="":
return [{}]
req=req.strip()
if detect_arabic(req):
search = '%' + transString(req) + '%'
textual = text("select nSura, nVerse, ar, eng, translit from q_tbl " +
"where translit like :x1")
else:
search = '%' + req + '%'
textual = text("select nSura, nVerse, ar, eng, translit from q_tbl " +
"where eng like :x1")
print("req is: ", req)
print("search is: ", search)
rv = db.session.execute(textual, {"x1": search}).fetchall()
return load_query_into_dict(rv)
db.session.commit()
qdf = solver.quran_as_df(quran_file, sura_order_table)
if quran_test: # if testing mode, only work with first few rows of data
n = 111
else:
n = 50100
i = 0
for verse in qdf[0:n].itertuples():
temp_verse = Verse(nSura=verse.sura,
nVerse=verse.verse,
ar=verse.arabic,
eng=verse.english,
translit=transString(verse.arabic),
seq_order=verse.seq_index,
chron_order=verse.chron_index)
db.session.add(temp_verse)
i += 1
db.session.commit()
# Build Quran dictionary and TGV dictionary
quran_dict = scrape_quran_into_dict(quran_file)
print("assembled quran_dict")
all_ngrams = get_ngrams_quran(quran_dict)
print("assembled ngrams phase 1...")
ngrams_sub = all_ngrams[0:6]
#tgv_dict = build_tgv_dict(ngrams_sub)
from pandasql import sqldf
from funcs import calc_val, transString
# Read in datafiles
ldf = pd.read_csv('letter_df.csv')
# Create summary by letter
q = '''select sura, letters as letter, count(*) as letter_count
from ldf
group by sura, letters
order by sura, letters;
'''
pysqldf = lambda q: sqldf(q, globals())
ldfsum = pysqldf(q)
ldfsum['tgv'] = ldfsum['letter'].apply(
lambda x: calc_val(transString(x), 'tgv'))
ldfsum = ldfsum.query('tgv > 0')
# Create TGV plotting dataset
q = '''select min(letter) as letter, tgv, sum(letter_count) as count
from ldfsum
group by tgv
order by tgv;
'''
tgv_plotting_df = pysqldf(q)
letters = [transString(l) for l in tgv_plotting_df['letter'].tolist()]
# Create plot
sns.lmplot(x="tgv",
y="count",
data=tgv_plotting_df,
