def process_files_law(basename, name):
filtertext(app.config['UPLOAD_FOLDER'] + '/' + basename + '_hin.txt',
app.config['UPLOAD_FOLDER'] + '/' + basename + '_hin_filtered.txt')
filtertext(app.config['UPLOAD_FOLDER'] + '/' + basename + '_eng.txt',
app.config['UPLOAD_FOLDER'] + '/' + basename + '_eng_filtered.txt')
processhindi(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_hin_filtered.txt')
processenglish(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_eng_filtered.txt')
translatewithgoogle(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_hin_filtered.txt', app.config['UPLOAD_FOLDER'] +
'/' + basename + '_eng_tran.txt')
os.system(
'./helpers/bleualign.py -s ' + os.getcwd() + '/upload/' + basename + '_hin_filtered' + '.txt' + ' -t ' + os.getcwd() + '/upload/' + basename +
'_eng_filtered' + '.txt' + ' --srctotarget ' + os.getcwd() + '/upload/' + basename + '_eng_tran' + '.txt' + ' -o ' + os.getcwd() + '/upload/' + basename + '_output')
english_res = []
hindi_res = []
english_points = []
english_points_words = []
hindi_points = []
hindi_points_words = []
f_eng = open(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_output-t', 'r')
for f in f_eng:
english_res.append(f)
point = fetchwordsfromsentence(f, basename)
english_points.append(point['avg'])
english_points_words.append(point['values'])
f_eng.close()
f_hin = open(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_output-s', 'r')
for f in f_hin:
hindi_res.append(f)
point = fetchwordsfromsentence(f, basename)
hindi_points.append(point['avg'])
hindi_points_words.append(point['values'])
f_hin.close()
data = {'hindi': hindi_res, 'english': english_res,
'english_scores': english_points, 'hindi_scores': hindi_points}
sentences = []
for i in range(0, len(hindi_res)):
sentence = Sentence(status=STATUS_PENDING, alignment_accuracy=english_res[i].split(':::::')[1], basename=name,
source=hindi_res[i], target=english_res[i].split(':::::')[0],
source_ocr_words=hindi_points_words[i], source_ocr=str(hindi_points[i]),
target_ocr_words=english_points_words[i], target_ocr=str(english_points[i]))
sentences.append(sentence)
# sentence.save()
Sentence.objects.insert(sentences)
for f in glob.glob(app.config['UPLOAD_FOLDER'] + '/' + basename + '*'):
os.remove(f)
res = CustomResponse(Status.SUCCESS.value, data)
# corpus = Corpus.objects(basename=basename)
# corpus.update(set__status=STATUS_PROCESSED,
# set__no_of_sentences=len(hindi_res))
return res.getres()
def getfiledata():
pool = mp.Pool(mp.cpu_count())
basename = str(int(time.time()))
current_time = datetime.now().strftime("%m/%d/%Y, %H:%M:%S")
f = request.files['file']
filepath = os.path.join(
app.config['UPLOAD_FOLDER'], basename + '.pdf')
# translationProcess = TranslationProcess(
# status=STATUS_PROCESSING, name=f.filename, created_on=current_time, basename=basename)
# translationProcess.save()
f.save(filepath)
pool.apply_async(converttoimage, args=(
filepath, app.config['UPLOAD_FOLDER'], basename, '_eng'), callback=capturetext)
pool.close()
pool.join()
filtertext(app.config['UPLOAD_FOLDER'] + '/' + basename + '_eng.txt',
app.config['UPLOAD_FOLDER'] + '/' + basename + '_eng_filtered.txt')
processenglish(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_eng_filtered.txt')
# translatewithanuvadaeng(app.config['UPLOAD_FOLDER'] +
# '/'+basename+'_hin_filtered.txt', app.config['UPLOAD_FOLDER'] +
# '/'+basename+'_eng_tran.txt')
# f_eng = open(app.config['UPLOAD_FOLDER']+'/' +
# basename + '_eng_filtered.txt', 'r')
english_res = []
# hindi_res = []
# for f in f_eng:
# english_res.append(f)
# f_eng.close()
f_eng = open(app.config['UPLOAD_FOLDER'] + '/' +
basename + '_eng_filtered.txt', 'r')
for f in f_eng:
english_res.append(f)
f_eng.close()
data = {'english': english_res}
# translations = []
# for i in range(0, len(hindi_res)):
# translation = Translation(basename=str(
# basename), source=hindi_res[i], target=english_res[i])
# translations.append(translation)
# Translation.objects.insert(translations)
res = CustomResponse(Status.SUCCESS.value, data)
result = flask.send_file(os.path.join('upload/', basename + '_eng_filtered.txt'), as_attachment=True)
result.headers["x-suggested-filename"] = basename + '.txt'
# translationProcess = TranslationProcess.objects(basename=basename)
# translationProcess.update(set__status=STATUS_PROCESSED)
return result
def translate():
pool = mp.Pool(mp.cpu_count())
basename = str(int(time.time()))
current_time = datetime.now().strftime("%m/%d/%Y, %H:%M:%S")
f = request.files['file']
filepath = os.path.join(
app.config['UPLOAD_FOLDER'], basename + '.pdf')
translationProcess = TranslationProcess(
status=STATUS_PROCESSING, name=f.filename, created_on=current_time, basename=basename)
translationProcess.save()
f.save(filepath)
pool.apply_async(converttoimage, args=(
filepath, app.config['UPLOAD_FOLDER'], basename, '_hin'), callback=capturetext)
pool.close()
pool.join()
filtertext(app.config['UPLOAD_FOLDER'] + '/' + basename + '_hin.txt',
app.config['UPLOAD_FOLDER'] + '/' + basename + '_hin_filtered.txt')
processenglish(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_hin_filtered.txt')
translatewithanuvadaeng(app.config['UPLOAD_FOLDER'] +
'/' + basename + '_hin_filtered.txt', app.config['UPLOAD_FOLDER'] +
'/' + basename + '_eng_tran.txt')
f_eng = open(app.config['UPLOAD_FOLDER'] + '/' +
basename + '_eng_tran.txt', 'r')
english_res = []
hindi_res = []
for f in f_eng:
english_res.append(f)
f_eng.close()
f_hin = open(app.config['UPLOAD_FOLDER'] + '/' +
basename + '_hin_filtered.txt', 'r')
for f in f_hin:
hindi_res.append(f)
f_hin.close()
data = {'hindi': hindi_res, 'english': english_res}
translations = []
for i in range(0, len(hindi_res)):
translation = Translation(basename=str(
basename), source=hindi_res[i], target=english_res[i])
translations.append(translation)
Translation.objects.insert(translations)
for f in glob.glob(app.config['UPLOAD_FOLDER'] + '/' + basename + '*'):
os.remove(f)
res = CustomResponse(Status.SUCCESS.value, data)
translationProcess = TranslationProcess.objects(basename=basename)
translationProcess.update(set__status=STATUS_PROCESSED)
return res.getres()
def translateFile():
pool = mp.Pool(mp.cpu_count())
basename = str(int(time.time()))
current_time = datetime.now().strftime("%m/%d/%Y, %H:%M:%S")
f = request.files['file']
filepath = os.path.join(
app.config['UPLOAD_FOLDER'], basename + '.pdf')
translationProcess = TranslationProcess(
status=STATUS_PROCESSING, name=f.filename, created_on=current_time, basename=basename)
translationProcess.save()
f.save(filepath)
pool.apply_async(converttoimage, args=(
filepath, app.config['UPLOAD_FOLDER'], basename, '_hin'), callback=capturetext)
pool.close()
pool.join()
filtertext(app.config['UPLOAD_FOLDER'] + '/'+basename+'_hin.txt',
app.config['UPLOAD_FOLDER'] + '/'+basename+'_hin_filtered.txt')
processenglish(app.config['UPLOAD_FOLDER'] +
'/'+basename+'_hin_filtered.txt')
translatewithanuvadaeng(app.config['UPLOAD_FOLDER'] +
'/'+basename+'_hin_filtered.txt', app.config['UPLOAD_FOLDER'] +
'/'+basename+'_eng_tran.txt')
f_eng = open(app.config['UPLOAD_FOLDER']+'/' +
basename + '_eng_tran.txt', 'r')
english_res = []
hindi_res = []
for f in f_eng:
english_res.append(f)
f_eng.close()
f_hin = open(app.config['UPLOAD_FOLDER']+'/' +
basename + '_hin_filtered.txt', 'r')
index = 0
previousY = 0
previousX = 0
previousH = 0
previousP = ''
text_y = {}
text_x = 0
for f in f_hin:
hindi_res.append(f)
print(f)
point = fetchwordhocrfromsentence(f, basename)
english = english_res[index]
words = english.split(' ')
wordIndex = 0
for word in words:
try:
if point['values'] is not None and point['values'][wordIndex] is not None and point['values'][wordIndex]['height'] is not None:
previousY = point['values'][wordIndex]['left']
previousX = point['values'][wordIndex]['top']
previousH = point['values'][wordIndex]['height']
try:
if text_y[point['values'][wordIndex]['imagepath']] is None:
text_y[point['values'][wordIndex]['imagepath']] = 200
except Exception as e:
text_y[point['values'][wordIndex]['imagepath']] = 200
(text_x, vertical) = puttext(point['values'][wordIndex]['height'],200,text_y[point['values'][wordIndex]['imagepath']],english,point['values'][wordIndex]['imagepath'])
text_y[point['values'][wordIndex]['imagepath']] = vertical
# else:
# (text_x, text_y) = puttext(point['values'][wordIndex]['height'],point['values'][wordIndex]['left'],point['values'][wordIndex]['top'],english,point['values'][wordIndex]['imagepath'])
previousP = point['values'][wordIndex]['imagepath']
break
except Exception as e:
previousY = previousY + 200
# puttext(previousH,previousY,previousX,word,previousP)
wordIndex = wordIndex + 1
# puttext(point['values'][wordIndex]['left'],point['values'][wordIndex]['top'],word,point['values'][wordIndex]['imagepath'])
index = index + 1
f_hin.close()
data = {'hindi': hindi_res, 'english': english_res}
translations = []
for i in range(0, len(hindi_res)):
translation = Translation(basename=str(
basename), source=hindi_res[i], target=english_res[i])
translations.append(translation)
Translation.objects.insert(translations)
# for f in glob.glob(app.config['UPLOAD_FOLDER']+'/'+basename+'*'):
# os.remove(f)
res = CustomResponse(Status.SUCCESS.value, data)
translationProcess = TranslationProcess.objects(basename=basename)
translationProcess.update(set__status=STATUS_PROCESSED)
return res.getres()