I'm working on a python project with Python(3.6) and Flask in which I have to return a text in Arabic. When I print the text in the console it works well but when I return it as response it's order changes to right-to-left.
Here's what I have tried:
from odoa import ODOA
import arabic_reshaper
from bidi.algorithm import get_display
from flask import Flask
import json
app = Flask(__name__)
app.config['JSON_AS_ASCII'] = False
@app.route('/', methods=['GET'])
def get_an_ayah():
odoa = ODOA()
surah = odoa.get_random_surah(lang='en')
text = surah.ayah.decode("utf-8")
reshaped_text = arabic_reshaper.reshape(text) # correct its shape
arabic_text = get_display(reshaped_text, base_dir='R')
print(arabic_text)
translation = str(surah.translate)
sound_file_url = str(surah.sound)
reference = str(str(surah.surah_number) + ':' + str(surah.ayah_number))
response_dict = {
'text': arabic_text,
'translation': translation,
'sound': sound_file_url,
'ref': reference
}
return response_dict
result for print(arabix_text
is:
and here's how its in response:
{
"ref": "94:2",
"sound": "https://raw.githubusercontent.com/semarketir/quranjson/master/source/audio/094/002.mp3",
"text": "ﻙﺭﺯﻭ ﻚﻨﻋ ﺎﻨﻌﺿﻭﻭ",
"translation": "And lift from you your burden."
}
how can I get the correct orientation for Arabic text?
Do you really need arabic_reshaper and python-bidi ?
Try the following code instead:
from odoa import ODOA
from flask import Flask
import json
app = Flask(__name__)
app.config['JSON_AS_ASCII'] = False
@app.route('/', methods=['GET'])
def get_an_ayah():
odoa = ODOA()
surah = odoa.get_random_surah(lang='en')
text = surah.ayah.decode("utf-8")
translation = str(surah.translate)
sound_file_url = str(surah.sound)
reference = str(str(surah.surah_number) + ':' + str(surah.ayah_number))
response_dict = {
'text': text,
'translation': translation,
'sound': sound_file_url,
'ref': reference
}
return response_dict
The output text will be with Arabic diacritics and Quranic annotation. If you want to remove them, use the following regex replacement:
import re
PATTERN = re.compile(
'['
'\u0610-\u061a'
'\u064b-\u065f'
'\u0670'
'\u06d6-\u06dc'
'\u06df-\u06e8'
'\u06ea-\u06ed'
'\u08d4-\u08e1'
'\u08d4-\u08ed'
'\u08e3-\u08ff'
']',
)
letters_only_text = re.sub(PATTERN, '', text)
See Arabic Unicode Chart and Arabic Extended-A Unicode Chart for more info about those replacements.