Spaces:
Running
Running
File size: 3,835 Bytes
9984001 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 |
from polyglot.detect import Detector
from polyglot.text import Text
from difflib import Differ
from icecream import ic
from patch import *
from llama_index.core.node_parser import SentenceSplitter
def lang_detector(text):
min_chars = 5
if len(text) < min_chars:
return "Input text too short"
try:
detector = Detector(text).language
lang_info = str(detector)
code = re.search(r"name: (\w+)", lang_info).group(1)
return code
except Exception as e:
return f"ERROR：{str(e)}"
def tokenize(text):
# Use polyglot to tokenize the text
polyglot_text = Text(text)
words = polyglot_text.words
# Check if the text contains spaces
if ' ' in text:
# Create a list of words and spaces
tokens = []
for word in words:
tokens.append(word)
tokens.append(' ') # Add space after each word
return tokens[:-1] # Remove the last space
else:
return words
def diff_texts(text1, text2):
tokens1 = tokenize(text1)
tokens2 = tokenize(text2)
d = Differ()
diff_result = list(d.compare(tokens1, tokens2))
highlighted_text = []
for token in diff_result:
word = token[2:]
category = None
if token[0] == '+':
category = 'added'
elif token[0] == '-':
category = 'removed'
elif token[0] == '?':
continue # Ignore the hints line
highlighted_text.append((word, category))
return highlighted_text
#modified from src.translaation-agent.utils.tranlsate
def translator(
source_lang,
target_lang,
source_text,
country,
max_tokens=MAX_TOKENS_PER_CHUNK
):
"""Translate the source_text from source_lang to target_lang."""
num_tokens_in_text = num_tokens_in_string(source_text)
ic(num_tokens_in_text)
if num_tokens_in_text < max_tokens:
ic("Translating text as single chunk")
#Note: use yield from B() if put yield in function B()
init_translation = one_chunk_initial_translation(
source_lang, target_lang, source_text
)
reflection = one_chunk_reflect_on_translation(
source_lang, target_lang, source_text, init_translation, country
)
final_translation = one_chunk_improve_translation(
source_lang, target_lang, source_text, init_translation, reflection
)
return init_translation, reflection, final_translation
else:
ic("Translating text as multiple chunks")
token_size = calculate_chunk_size(
token_count=num_tokens_in_text, token_limit=max_tokens
)
ic(token_size)
#using sentence splitter
text_parser = SentenceSplitter(
chunk_size=token_size,
)
source_text_chunks = text_parser.split_text(source_text)
translation_1_chunks = multichunk_initial_translation(
source_lang, target_lang, source_text_chunks
)
init_translation = "".join(translation_1_chunks)
reflection_chunks = multichunk_reflect_on_translation(
source_lang,
target_lang,
source_text_chunks,
translation_1_chunks,
country,
)
reflection = "".join(reflection_chunks)
translation_2_chunks = multichunk_improve_translation(
source_lang,
target_lang,
source_text_chunks,
translation_1_chunks,
reflection_chunks,
)
final_translation = "".join(translation_2_chunks)
return init_translation, reflection, final_translation
|