Spaces:
Runtime error
Runtime error
Commit
•
e1bd857
1
Parent(s):
f772a38
Update app.py
Browse files
app.py
CHANGED
@@ -1,67 +1,53 @@
|
|
1 |
import gradio as gr
|
2 |
-
import nltk
|
3 |
-
import PyPDF2
|
4 |
-
from nltk.corpus import stopwords
|
5 |
-
from nltk.tokenize import word_tokenize
|
6 |
-
import os
|
7 |
import tempfile
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
def clean_text(text):
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
#
|
23 |
-
|
24 |
-
|
25 |
-
return
|
26 |
-
|
27 |
def pdf_to_text(file):
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
# Clean the text
|
45 |
-
text = clean_text(text)
|
46 |
-
|
47 |
-
# Add the text to the full text
|
48 |
-
full_text += text + "\n\n----------\n\n" # Adding page break
|
49 |
-
|
50 |
-
# Write the full_text to a temp file
|
51 |
-
temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".txt")
|
52 |
-
with open(temp_file.name, 'w') as f:
|
53 |
f.write(full_text)
|
|
|
54 |
|
55 |
-
# Close the PDF file
|
56 |
-
pdf_file.close()
|
57 |
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
inputs=gr.inputs.File(type="file", label="Your PDF"),
|
63 |
-
outputs=gr.outputs.File(), # Changing the output to File type
|
64 |
-
title="PDF to TXT",
|
65 |
description="Convert your PDF files to clean text")
|
66 |
-
|
67 |
-
iface.launch(share=True)
|
|
|
1 |
import gradio as gr
|
|
|
|
|
|
|
|
|
|
|
2 |
import tempfile
|
3 |
+
import re
|
4 |
+
from PyPDF2 import PdfReader, PdfFileReader
|
5 |
+
import os
|
6 |
+
import spacy
|
7 |
+
import pytesseract
|
8 |
+
import pdf2image
|
9 |
+
import subprocess
|
10 |
+
from pdf2image.exceptions import (
|
11 |
+
PDFInfoNotInstalledError,
|
12 |
+
PDFPageCountError,
|
13 |
+
PDFSyntaxError
|
14 |
+
)
|
15 |
|
16 |
def clean_text(text):
|
17 |
+
nlp = spacy.load("en_core_web_sm", disable=["tagger", "parser", "ner", "textcat"])
|
18 |
+
text = re.sub(r'\n+', '\n', text)
|
19 |
+
text = re.sub(r'\s+', ' ', text)
|
20 |
+
return text.strip()
|
21 |
+
|
22 |
+
def image_to_latex(image):
|
23 |
+
image_path = "/tmp/equation.png" # Modify as needed
|
24 |
+
image.save(image_path)
|
25 |
+
result = subprocess.run(["pix2tex", image_path], capture_output=True, text=True)
|
26 |
+
return result.stdout
|
|
|
27 |
def pdf_to_text(file):
|
28 |
+
with open(file.name, 'rb') as f:
|
29 |
+
reader = PdfReader(f)
|
30 |
+
full_text = ''
|
31 |
+
for i, page in enumerate(reader.pages):
|
32 |
+
page_text = page.extract_text()
|
33 |
+
if page_text is None:
|
34 |
+
images = pdf2image.convert_from_path(file.name, first_page=i+1, last_page=i+2)
|
35 |
+
for image in images:
|
36 |
+
page_text = image_to_latex(image)
|
37 |
+
page_text = clean_text(page_text)
|
38 |
+
if len(page_text.split()) > 5:
|
39 |
+
page_text = "## Metadata: Page Number " + str(i+1) + "\n" + page_text
|
40 |
+
full_text += page_text + "\n\n"
|
41 |
+
base_name = os.path.splitext(os.path.basename(file.name))[0]
|
42 |
+
output_file_name = base_name + ".txt"
|
43 |
+
with open(output_file_name, 'w') as f:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
f.write(full_text)
|
45 |
+
return output_file_name
|
46 |
|
|
|
|
|
47 |
|
48 |
+
iface = gr.Interface(fn=pdf_to_text,
|
49 |
+
inputs=gr.inputs.File(label="Your PDF"),
|
50 |
+
outputs=gr.outputs.File(label="Download TXT"),
|
51 |
+
title="PDF to TXT",
|
|
|
|
|
|
|
52 |
description="Convert your PDF files to clean text")
|
53 |
+
iface.launch()
|
|