Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -49,7 +49,7 @@ def get_face_locations(binary_data):
|
|
| 49 |
print(3)
|
| 50 |
return face_locations
|
| 51 |
|
| 52 |
-
def
|
| 53 |
# List to store page information
|
| 54 |
pages_info = []
|
| 55 |
|
|
@@ -57,13 +57,18 @@ def seperate_image_text_from_pdf(pdf_url):
|
|
| 57 |
response = requests.get(pdf_url)
|
| 58 |
|
| 59 |
if response.status_code == 200:
|
| 60 |
-
# Create a temporary
|
| 61 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
tmp_file.write(response.content)
|
| 63 |
-
tmp_file_path = tmp_file.name
|
| 64 |
|
| 65 |
# Open the PDF
|
| 66 |
-
pdf = fitz.open(
|
| 67 |
|
| 68 |
# Iterate through each page
|
| 69 |
for page_num in range(len(pdf)):
|
|
@@ -96,9 +101,9 @@ def seperate_image_text_from_pdf(pdf_url):
|
|
| 96 |
# Close the PDF
|
| 97 |
pdf.close()
|
| 98 |
|
| 99 |
-
# Clean up the temporary
|
| 100 |
-
|
| 101 |
-
os.
|
| 102 |
else:
|
| 103 |
print("Failed to fetch the PDF from the URL.")
|
| 104 |
|
|
@@ -148,16 +153,18 @@ def separate_audio_from_video(video_url):
|
|
| 148 |
# Extract audio
|
| 149 |
audio = video.audio
|
| 150 |
|
| 151 |
-
# Create a temporary
|
| 152 |
-
|
| 153 |
-
|
|
|
|
|
|
|
| 154 |
|
| 155 |
-
|
| 156 |
-
|
| 157 |
|
| 158 |
-
|
| 159 |
-
|
| 160 |
-
|
| 161 |
|
| 162 |
return audio_bytes
|
| 163 |
|
|
|
|
| 49 |
print(3)
|
| 50 |
return face_locations
|
| 51 |
|
| 52 |
+
def separate_image_text_from_pdf(pdf_url):
|
| 53 |
# List to store page information
|
| 54 |
pages_info = []
|
| 55 |
|
|
|
|
| 57 |
response = requests.get(pdf_url)
|
| 58 |
|
| 59 |
if response.status_code == 200:
|
| 60 |
+
# Create a temporary directory to store the PDF data
|
| 61 |
+
temp_dir = tempfile.mkdtemp()
|
| 62 |
+
|
| 63 |
+
# Define the temporary file path for the PDF
|
| 64 |
+
temp_pdf_path = os.path.join(temp_dir, "temp.pdf")
|
| 65 |
+
|
| 66 |
+
# Write the PDF data to the temporary file
|
| 67 |
+
with open(temp_pdf_path, "wb") as tmp_file:
|
| 68 |
tmp_file.write(response.content)
|
|
|
|
| 69 |
|
| 70 |
# Open the PDF
|
| 71 |
+
pdf = fitz.open(temp_pdf_path)
|
| 72 |
|
| 73 |
# Iterate through each page
|
| 74 |
for page_num in range(len(pdf)):
|
|
|
|
| 101 |
# Close the PDF
|
| 102 |
pdf.close()
|
| 103 |
|
| 104 |
+
# Clean up the temporary files
|
| 105 |
+
os.unlink(temp_pdf_path)
|
| 106 |
+
os.rmdir(temp_dir)
|
| 107 |
else:
|
| 108 |
print("Failed to fetch the PDF from the URL.")
|
| 109 |
|
|
|
|
| 153 |
# Extract audio
|
| 154 |
audio = video.audio
|
| 155 |
|
| 156 |
+
# Create a temporary directory to store temporary files
|
| 157 |
+
temp_dir = tempfile.mkdtemp()
|
| 158 |
+
|
| 159 |
+
# Define the temporary file path for the audio
|
| 160 |
+
temp_audio_filename = os.path.join(temp_dir, "audio.wav")
|
| 161 |
|
| 162 |
+
# Write the audio data to the temporary file
|
| 163 |
+
audio.write_audiofile(temp_audio_filename)
|
| 164 |
|
| 165 |
+
# Read the audio data from the temporary file as bytes
|
| 166 |
+
with open(temp_audio_filename, "rb") as f:
|
| 167 |
+
audio_bytes = f.read()
|
| 168 |
|
| 169 |
return audio_bytes
|
| 170 |
|