Spaces:
No application file
No application file
Update app.py
Browse files
app.py
CHANGED
|
@@ -492,63 +492,110 @@ def export_bookmarked_models():
|
|
| 492 |
return df.to_csv(index=False)
|
| 493 |
|
| 494 |
def display_pdf_preview(file_path, model_name):
|
| 495 |
-
"""Display PDF
|
| 496 |
import os
|
|
|
|
|
|
|
| 497 |
|
| 498 |
# Extract filename
|
| 499 |
normalized_path = file_path.replace('\\', '/')
|
| 500 |
pdf_filename = normalized_path.split('/')[-1]
|
| 501 |
|
| 502 |
-
#
|
| 503 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 504 |
|
| 505 |
# Control buttons
|
| 506 |
col1, col2, col3 = st.columns([2, 1, 1])
|
| 507 |
|
| 508 |
with col1:
|
| 509 |
-
st.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 510 |
|
| 511 |
with col2:
|
| 512 |
-
st.success("✅ PDF ready")
|
| 513 |
|
| 514 |
with col3:
|
| 515 |
if st.button("❌ Close Preview", use_container_width=True):
|
| 516 |
st.session_state[f'show_pdf_{model_name}'] = False
|
| 517 |
st.rerun()
|
| 518 |
|
| 519 |
-
# Display PDF - NO GREEN BORDER
|
| 520 |
st.markdown("---")
|
| 521 |
|
| 522 |
-
#
|
| 523 |
-
|
| 524 |
-
|
| 525 |
-
|
| 526 |
-
|
| 527 |
-
|
| 528 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 529 |
"""
|
| 530 |
|
| 531 |
-
|
| 532 |
-
|
| 533 |
-
#
|
| 534 |
-
|
| 535 |
-
|
| 536 |
-
|
| 537 |
-
|
| 538 |
-
|
| 539 |
-
|
| 540 |
-
|
| 541 |
-
|
| 542 |
-
# pdf_viewer = f"""
|
| 543 |
-
# <iframe
|
| 544 |
-
# src="https://docs.google.com/viewer?url={pdf_url}&embedded=true"
|
| 545 |
-
# width="100%"
|
| 546 |
-
# height="900px"
|
| 547 |
-
# frameborder="0">
|
| 548 |
-
# </iframe>
|
| 549 |
-
# """
|
| 550 |
-
|
| 551 |
-
st.markdown(pdf_viewer, unsafe_allow_html=True)
|
| 552 |
|
| 553 |
def get_high_accuracy_models(limit=8):
|
| 554 |
"""Get models with highest extraction quality"""
|
|
|
|
| 492 |
return df.to_csv(index=False)
|
| 493 |
|
| 494 |
def display_pdf_preview(file_path, model_name):
|
| 495 |
+
"""Display PDF - works on Hugging Face with Git LFS"""
|
| 496 |
import os
|
| 497 |
+
import requests
|
| 498 |
+
import base64
|
| 499 |
|
| 500 |
# Extract filename
|
| 501 |
normalized_path = file_path.replace('\\', '/')
|
| 502 |
pdf_filename = normalized_path.split('/')[-1]
|
| 503 |
|
| 504 |
+
# Check if running locally or on Hugging Face
|
| 505 |
+
is_local = not os.environ.get('SPACE_ID')
|
| 506 |
+
|
| 507 |
+
if is_local:
|
| 508 |
+
# Local file system access
|
| 509 |
+
pdf_path = os.path.join('pdfs', pdf_filename)
|
| 510 |
+
if os.path.exists(pdf_path):
|
| 511 |
+
with open(pdf_path, 'rb') as f:
|
| 512 |
+
pdf_data = f.read()
|
| 513 |
+
file_size_mb = len(pdf_data) / (1024 * 1024)
|
| 514 |
+
else:
|
| 515 |
+
st.error(f"PDF not found locally: {pdf_path}")
|
| 516 |
+
return
|
| 517 |
+
else:
|
| 518 |
+
# Hugging Face - need to download from URL
|
| 519 |
+
pdf_url = f"https://huggingface.co/spaces/TurboAir/TurboAirViewer/resolve/main/pdfs/{pdf_filename}"
|
| 520 |
+
|
| 521 |
+
try:
|
| 522 |
+
with st.spinner("Loading PDF..."):
|
| 523 |
+
response = requests.get(pdf_url, timeout=30)
|
| 524 |
+
response.raise_for_status()
|
| 525 |
+
pdf_data = response.content
|
| 526 |
+
file_size_mb = len(pdf_data) / (1024 * 1024)
|
| 527 |
+
|
| 528 |
+
# Verify we got actual PDF data, not Git LFS pointer
|
| 529 |
+
if len(pdf_data) < 1000:
|
| 530 |
+
st.error("PDF file is too small - might be Git LFS pointer")
|
| 531 |
+
# Use external viewer as fallback
|
| 532 |
+
st.markdown(f"[📄 Open PDF in new tab]({pdf_url})")
|
| 533 |
+
return
|
| 534 |
+
|
| 535 |
+
except Exception as e:
|
| 536 |
+
st.error(f"Error loading PDF: {str(e)}")
|
| 537 |
+
# Fallback to external link
|
| 538 |
+
st.markdown(f"[📄 Open PDF in new tab]({pdf_url})")
|
| 539 |
+
return
|
| 540 |
|
| 541 |
# Control buttons
|
| 542 |
col1, col2, col3 = st.columns([2, 1, 1])
|
| 543 |
|
| 544 |
with col1:
|
| 545 |
+
st.download_button(
|
| 546 |
+
"📥 Download PDF",
|
| 547 |
+
data=pdf_data,
|
| 548 |
+
file_name=pdf_filename,
|
| 549 |
+
mime="application/pdf",
|
| 550 |
+
use_container_width=True,
|
| 551 |
+
type="primary"
|
| 552 |
+
)
|
| 553 |
|
| 554 |
with col2:
|
| 555 |
+
st.success(f"✅ PDF ready ({file_size_mb:.1f} MB)")
|
| 556 |
|
| 557 |
with col3:
|
| 558 |
if st.button("❌ Close Preview", use_container_width=True):
|
| 559 |
st.session_state[f'show_pdf_{model_name}'] = False
|
| 560 |
st.rerun()
|
| 561 |
|
|
|
|
| 562 |
st.markdown("---")
|
| 563 |
|
| 564 |
+
# Display PDF using base64 encoding (works everywhere)
|
| 565 |
+
base64_pdf = base64.b64encode(pdf_data).decode('utf-8')
|
| 566 |
+
|
| 567 |
+
# Try multiple display methods
|
| 568 |
+
pdf_display = f"""
|
| 569 |
+
<style>
|
| 570 |
+
.pdf-viewer {{
|
| 571 |
+
width: 100%;
|
| 572 |
+
height: 900px;
|
| 573 |
+
border: none;
|
| 574 |
+
}}
|
| 575 |
+
</style>
|
| 576 |
+
|
| 577 |
+
<!-- Method 1: Object with embed fallback -->
|
| 578 |
+
<object data="data:application/pdf;base64,{base64_pdf}"
|
| 579 |
+
type="application/pdf"
|
| 580 |
+
class="pdf-viewer">
|
| 581 |
+
<embed src="data:application/pdf;base64,{base64_pdf}"
|
| 582 |
+
type="application/pdf"
|
| 583 |
+
class="pdf-viewer" />
|
| 584 |
+
<p>PDF preview not available. Please use the download button above.</p>
|
| 585 |
+
</object>
|
| 586 |
"""
|
| 587 |
|
| 588 |
+
st.markdown(pdf_display, unsafe_allow_html=True)
|
| 589 |
+
|
| 590 |
+
# If base64 doesn't work, show alternative
|
| 591 |
+
st.markdown("---")
|
| 592 |
+
st.info("💡 If PDF doesn't display above, you can:")
|
| 593 |
+
col1, col2 = st.columns(2)
|
| 594 |
+
with col1:
|
| 595 |
+
st.markdown("• Use the Download button above")
|
| 596 |
+
with col2:
|
| 597 |
+
if not is_local:
|
| 598 |
+
st.markdown(f"• [Open in new tab ↗]({pdf_url})")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 599 |
|
| 600 |
def get_high_accuracy_models(limit=8):
|
| 601 |
"""Get models with highest extraction quality"""
|