Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -31,6 +31,7 @@ from xml.etree import ElementTree as ET
|
|
31 |
|
32 |
import openai
|
33 |
from openai import OpenAI
|
|
|
34 |
|
35 |
|
36 |
|
@@ -63,6 +64,16 @@ if st.button("Clear Session"):
|
|
63 |
st.session_state.messages = []
|
64 |
|
65 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
|
67 |
def download_and_save_abstract(url, title):
|
68 |
response = requests.get(url)
|
@@ -118,10 +129,18 @@ def parse_to_markdown(text):
|
|
118 |
return text
|
119 |
|
120 |
def load_file(file_name):
|
121 |
-
|
122 |
-
|
123 |
-
|
124 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
125 |
|
126 |
def extract_urls_old(text):
|
127 |
try:
|
@@ -543,10 +562,9 @@ def compare_and_delete_files(files):
|
|
543 |
def get_file_size(file_path):
|
544 |
return os.path.getsize(file_path)
|
545 |
|
546 |
-
|
547 |
def FileSidebar():
|
548 |
# File Sidebar for files 🌐View, 📂Open, ▶️Run, and 🗑Delete per file
|
549 |
-
all_files = glob.glob("*.md") + glob.glob("*_abstract.html") + glob.glob("
|
550 |
all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 10] # exclude files with short names
|
551 |
all_files.sort(key=lambda x: (os.path.splitext(x)[1], x), reverse=True) # sort by filename length which puts similar prompts together
|
552 |
|
@@ -561,9 +579,9 @@ def FileSidebar():
|
|
561 |
if st.button("⬇️ Download"):
|
562 |
zip_file = create_zip_of_files(all_files)
|
563 |
st.sidebar.markdown(get_zip_download_link(zip_file), unsafe_allow_html=True)
|
564 |
-
file_contents=''
|
565 |
-
file_name=''
|
566 |
-
next_action=''
|
567 |
|
568 |
# Add files 🌐View, 📂Open, ▶️Run, and 🗑Delete per file
|
569 |
for file in all_files:
|
@@ -573,17 +591,27 @@ def FileSidebar():
|
|
573 |
col1, col2, col3, col4, col5 = st.sidebar.columns([1,6,1,1,1]) # adjust the ratio as needed
|
574 |
with col1:
|
575 |
if st.button("🌐", key=f"view_{timestamp}_{file}"): # view emoji button
|
576 |
-
|
577 |
-
|
578 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
579 |
st.session_state['next_action'] = next_action
|
580 |
with col2:
|
581 |
st.markdown(get_table_download_link(file), unsafe_allow_html=True)
|
582 |
with col3:
|
583 |
if st.button("📂", key=f"open_{timestamp}_{file}"): # open emoji button
|
584 |
file_contents = load_file(file)
|
585 |
-
file_name=file
|
586 |
-
next_action='open'
|
587 |
st.session_state['lastfilename'] = file
|
588 |
st.session_state['filename'] = file
|
589 |
st.session_state['filetext'] = file_contents
|
@@ -591,15 +619,15 @@ def FileSidebar():
|
|
591 |
with col4:
|
592 |
if st.button("▶️", key=f"read_{timestamp}_{file}"): # search emoji button
|
593 |
file_contents = load_file(file)
|
594 |
-
file_name=file
|
595 |
-
next_action='search'
|
596 |
st.session_state['next_action'] = next_action
|
597 |
with col5:
|
598 |
if st.button("🗑", key=f"delete_{timestamp}_{file}"):
|
599 |
os.remove(file)
|
600 |
-
file_name=file
|
601 |
st.rerun()
|
602 |
-
next_action='delete'
|
603 |
st.session_state['next_action'] = next_action
|
604 |
|
605 |
|
|
|
31 |
|
32 |
import openai
|
33 |
from openai import OpenAI
|
34 |
+
import fitz # pymupdf
|
35 |
|
36 |
|
37 |
|
|
|
64 |
st.session_state.messages = []
|
65 |
|
66 |
|
67 |
+
def convert_pdf_to_markdown(pdf_path):
|
68 |
+
try:
|
69 |
+
doc = fitz.open(pdf_path)
|
70 |
+
markdown_text = ""
|
71 |
+
for page in doc:
|
72 |
+
markdown_text += page.get_text("markdown")
|
73 |
+
doc.close()
|
74 |
+
return markdown_text
|
75 |
+
except Exception as e:
|
76 |
+
return f"Error converting PDF to markdown: {str(e)}"
|
77 |
|
78 |
def download_and_save_abstract(url, title):
|
79 |
response = requests.get(url)
|
|
|
129 |
return text
|
130 |
|
131 |
def load_file(file_name):
|
132 |
+
file_extension = os.path.splitext(file_name)[1].lower()
|
133 |
+
try:
|
134 |
+
if file_extension == '.pdf':
|
135 |
+
return convert_pdf_to_markdown(file_name)
|
136 |
+
elif file_extension == '.html':
|
137 |
+
with open(file_name, 'r', encoding='utf-8') as file:
|
138 |
+
return file.read()
|
139 |
+
else: # For .md and other text files
|
140 |
+
with open(file_name, 'r', encoding='utf-8') as file:
|
141 |
+
return file.read()
|
142 |
+
except Exception as e:
|
143 |
+
return f"Error loading file {file_name}: {str(e)}"
|
144 |
|
145 |
def extract_urls_old(text):
|
146 |
try:
|
|
|
562 |
def get_file_size(file_path):
|
563 |
return os.path.getsize(file_path)
|
564 |
|
|
|
565 |
def FileSidebar():
|
566 |
# File Sidebar for files 🌐View, 📂Open, ▶️Run, and 🗑Delete per file
|
567 |
+
all_files = glob.glob("*.md") + glob.glob("*_abstract.html") + glob.glob("*_abstract.pdf")
|
568 |
all_files = [file for file in all_files if len(os.path.splitext(file)[0]) >= 10] # exclude files with short names
|
569 |
all_files.sort(key=lambda x: (os.path.splitext(x)[1], x), reverse=True) # sort by filename length which puts similar prompts together
|
570 |
|
|
|
579 |
if st.button("⬇️ Download"):
|
580 |
zip_file = create_zip_of_files(all_files)
|
581 |
st.sidebar.markdown(get_zip_download_link(zip_file), unsafe_allow_html=True)
|
582 |
+
file_contents = ''
|
583 |
+
file_name = ''
|
584 |
+
next_action = ''
|
585 |
|
586 |
# Add files 🌐View, 📂Open, ▶️Run, and 🗑Delete per file
|
587 |
for file in all_files:
|
|
|
591 |
col1, col2, col3, col4, col5 = st.sidebar.columns([1,6,1,1,1]) # adjust the ratio as needed
|
592 |
with col1:
|
593 |
if st.button("🌐", key=f"view_{timestamp}_{file}"): # view emoji button
|
594 |
+
file_extension = os.path.splitext(file)[1].lower()
|
595 |
+
if file_extension == '.pdf':
|
596 |
+
markdown_content = convert_pdf_to_markdown(file)
|
597 |
+
st.markdown(markdown_content)
|
598 |
+
elif file_extension == '.html':
|
599 |
+
with open(file, 'r', encoding='utf-8') as f:
|
600 |
+
html_content = f.read()
|
601 |
+
st.components.v1.html(html_content, height=600, scrolling=True)
|
602 |
+
else:
|
603 |
+
file_contents = load_file(file)
|
604 |
+
st.markdown(file_contents)
|
605 |
+
file_name = file
|
606 |
+
next_action = 'md'
|
607 |
st.session_state['next_action'] = next_action
|
608 |
with col2:
|
609 |
st.markdown(get_table_download_link(file), unsafe_allow_html=True)
|
610 |
with col3:
|
611 |
if st.button("📂", key=f"open_{timestamp}_{file}"): # open emoji button
|
612 |
file_contents = load_file(file)
|
613 |
+
file_name = file
|
614 |
+
next_action = 'open'
|
615 |
st.session_state['lastfilename'] = file
|
616 |
st.session_state['filename'] = file
|
617 |
st.session_state['filetext'] = file_contents
|
|
|
619 |
with col4:
|
620 |
if st.button("▶️", key=f"read_{timestamp}_{file}"): # search emoji button
|
621 |
file_contents = load_file(file)
|
622 |
+
file_name = file
|
623 |
+
next_action = 'search'
|
624 |
st.session_state['next_action'] = next_action
|
625 |
with col5:
|
626 |
if st.button("🗑", key=f"delete_{timestamp}_{file}"):
|
627 |
os.remove(file)
|
628 |
+
file_name = file
|
629 |
st.rerun()
|
630 |
+
next_action = 'delete'
|
631 |
st.session_state['next_action'] = next_action
|
632 |
|
633 |
|