Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -9,6 +9,8 @@ from huggingface_hub import InferenceClient,HfApi
|
|
| 9 |
import random
|
| 10 |
import json
|
| 11 |
import datetime
|
|
|
|
|
|
|
| 12 |
#from query import tasks
|
| 13 |
from agent import (
|
| 14 |
PREFIX,
|
|
@@ -89,7 +91,6 @@ def read_txt(txt_path):
|
|
| 89 |
return text
|
| 90 |
|
| 91 |
def read_pdf(pdf_path):
|
| 92 |
-
from pypdf import PdfReader
|
| 93 |
text=""
|
| 94 |
reader = PdfReader(f'{pdf_path}')
|
| 95 |
number_of_pages = len(reader.pages)
|
|
@@ -100,6 +101,26 @@ def read_pdf(pdf_path):
|
|
| 100 |
return text
|
| 101 |
|
| 102 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 103 |
VERBOSE = True
|
| 104 |
MAX_HISTORY = 100
|
| 105 |
MAX_DATA = 25000
|
|
@@ -207,12 +228,15 @@ def compress_data(c, instruct, history):
|
|
| 207 |
|
| 208 |
|
| 209 |
|
| 210 |
-
def summarize(inp,history,data=None,file=None,url=None):
|
| 211 |
if inp == "":
|
| 212 |
inp = "Process this data"
|
| 213 |
history.clear()
|
| 214 |
history = [(inp,"Working on it...")]
|
| 215 |
yield "",history
|
|
|
|
|
|
|
|
|
|
| 216 |
if url != "":
|
| 217 |
val, out = find_all(url)
|
| 218 |
if not val:
|
|
@@ -280,10 +304,11 @@ with gr.Blocks() as app:
|
|
| 280 |
file=gr.File(label="Input File (.pdf .txt)")
|
| 281 |
with gr.Tab("URL"):
|
| 282 |
url = gr.Textbox(label="URL")
|
| 283 |
-
|
|
|
|
| 284 |
#text=gr.JSON()
|
| 285 |
#inp_query.change(search_models,inp_query,models_dd)
|
| 286 |
clear_btn.click(clear_fn,None,[prompt,chatbot])
|
| 287 |
-
go=button.click(summarize,[prompt,chatbot,data,file,url],[prompt,chatbot])
|
| 288 |
stop_button.click(None,None,None,cancels=[go])
|
| 289 |
app.launch(server_port=7860,show_api=False)
|
|
|
|
| 9 |
import random
|
| 10 |
import json
|
| 11 |
import datetime
|
| 12 |
+
from pypdf import PdfReader
|
| 13 |
+
import uuid
|
| 14 |
#from query import tasks
|
| 15 |
from agent import (
|
| 16 |
PREFIX,
|
|
|
|
| 91 |
return text
|
| 92 |
|
| 93 |
def read_pdf(pdf_path):
|
|
|
|
| 94 |
text=""
|
| 95 |
reader = PdfReader(f'{pdf_path}')
|
| 96 |
number_of_pages = len(reader.pages)
|
|
|
|
| 101 |
return text
|
| 102 |
|
| 103 |
|
| 104 |
+
def read_pdf_online(url):
|
| 105 |
+
uid=uuid.uuid4()
|
| 106 |
+
response = requests.get(url, stream=True)
|
| 107 |
+
if response.status_code == 200:
|
| 108 |
+
with open(f"{uid}.pdf", "wb") as f:
|
| 109 |
+
f.write(response.content)
|
| 110 |
+
f.close()
|
| 111 |
+
#out = Path("./data.pdf")
|
| 112 |
+
#print (out)
|
| 113 |
+
reader = PdfReader(f"{uid}.pdf")
|
| 114 |
+
number_of_pages = len(reader.pages)
|
| 115 |
+
for i in range(number_of_pages-1):
|
| 116 |
+
page = reader.pages[i]
|
| 117 |
+
text = f'{text}\n{page.extract_text()}'
|
| 118 |
+
print (text)
|
| 119 |
+
else:
|
| 120 |
+
text = response.status_code
|
| 121 |
+
print(text)
|
| 122 |
+
return text
|
| 123 |
+
|
| 124 |
VERBOSE = True
|
| 125 |
MAX_HISTORY = 100
|
| 126 |
MAX_DATA = 25000
|
|
|
|
| 228 |
|
| 229 |
|
| 230 |
|
| 231 |
+
def summarize(inp,history,data=None,file=None,url=None,pdf_url=None):
|
| 232 |
if inp == "":
|
| 233 |
inp = "Process this data"
|
| 234 |
history.clear()
|
| 235 |
history = [(inp,"Working on it...")]
|
| 236 |
yield "",history
|
| 237 |
+
if pdf_url.startswith("http"):
|
| 238 |
+
out = read_pdf_online(url)
|
| 239 |
+
data=out
|
| 240 |
if url != "":
|
| 241 |
val, out = find_all(url)
|
| 242 |
if not val:
|
|
|
|
| 304 |
file=gr.File(label="Input File (.pdf .txt)")
|
| 305 |
with gr.Tab("URL"):
|
| 306 |
url = gr.Textbox(label="URL")
|
| 307 |
+
with gr.Tab("PDF URL"):
|
| 308 |
+
pdf_url = gr.Textbox(label="PDF URL")
|
| 309 |
#text=gr.JSON()
|
| 310 |
#inp_query.change(search_models,inp_query,models_dd)
|
| 311 |
clear_btn.click(clear_fn,None,[prompt,chatbot])
|
| 312 |
+
go=button.click(summarize,[prompt,chatbot,data,file,url,pdf_url],[prompt,chatbot])
|
| 313 |
stop_button.click(None,None,None,cancels=[go])
|
| 314 |
app.launch(server_port=7860,show_api=False)
|