Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,81 +1,81 @@
|
|
1 |
-
import json
|
2 |
-
import os
|
3 |
-
import pdf2doi
|
4 |
-
import gradio as gr
|
5 |
-
import requests
|
6 |
-
import html
|
7 |
-
|
8 |
-
def download_pdf(url):
|
9 |
-
file_path = f"{url.split('/')[-1]}.pdf"
|
10 |
-
response = requests.get(url)
|
11 |
-
with open(file_path, 'wb') as file:
|
12 |
-
file.write(response.content)
|
13 |
-
return file_path
|
14 |
-
|
15 |
-
def get_doi(pdf_path):
|
16 |
-
pdf2doi.config.set('verbose', False)
|
17 |
-
results = pdf2doi.pdf2doi(pdf_path)
|
18 |
-
validation_info = json.loads(results['validation_info'])
|
19 |
-
doi = validation_info.get('DOI', None)
|
20 |
-
title = validation_info.get('title', None)
|
21 |
-
url = validation_info.get('URL', None)
|
22 |
-
return doi, title, url
|
23 |
-
|
24 |
-
def get_paper_data(doi):
|
25 |
-
api_url = f"https://api.citeas.org/product/{doi}"
|
26 |
-
response = requests.get(api_url)
|
27 |
-
return response.json()
|
28 |
-
|
29 |
-
def main(pdf_url):
|
30 |
-
pdf_path = download_pdf(pdf_url)
|
31 |
-
doi, title, url = get_doi(pdf_path)
|
32 |
-
if doi is None:
|
33 |
-
return json.dumps({"error": "DOI not found"}, indent=4)
|
34 |
-
paper_data = get_paper_data(doi)
|
35 |
-
if not paper_data:
|
36 |
-
return json.dumps({"error": "Paper data not found"}, indent=4)
|
37 |
-
citation_text = None
|
38 |
-
for citation in paper_data.get('citations', []):
|
39 |
-
if citation.get('style_shortname') == 'apa':
|
40 |
-
citation_text = citation.get('citation')
|
41 |
-
break
|
42 |
-
title = title or paper_data.get('name')
|
43 |
-
url = url or f"https://doi.org/{doi}"
|
44 |
-
if citation_text:
|
45 |
-
citation_text = citation_text.encode('utf-8').decode('utf-8')
|
46 |
-
citation_text = html.unescape(
|
47 |
-
citation_text.replace("<i>", "").replace("</i>", "").replace("\u2026", "...").replace("\n", " ")
|
48 |
-
)
|
49 |
-
else:
|
50 |
-
citation_text = "Citation not found"
|
51 |
-
data = {
|
52 |
-
"doi": doi,
|
53 |
-
"title": title if title else "Title not found",
|
54 |
-
"citation_text": citation_text,
|
55 |
-
"url": url
|
56 |
-
}
|
57 |
-
os.remove(pdf_path)
|
58 |
-
return json.dumps(data, ensure_ascii=False, indent=4)
|
59 |
-
|
60 |
-
|
61 |
-
theme = gr.themes.Soft(
|
62 |
-
primary_hue="purple",
|
63 |
-
secondary_hue="cyan",
|
64 |
-
neutral_hue="slate",
|
65 |
-
font=[
|
66 |
-
gr.themes.GoogleFont("Syne"),
|
67 |
-
gr.themes.GoogleFont("Poppins"),
|
68 |
-
gr.themes.GoogleFont("Poppins"),
|
69 |
-
gr.themes.GoogleFont("Poppins")
|
70 |
-
],
|
71 |
-
)
|
72 |
-
|
73 |
-
with gr.Blocks(theme=theme) as app:
|
74 |
-
with gr.Row():
|
75 |
-
pdf_path = gr.Textbox(lines=1, label="PDF URL", placeholder="Enter the URL of the PDF")
|
76 |
-
doi_data = gr.Textbox(lines=7, label="DOI Data", placeholder="DOI data will be displayed here", show_copy_button=True)
|
77 |
-
get_data = gr.Button(value="Get DOI Data", variant='primary')
|
78 |
-
|
79 |
-
get_data.click(main, inputs=[pdf_path], outputs=[doi_data], api_name="getDOIData")
|
80 |
-
|
81 |
app.queue(default_concurrency_limit=250).launch()
|
|
|
1 |
+
import json
|
2 |
+
import os
|
3 |
+
import pdf2doi
|
4 |
+
import gradio as gr
|
5 |
+
import requests
|
6 |
+
import html
|
7 |
+
|
8 |
+
def download_pdf(url):
|
9 |
+
file_path = f"{url.split('/')[-1]}.pdf"
|
10 |
+
response = requests.get(url)
|
11 |
+
with open(file_path, 'wb') as file:
|
12 |
+
file.write(response.content)
|
13 |
+
return file_path
|
14 |
+
|
15 |
+
def get_doi(pdf_path):
|
16 |
+
pdf2doi.config.set('verbose', False)
|
17 |
+
results = pdf2doi.pdf2doi(pdf_path)
|
18 |
+
validation_info = json.loads(str(results['validation_info']))
|
19 |
+
doi = validation_info.get('DOI', None)
|
20 |
+
title = validation_info.get('title', None)
|
21 |
+
url = validation_info.get('URL', None)
|
22 |
+
return doi, title, url
|
23 |
+
|
24 |
+
def get_paper_data(doi):
|
25 |
+
api_url = f"https://api.citeas.org/product/{doi}"
|
26 |
+
response = requests.get(api_url)
|
27 |
+
return response.json()
|
28 |
+
|
29 |
+
def main(pdf_url):
|
30 |
+
pdf_path = download_pdf(pdf_url)
|
31 |
+
doi, title, url = get_doi(pdf_path)
|
32 |
+
if doi is None:
|
33 |
+
return json.dumps({"error": "DOI not found"}, indent=4)
|
34 |
+
paper_data = get_paper_data(doi)
|
35 |
+
if not paper_data:
|
36 |
+
return json.dumps({"error": "Paper data not found"}, indent=4)
|
37 |
+
citation_text = None
|
38 |
+
for citation in paper_data.get('citations', []):
|
39 |
+
if citation.get('style_shortname') == 'apa':
|
40 |
+
citation_text = citation.get('citation')
|
41 |
+
break
|
42 |
+
title = title or paper_data.get('name')
|
43 |
+
url = url or f"https://doi.org/{doi}"
|
44 |
+
if citation_text:
|
45 |
+
citation_text = citation_text.encode('utf-8').decode('utf-8')
|
46 |
+
citation_text = html.unescape(
|
47 |
+
citation_text.replace("<i>", "").replace("</i>", "").replace("\u2026", "...").replace("\n", " ")
|
48 |
+
)
|
49 |
+
else:
|
50 |
+
citation_text = "Citation not found"
|
51 |
+
data = {
|
52 |
+
"doi": doi,
|
53 |
+
"title": title if title else "Title not found",
|
54 |
+
"citation_text": citation_text,
|
55 |
+
"url": url
|
56 |
+
}
|
57 |
+
os.remove(pdf_path)
|
58 |
+
return json.dumps(data, ensure_ascii=False, indent=4)
|
59 |
+
|
60 |
+
|
61 |
+
theme = gr.themes.Soft(
|
62 |
+
primary_hue="purple",
|
63 |
+
secondary_hue="cyan",
|
64 |
+
neutral_hue="slate",
|
65 |
+
font=[
|
66 |
+
gr.themes.GoogleFont("Syne"),
|
67 |
+
gr.themes.GoogleFont("Poppins"),
|
68 |
+
gr.themes.GoogleFont("Poppins"),
|
69 |
+
gr.themes.GoogleFont("Poppins")
|
70 |
+
],
|
71 |
+
)
|
72 |
+
|
73 |
+
with gr.Blocks(theme=theme) as app:
|
74 |
+
with gr.Row():
|
75 |
+
pdf_path = gr.Textbox(lines=1, label="PDF URL", placeholder="Enter the URL of the PDF")
|
76 |
+
doi_data = gr.Textbox(lines=7, label="DOI Data", placeholder="DOI data will be displayed here", show_copy_button=True)
|
77 |
+
get_data = gr.Button(value="Get DOI Data", variant='primary')
|
78 |
+
|
79 |
+
get_data.click(main, inputs=[pdf_path], outputs=[doi_data], api_name="getDOIData")
|
80 |
+
|
81 |
app.queue(default_concurrency_limit=250).launch()
|