Parse pdfs

app.py

@@ -1,16 +1,16 @@
 import tempfile
 from pathlib import Path
 
+import pymupdf
 import streamlit as st
 
-from refcheck.doi import get_dois, doi_to_url
-from refcheck.bibtex import fetch_bibtex
+from refcheck.doi import get_dois, doi_to_url, fetch_doi_json
 
-DOI_REGEX = "10.\\d{4,9}/[-._;()/:a-z0-9A-Z]+"
 FORMAT_TO_EXT = {
     "markdown": "md",
     "docx": "docx",
     "latex": "tex",
+    "pdf": "pdf",
 }
 
 
@@ -28,11 +28,23 @@
     path = Path(temp_dir) / uploaded_file.name
     with open(path, "wb") as f:
         f.write(uploaded_file.getvalue())
-    results = get_dois(path, format)
+    if format == "pdf":
+        doc = pymupdf.open(path)
+        text = (
+            "".join(page.get_text() for page in doc).replace("\n", " ").encode("utf-8")
+        )
+        txt_path = path.with_suffix(".txt")
+        with open(txt_path, "wb") as f:
+            f.write(text)
+        results = get_dois(txt_path, "rtf")
+    else:
+        results = get_dois(path, format)
     urls = [doi_to_url(doi) for doi in results]
-    bibtexes = [fetch_bibtex(url) for url in urls]
-
+    doi_jsons = [fetch_doi_json(url) for url in urls]
     st.write("Here are the dois in your document. Please check if they are correct!")
-    for doi, url, bibtex in zip(results, urls, bibtexes):
-        st.write(f"[**{doi}**]({url}):", bibtex)
+    for doi, url, doi_json in zip(results, urls, doi_jsons):
+        title = f"*{doi_json["title"]}*"
+        authors = ", ".join(f"{a["given"]} {a["family"]}" for a in doi_json["author"])
+        st.write(f"[**{doi}**]({url}): {title}", unsafe_allow_html=True)
+        st.write(authors)
         st.write("")

pyproject.toml

@@ -5,6 +5,7 @@ description = "Add your description here"
 readme = "README.md"
 requires-python = ">=3.13"
 dependencies = [
+    "pymupdf>=1.24.14",
     "pypandoc-binary>=1.14",
     "streamlit>=1.40.2",
 ]

src/refcheck/doi.py

@@ -1,4 +1,6 @@
+import json
 import re
+import requests
 from pathlib import Path
 
 import pypandoc
@@ -14,3 +16,9 @@ def get_dois(path: Path, format) -> list[str]:
 
 def doi_to_url(doi: str) -> str:
     return "http://dx.doi.org/" + doi
+
+
+def fetch_doi_json(doi_url: str) -> dict:
+    headers = {"accept": "application/json"}
+    r = requests.get(doi_url, headers=headers)
+    return json.loads(r.text) if r is not None else dict()