Do not find DOI for neurips papers. Add test and fixes #6

DuBibtex.py

@@ -67,6 +67,7 @@ class Re:
   acmBib = re.compile('<PRE id="[\d\.]+">(.+)<\/pre>',
                       flags=re.MULTILINE | re.IGNORECASE | re.S)
   ieee = re.compile('ieee\.org(?:\/abstract)?\/document\/(\d+)', flags=re.MULTILINE)
+  neurips = re.compile(r'proceedings.neurips.cc\/paper', flags=re.MULTILINE)
   year = re.compile('\w+(\d+)')
 
 
@@ -351,18 +352,22 @@ def google_lookup(s, parser, use_scholar=False):
     f.write(html)
 
 
-  url_regexes = ['doiAcmUrl', 'acm', 'doiSpringer', 'doiWiley', 'doiUrl', 'ieee', 'doiCaltech', 'doiPubmed']
+  url_regexes = ['doiAcmUrl', 'acm', 'doiSpringer', 'doiWiley', 'doiUrl', 'ieee', 'doiCaltech', 'doiPubmed', 'neurips']
 
   found_urls = []
   for url_regex in url_regexes:
     m = getattr(Re, url_regex).search(html)
-    if m and len(m.groups()) > 0:
+    if m:
       found_urls.append((url_regex, m, m.start()))
   # Sort by start position
   found_urls.sort(key=lambda x: x[2])
 
   for url_regex, m, _ in found_urls:
 
+    if url_regex == 'neurips':
+      # NeurIPS does not have a DOI.
+      return None
+
     if url_regex == 'doiAcmUrl' and m and len(m.groups()) > 0:
       res = m.groups()[0].replace('\\', '')
       print("DOI from Google and ACM DOI: %s\n" % res)

tests/inputs/ddpm.bib

@@ -0,0 +1,11 @@
+@inproceedings{NEURIPS2020_4c5bcfec,
+ author = {Ho, Jonathan and Jain, Ajay and Abbeel, Pieter},
+ booktitle = {Advances in Neural Information Processing Systems},
+ editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
+ pages = {6840--6851},
+ publisher = {Curran Associates, Inc.},
+ title = {Denoising Diffusion Probabilistic Models},
+ url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/4c5bcfec8584af0d967f1ab10179ca4b-Paper.pdf},
+ volume = {33},
+ year = {2020}
+}

tests/test_find_dois.py

@@ -25,8 +25,11 @@ def search_and_check_doi(filename, correct_doi):
         assert len(generated_library.entries) == len(
             library.entries), "Number of entries should be the same"
         for entry in generated_library.entries:
-            assert "doi" in entry
-            assert entry.fields_dict["doi"].value == correct_doi
+            if correct_doi:
+                assert "doi" in entry
+                assert entry.fields_dict["doi"].value == correct_doi
+            else:
+                assert "doi" not in entry
 
 
 @pytest.mark.parametrize("filename,correct_doi", [
@@ -42,3 +45,10 @@ def test_iccv_doi(filename, correct_doi):
 ])
 def test_tvcg_doi(filename, correct_doi):
     search_and_check_doi(filename, correct_doi)
+
+@pytest.mark.parametrize("filename", [
+    ("ddpm.bib"),
+])
+def test_neurips_doi(filename):
+    # NeuroIPS does not have a DOI
+    search_and_check_doi(filename, "")