From d656ad6531bd117aa2dc077d35f7df9c996376d4 Mon Sep 17 00:00:00 2001
From: Ian Johnson <enjalot@gmail.com>
Date: Wed, 16 Oct 2024 10:17:01 -0400
Subject: [PATCH] Accomodate date-type in parse_date

180/955 articles in oa_comm_xml.incr.2024-10-04 had dates recorded like:
```<pub-date date-type="collection" publication-format="electronic"><month>9</month><year>2024</year></pub-date>```
instead of
```
<pub-date pub-type="collection" publication-format="electronic"><month>9</month><year>2024</year></pub-date>
```
these changes will now properly parse those dates and not error on parsing the whole document
---
 pubmed_parser/pubmed_oa_parser.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/pubmed_parser/pubmed_oa_parser.py b/pubmed_parser/pubmed_oa_parser.py
index 8b079f6..49154eb 100644
--- a/pubmed_parser/pubmed_oa_parser.py
+++ b/pubmed_parser/pubmed_oa_parser.py
@@ -92,10 +92,15 @@ def parse_date(tree, date_type):
     def get_text(node):
         return node.text if node is not None else None
 
-    pub_date_path = f".//pub-date[@pub-type=\"{date_type}\"]"
+    pub_date_path = f".//pub-date[@pub-type='{date_type}' or @date-type='{date_type}']"
+    date_node = tree.xpath(pub_date_path)
+    
+    if not date_node:
+        return {}
+
     date_dict = {}
     for part in ["year", "month", "day"]:
-        text = get_text(tree.find(f"{pub_date_path}/{part}"))
+        text = get_text(date_node[0].find(part))
         if text is not None:
             date_dict[part] = text