Merge pull request #57 from adelavega/fix/pmcid

Fix: PMCID extraction

Author: adelavega

src/pubget/_metadata.py

@@ -90,7 +90,7 @@ def _add_license(article: etree.Element, metadata: Dict[str, Any]) -> None:
 
 def _add_id(article_id: etree.Element, metadata: Dict[str, Any]) -> None:
     id_type = article_id.get("pub-id-type")
-    if id_type not in ["pmc", "pmid", "doi"]:
+    if id_type not in ["pmc", "pmid", "doi", "pmcid"]:
         return
     if id_type == "pmc":
         id_type = "pmcid"

src/pubget/_utils.py

@@ -136,12 +136,17 @@ def get_pmcid(article: Union[etree.ElementTree, etree.Element]) -> int:
     pmcid = article.find(
         "front/article-meta/article-id[@pub-id-type='pmcid']"
     )
-    if pmc is None and pmcid is None:
-        raise ValueError("No PMC ID found in the article XML.")
-    if pmc:
+
+    val = None
+    if pmc is not None:
         val = pmc.text
-    else:
-        val = pmcid.text.replace("PMC", "")
+    elif pmcid is not None:
+        val = pmcid.text
+        if val.startswith("PMC"):
+            val = val[3:]
+
+    if val is None or not val.isdigit():
+        raise ValueError("No valid PMCID found in article XML.")
 
     return int(val)