manubot#337 Add timeout to zotero searches.

manubot/cite/zotero.py

@@ -25,15 +25,15 @@
 """URL that provides access to the Zotero translation-server API"""
 
 
-def web_query(url: str) -> ZoteroData:
+def web_query(url: str, timeout=3) -> ZoteroData:
     """
     Return Zotero citation metadata for a URL as a list containing a single element that
     is a dictionary with the URL's metadata.
     """
     headers = {"User-Agent": get_manubot_user_agent(), "Content-Type": "text/plain"}
     params = {"single": 1}
     api_url = f"{base_url}/web"
-    response = requests.post(api_url, params=params, headers=headers, data=str(url))
+    response = requests.post(api_url, params=params, headers=headers, data=str(url), timeout=3)
     try:
         zotero_data = response.json()
     except Exception as error:
@@ -52,7 +52,7 @@ def web_query(url: str) -> ZoteroData:
     return zotero_data
 
 
-def search_query(identifier: str) -> ZoteroData:
+def search_query(identifier: str, timeout=3) -> ZoteroData:
     """
     Retrive Zotero metadata for a DOI, ISBN, PMID, or arXiv ID.
     Example usage:
@@ -66,7 +66,7 @@ def search_query(identifier: str) -> ZoteroData:
     """
     api_url = f"{base_url}/search"
     headers = {"User-Agent": get_manubot_user_agent(), "Content-Type": "text/plain"}
-    response = requests.post(api_url, headers=headers, data=str(identifier))
+    response = requests.post(api_url, headers=headers, data=str(identifier), timeout=timeout)
     try:
         zotero_data = response.json()
     except Exception as error:
@@ -93,7 +93,7 @@ def _passthrough_zotero_data(zotero_data: ZoteroData) -> ZoteroData:
     return zotero_data
 
 
-def export_as_csl(zotero_data: ZoteroData) -> CSLItems:
+def export_as_csl(zotero_data: ZoteroData, timeout=3) -> CSLItems:
     """
     Export Zotero JSON data to CSL JSON using a translation-server /export query.
     Performs a similar query to the following curl command:
@@ -107,7 +107,7 @@ def export_as_csl(zotero_data: ZoteroData) -> CSLItems:
     api_url = f"{base_url}/export"
     params = {"format": "csljson"}
     headers = {"User-Agent": get_manubot_user_agent()}
-    response = requests.post(api_url, params=params, headers=headers, json=zotero_data)
+    response = requests.post(api_url, params=params, headers=headers, json=zotero_data, timeout=timeout)
     if not response.ok:
         message = f"export_as_csl: translation-server returned status code {response.status_code}"
         logging.warning(f"{message} with the following output:\n{response.text}")
@@ -131,14 +131,14 @@ def get_csl_item(identifier: str) -> CSLItem:
     return csl_item
 
 
-def search_or_web_query(identifier: str) -> ZoteroData:
+def search_or_web_query(identifier: str, timeout=3) -> ZoteroData:
     """
     Detect whether `identifier` is a URL. If so,
     retrieve zotero metadata using a /web query.
     Otherwise, retrieve zotero metadata using a /search query.
     """
     if is_http_url(identifier):
-        zotero_data = web_query(identifier)
+        zotero_data = web_query(identifier, timeout=timeout)
     else:
-        zotero_data = search_query(identifier)
+        zotero_data = search_query(identifier, timeout=timeout)
     return zotero_data