started process to get formatted refs only once and not every time in the loop

news/patch_read_lists.rst

@@ -0,0 +1,23 @@
+**Added:**
+
+* <news item>
+
+**Changed:**
+
+* <news item>
+
+**Deprecated:**
+
+* <news item>
+
+**Removed:**
+
+* <news item>
+
+**Fixed:**
+
+* Changed how the reading-list builder fetches the references from Crossref so that it only fetches each needed reference once.
+
+**Security:**
+
+* <news item>

regolith/builders/readinglistsbuilder.py

@@ -41,6 +41,22 @@
         """Render latex template"""
         rc = self.rc
 
+        # build the collection of formatted references so that we only go
+        # and fetch the formatted references once per doi
+        dois, formatted_refs = [], {}
+        for rlist in self.gtx["reading_lists"]:
+            for paper in rlist['papers']:
+                dois.append(paper.get('doi', ''))
+        dois = list(set(dois))
+        for item in ['tbd', '']:
+            if item in dois:
+                dois.remove('tbd')
+                dois.remove('')
+        for doi in dois:
+            ref_and_date = get_formatted_crossref_reference(doi)
+            formatted_refs.update({doi: ref_and_date})
+
+        # loop through the reading lists to build the files
         for rlist in self.gtx["reading_lists"]:
             listid = rlist["_id"]
             outfile_bib = listid
@@ -48,20 +64,14 @@
             n = 1
             for paper in rlist['papers']:
                 doi = paper.get('doi')
                 paper['text'] = paper['text'].strip('.').strip()
                 if doi == 'tbd' or doi == '':
                     doi = None
                 url = paper.get('url')
                 if doi:
-                    # if rc.verbose:
-                    #     print(f"getting {doi} for {paper.get('tite')}")
-                    ref, ref_date = get_formatted_crossref_reference(doi)
-                    # if rc.verbose:
-                    #     try:
-                    #         print(f"got ref: {ref}")
-                    #     except:
-                    #         print("obtained ref but print error")
-                    paper.update({'reference': ref, 'ref_date': ref_date, 'n': n, 'label': 'DOI'})
+                    paper.update({'reference': formatted_refs.get(doi)[0],
+                                   'ref_date': formatted_refs.get(doi)[1],
+                                   'n': n, 'label': 'DOI'})
                     n += 1
                 elif url:
                     paper['doi'] = url

regolith/tools.py

@@ -1980,7 +1980,7 @@ def get_formatted_crossref_reference(doi):
 
     authorlist = [
         f"{a['given'].strip()} {a['family'].strip()}"
-        for a in article.get('message').get('author')]
+        for a in article.get('message',{}).get('author','')]
     try:
         journal = \
             article.get('message').get('short-container-title')[0]

tests/test_builders.py

@@ -26,6 +26,7 @@
     "grantreport",
     "resume",
     "review-man",
+    # reading-lists    need tests for this
     "reimb"
 ]
 db_srcs = ["mongo", "fs"]