Add PubMed fetching rule

examples/common/.gitignore

@@ -2,6 +2,7 @@ Crossref-April-2022
 .depend
 .depend.all
 ORCID_2022_10_summaries.tar.gz
+pubmed-data
 reports
 rolap
 ror-v1.17.1-2022-12-16.zip

examples/common/Makefile

@@ -11,6 +11,7 @@ CROSSREF_DIR?=../common/Crossref-April-2022
 USPTO_DIR?=../common/uspto-data
 ORCID_SUMMARIES?=../common/ORCID_2022_10_summaries.tar.gz
 ROR?=../common/ror-v1.17.1-2022-12-16.zip
+PUBMED_DIR?=../common/pubmed
 
 V?=1
 TIME?=time
@@ -26,6 +27,9 @@ $(CROSSREF_DIR):
 $(USPTO_DIR):
 	cd ../common && ./fetch-uspto.sh
 
+$(PUBMED_DIR):
+	cd ../common && ./fetch-pubmed.sh
+
 $(ORCID_SUMMARIES):
 	curl -L https://orcid.figshare.com/ndownloader/files/37635374 >$@
 

examples/common/fetch-pubmed.sh

@@ -0,0 +1,22 @@
+#!/bin/sh
+#
+# Fetch baseline PubMed bibliographic data
+#
+
+# Fail on command errors and unset variables
+set -eu
+
+BASE=https://ftp.ncbi.nlm.nih.gov/pubmed/baseline/
+
+mkdir pubmed-data
+cd pubmed-data
+
+# Obtain last baseline file from README.txt
+last=$(curl --silent https://ftp.ncbi.nlm.nih.gov/pubmed/baseline/README.txt |
+  sed -n '/pubmed24n0001/s/.*pubmed24n\([^.]*\)\.xml.*/\1/p')
+
+for n in $(seq 1 $last) ; do
+  file_name=$(printf 'pubmed24n%04d.xml.gz' $n)
+
+  curl --silent $BASE/$file_name >$file_name
+done