From 716b56b71b8513faa617acb38e2a841b59910b44 Mon Sep 17 00:00:00 2001
From: Mohammad Akhlaghi <mohammad@akhlaghi.org>
Date: Sun, 8 Mar 2020 18:15:32 +0000
Subject: Menke+20 example: properly count number of papers with software

Until now, I was mistakenly multiplying the fraction of papers in that
journal. This is corrected with this commit.
---
 reproduce/analysis/make/analysis-1.mk | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

(limited to 'reproduce/analysis/make')

diff --git a/reproduce/analysis/make/analysis-1.mk b/reproduce/analysis/make/analysis-1.mk
index 9d0018e..f739306 100644
--- a/reproduce/analysis/make/analysis-1.mk
+++ b/reproduce/analysis/make/analysis-1.mk
@@ -47,9 +47,10 @@ $(mk20tab3): $(indir)/menke20.xlsx | $(a1dir)
         # count as a delimter.
 	echo "# Column 1: YEAR [counter, i16] Year of journal's publication." > $@.tmp
 	echo "# Column 2: NUM_PAPERS [counter, i16] Number of studied papers in that journal." >> $@.tmp
-	echo "# Column 3: NUM_ID_TOOLS [counter, i16] Number of software/tools that were identified." >> $@.tmp
-	echo "# Column 4: JOURNAL_NAME [string, str150] Name of journal." >> $@.tmp
-	awk 'NR>1{printf("%-10d%-10d%-10d %s\n", $$2, $$3, $$(NF-1)*$$NF, $$1)}' \
+	echo "# Column 3: NUM_PAPERS_WITH_TOOLS [counter, i16] Number of papers with an identified tool." >> $@.tmp
+	echo "# Column 4: NUM_ID_TOOLS [counter, i16] Number of software/tools that were identified." >> $@.tmp
+	echo "# Column 5: JOURNAL_NAME [string, str150] Name of journal." >> $@.tmp
+	awk 'NR>1{printf("%-10d%-10d%-10d%-10d %s\n", $$2, $$3, $$3*$$NF, $$(NF-1), $$1)}' \
 	    FPAT='([^,]+)|("[^"]+")' $(indir)/$$base.csv >> $@.tmp
 
         # Set the temporary file as the final target. This was done so if
@@ -60,6 +61,16 @@ $(mk20tab3): $(indir)/menke20.xlsx | $(a1dir)
 
 
 
+############################
+# Recreate Figure 1C of Menke+20.
+############################
+
+# awk '!/^#/{all[$1]+=$2; id[$1]+=$3} END{for(year in all){print year, id[year]/all[year]}}' menke20-table-3.txt.tmp
+############################
+
+
+
+
 
 # Main LaTeX macro file
 $(mtexdir)/analysis-1.tex: $(mk20tab3) | $(mtexdir)
@@ -73,6 +84,6 @@ $(mtexdir)/analysis-1.tex: $(mk20tab3) | $(mtexdir)
         # column and separated the last numeric column from the string
         # column with a space. If the number of numeric columns change in
         # the future, the `31' also has to change.
-	v=$$(awk 'BEGIN{FIELDWIDTHS="31 10000"} !/^#/{print $$2}' \
+	v=$$(awk 'BEGIN{FIELDWIDTHS="41 10000"} !/^#/{print $$2}' \
 	         $(mk20tab3) | uniq | wc -l)
 	echo "\newcommand{\menkenumjournals}{$$v}" >> $@
-- 
cgit v1.2.1