Rename bookgraphv2.sh to the canonical bookgraph

Add word count to the graph. Use a scaled figure so it's easy to compare with the confidence.

1 files changed, 8 insertions, 3 deletions

diff --git a/bookgraphv2.sh b/bookgraph
index 3f6cfaa..8010bdc 100644..100755
--- a/bookgraphv2.sh
+++ b/bookgraph
@@ -1,7 +1,10 @@
 #!/bin/sh
 usage="$0 bookdir
 
-Creates a graph showing the average confidence of each page in a book."
+Creates a graph showing the average confidence and a (scaled) word
+count of each page in a book. The word count is scaled (divided by
+10 and plus 50) to make it easy to compare to the confidence, by
+generally occupying a similar scale."
 
 test $# -ne 1 && echo "$usage" && exit 1
 
@@ -15,7 +18,8 @@ t=`mktemp`
 find "$1/best" -maxdepth 1 -type f -name '*hocr' | while read i; do
 	c=`pgconf "$i"`
 	n=`basename "$i" .hocr | sed 's/_.*//'`
-	printf '%s\t%d\n' "$n" "$c" >> "$t"
+	w=`grep ocrx_word "$i" | wc -l`
+	printf '%s\t%d\t%d\n' "$n" "$c" "$w" >> "$t"
 done
 
 r=`readlink -f "$1"`
@@ -32,7 +36,8 @@ set mxtics
 set terminal png truecolor size 4600,1700
 set output "%s"
 plot "%s" using 1:2 with lines title "Confidence",\\
-     "" using 1:2:(sprintf("%%d", $1)) with labels point pt 2 notitle
+     "" using 1:2:(sprintf("%%d", $1)) with labels point pt 2 notitle,\\
+     "" using 1:($3/10+50) with lines title "Number of words (scaled)"
 quit
 ' "$b" "$1/${b}_bookgraph.png" "$t" >> "$gp"
 gnuplot "$gp"