Improve snippet checking.

snippet/bin/check_manuscript.py

@@ -1,15 +1,15 @@
 #!/usr/bin/env python
 
 # Check that the expected (or actual) snippets are in the manuscript. E.g.
-# bin/check_manuscript.py  ~/book-workspace/htdg2/ch11.xml expected/ch11/grunt/*
+# bin/check_manuscript.py  ~/book-workspace/htdg3/ch11.xml expected/ch11/grunt/*
 
-# TODO: ignore id elements in the XML in the book.
-# Need to use an XML parser for this.
 import sys
 
 manuscript = open(sys.argv[1], 'r').read()
 
 for snippet_file in sys.argv[2:]:
-  index = manuscript.find(open(snippet_file, 'r').read())
+  lines = open(snippet_file, 'r').readlines()
+  snippet = "".join(lines[1:]).strip() # remove first line (comment) and strip spaces
+  index = manuscript.find(snippet)
   if index == -1:
     print "Snippet not found", snippet_file

snippet/bin/check_manuscript.sh

@@ -5,9 +5,16 @@ bin=`cd "$bin"; pwd`
 
 actual="$bin"/../actual
 
-$bin/check_manuscript.py  ~/book-workspace/htdg3/ch02.xml $actual/ch02/*
-$bin/check_manuscript.py  ~/book-workspace/htdg3/ch04.xml $actual/ch04/*
-$bin/check_manuscript.py  ~/book-workspace/htdg3/ch05.xml $actual/ch05/*
-$bin/check_manuscript.py  ~/book-workspace/htdg3/ch07.xml $actual/ch07/*
+for ch in ch02 ch04 ch05 ch07 ch08
+do
+  # remove any id attributes from program listings
+  sed '/<programlisting/s/ id=".*"//' ~/book-workspace/htdg3/$ch.xml > /tmp/$ch.xml
+  $bin/check_manuscript.py  /tmp/$ch.xml $actual/$ch/*
+done
+
+# Avro check
+sed '/<programlisting/s/ id=".*"//' ~/book-workspace/htdg3/ch04.xml > /tmp/ch04-avro.xml
+$bin/check_manuscript.py  /tmp/ch04-avro.xml $actual/ch04-avro/*
+
+# Common check
 $bin/check_manuscript.py  ~/book-workspace/htdg3/ch07.xml $actual/common/*
-$bin/check_manuscript.py  ~/book-workspace/htdg3/ch08.xml $actual/ch08/*