Add bash scripts for parallel execution.

MANIFEST.in

@@ -2,3 +2,5 @@ include README
 include TODO
 include COPYING
 include apachedex/jquery*.js
+include parallel_parse.sh
+include parallel_parse_helper.sh

README

@@ -170,6 +170,10 @@ For better performance...
     wait
     apachedex "$@" --out access.html --state-file access.*.json
 
+  If you have bash and have an xargs implementation supporting `-P`, you may
+  want to use `parallel_parse.sh` (which needs `parallel_parse_helper.sh`)
+  available in source distribution or from repository.
+
 Notes
 =====
 

parallel_parse.sh

+#!/bin/sh
+usage() {
+  echo "Usage:"
+  echo "  find [...] -print0 | $0 \\"
+  echo "    path_to_helper parallelism state_dir out_file command [arg1 [...]]"
+  echo "Reads filenames to process from stdin, null-delimited."
+}
+
+if [ $1 = '-h' -o $1 = '--help' ]; then
+  usage
+  exit 0
+fi
+
+HELPER=$1
+if [ ! -x "$HELPER" ]; then
+  echo "$HELPER cannot be executed"
+  exit 1
+fi
+shift
+PARALLELISM=$1
+shift
+STATE_DIR=$1
+mkdir -p "$STATE_DIR" || exit $?
+shift
+OUT_FILE=$1
+shift
+
+if [ $# -eq 0 ]; then
+  usage
+  exit 1
+fi
+
+xargs -0 -r -n 1 -P $PARALLELISM -I "@FILE@" "$HELPER" "@FILE@" "$STATE_DIR" "$@"
+"$@" --out "$OUT_FILE" --state-file "$STATE_DIR"/*

parallel_parse_helper.sh

+#!/bin/bash
+INFILE=$1
+shift
+STATE_DIR=$1
+shift
+echo -n '.'
+exec "$@" -Q --format json --out "$STATE_DIR/$(sed s:/:@:g <<< "$INFILE").json" "$INFILE"