perf-test: bump version to 0.4, use manifest files

The new revision of the performance test includes manifests for each corpus,
so update the support library to use these manifests at the same time.

performance-test/download/notmuch-email-corpus-0.4.tar.xz.asc

@@ -0,0 +1,14 @@
+-----BEGIN PGP SIGNATURE-----
+Version: GnuPG v1.4.15 (GNU/Linux)
+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+=eSMY
+-----END PGP SIGNATURE-----

performance-test/perf-test-lib.sh

@@ -41,52 +41,70 @@ add_email_corpus ()
 {
     rm -rf ${MAIL_DIR}
 
-    case "$corpus_size" in
-	small)
-	    mail_subdir="mail/enron/bailey-s"
-	    check_for="${TEST_DIRECTORY}/corpus/$mail_subdir"
-	    ;;
-	medium)
-	    mail_subdir="mail/notmuch-archive"
-	    check_for="${TEST_DIRECTORY}/corpus/$mail_subdir"
-	    ;;
-	*)
-	    mail_subdir=mail
-	    check_for="${TEST_DIRECTORY}/corpus/$mail_subdir/enron/wolfe-j"
-    esac
+    CORPUS_DIR=${TEST_DIRECTORY}/corpus
+    mkdir -p "${CORPUS_DIR}"
 
-    MAIL_CORPUS="${TEST_DIRECTORY}/corpus/$mail_subdir"
-    TAG_CORPUS="${TEST_DIRECTORY}/corpus/tags"
+    MAIL_CORPUS="${CORPUS_DIR}/mail.${corpus_size}"
+    TAG_CORPUS="${CORPUS_DIR}/tags"
 
-    args=()
+    if command -v pixz > /dev/null; then
+	XZ=pixz
+    else
+	XZ=xz
+    fi
+
+    if [ ! -d "${CORPUS_DIR}/manifest" ]; then
+
+	printf "Unpacking manifests\n"
+	tar --extract --use-compress-program ${XZ} --strip-components=1 \
+	    --directory ${TEST_DIRECTORY}/corpus \
+	    --wildcards --file ../download/notmuch-email-corpus-${PERFTEST_VERSION}.tar.xz \
+	    'notmuch-email-corpus/manifest/*'
+    fi
+
+    file_list=$(mktemp file_listXXXXXX)
     if [ ! -d "$TAG_CORPUS" ] ; then
-	args+=("notmuch-email-corpus/tags")
+	echo "notmuch-email-corpus/tags" >> $file_list
     fi
 
-    if [ ! -d "$check_for" ] ; then
-	args+=("notmuch-email-corpus/$mail_subdir")
-    fi
-
-    if [[ ${#args[@]} > 0 ]]; then
-	if command -v pixz > /dev/null; then
-	    XZ=pixz
+    if [ ! -d "$MAIL_CORPUS" ] ; then
+	if [[ "$corpus_size" != "large" ]]; then
+	    sed s,^,notmuch-email-corpus/, < \
+		${TEST_DIRECTORY}/corpus/manifest/MANIFEST.${corpus_size} >> $file_list
 	else
-	    XZ=xz
+	    echo "notmuch-email-corpus/mail" >> $file_list
 	fi
+    fi
+
+    if [[ -s $file_list ]]; then
 
 	printf "Unpacking corpus\n"
-	mkdir -p "${TEST_DIRECTORY}/corpus"
-
 	tar --checkpoint=.5000 --extract --strip-components=1 \
 	    --directory ${TEST_DIRECTORY}/corpus \
 	    --use-compress-program ${XZ} \
 	    --file ../download/notmuch-email-corpus-${PERFTEST_VERSION}.tar.xz \
-	    "${args[@]}"
+	    --anchored --recursion \
+	    --files-from $file_list
 
 	printf "\n"
 
+	if [[ ! -d ${MAIL_CORPUS} ]]; then
+	    printf "creating link farm\n"
+
+	    if [[ "$corpus_size" = large ]]; then
+		cp -rl ${TEST_DIRECTORY}/corpus/mail ${MAIL_CORPUS}
+	    else
+		while read -r file; do
+		    tdir=${MAIL_CORPUS}/$(dirname $file)
+		    mkdir -p $tdir
+		    ln ${TEST_DIRECTORY}/corpus/$file $tdir
+		done <${TEST_DIRECTORY}/corpus/manifest/MANIFEST.${corpus_size}
+	    fi
+	fi
+
     fi
 
+    rm $file_list
     cp -lr $TAG_CORPUS $TMP_DIRECTORY/corpus.tags
     cp -lr $MAIL_CORPUS $MAIL_DIR
 }

performance-test/version.sh

@@ -1,3 +1,3 @@
 # this should be both a valid Makefile fragment and valid POSIX(ish) shell.
 
-PERFTEST_VERSION=0.3
+PERFTEST_VERSION=0.4