3 files changed, 183 insertions, 0 deletions
diff --git a/taskcluster/docker/diffoscope/Dockerfile b/taskcluster/docker/diffoscope/Dockerfile
new file mode 100644
index 0000000000..20d79c016f
--- /dev/null
+++ b/taskcluster/docker/diffoscope/Dockerfile
@@ -0,0 +1,30 @@
+FROM $DOCKER_IMAGE_PARENT
+MAINTAINER Mike Hommey <mhommey@mozilla.com>
+
+VOLUME /builds/worker/checkouts
+VOLUME /builds/worker/workspace
+VOLUME /builds/worker/tooltool-cache
+
+ENV LANG=en_US.UTF-8
+
+RUN apt-get install \
+      binutils-multiarch \
+      bzip2 \
+      curl \
+      enjarify \
+      diffoscope/buster-backports \
+      jsbeautifier \
+      libc++abi1 \
+      locales \
+      default-jdk-headless \
+      python3-progressbar \
+      unzip \
+      zip \
+      && \
+    sed -i '/en_US.UTF-8/s/^# *//' /etc/locale.gen && \
+    locale-gen
+
+COPY get_and_diffoscope /builds/worker/bin/get_and_diffoscope
+COPY readelf /builds/worker/bin/readelf
+
+RUN chown -R worker:worker /builds/worker/bin && chmod 755 /builds/worker/bin/*
diff --git a/taskcluster/docker/diffoscope/get_and_diffoscope b/taskcluster/docker/diffoscope/get_and_diffoscope
new file mode 100644
index 0000000000..7409313275
--- /dev/null
+++ b/taskcluster/docker/diffoscope/get_and_diffoscope
@@ -0,0 +1,140 @@
+#!/bin/bash
+
+set -e
+set -x
+
+cd /builds/worker
+
+mkdir a b
+
+# /builds/worker/bin contains wrapper binaries to divert what diffoscope
+# needs to use, so it needs to appear first.
+export PATH=/builds/worker/bin:$PATH
+
+# Until https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=879010 is
+# implemented, it's better to first manually extract the data.
+# Plus dmg files are not supported yet.
+
+case "$ORIG_URL" in
+*.zip|*.apk)
+	curl -L "$ORIG_URL" > a.zip
+	curl -L "$NEW_URL" > b.zip
+	unzip -d a a.zip
+	unzip -d b b.zip
+	;;
+*.tar.bz2)
+	curl -L "$ORIG_URL" | tar -C a -jxf -
+	curl -L "$NEW_URL" | tar -C b -jxf -
+	;;
+*.tar.gz)
+	curl -L "$ORIG_URL" | tar -C a -zxf -
+	curl -L "$NEW_URL" | tar -C b -zxf -
+	;;
+*.dmg)
+	for tool in lipo otool; do
+		ln -s $MOZ_FETCHES_DIR/cctools/bin/x86_64-apple-darwin*-$tool bin/$tool
+	done
+	curl -L "$ORIG_URL" > a.dmg
+	curl -L "$NEW_URL" > b.dmg
+	for i in a b; do
+		$MOZ_FETCHES_DIR/dmg/dmg extract $i.dmg $i.hfs
+		$MOZ_FETCHES_DIR/dmg/hfsplus $i.hfs extractall / $i
+	done
+	;;
+*)
+	ARTIFACT=$(basename "${ORIG_URL}")
+	curl -L "$ORIG_URL" > "a/${ARTIFACT}"
+	curl -L "$NEW_URL" > "b/${ARTIFACT}"
+esac
+
+case "$ORIG_URL" in
+*/target.apk)
+	OMNIJAR=assets/omni.ja
+	;;
+*)
+	OMNIJAR=omni.ja
+	;;
+esac
+
+report_error() {
+	# We "parse" the diff output, so we look at the lines that contain a "tee", like:
+	# ├── firefox
+	# │ ├── libxul.so
+	# │ │ ├── readelf --wide --notes {}
+	# We ignore lines like the last one, to only report file names. And we ignore
+	# lines for directories such as the first one, but still look at them to report
+	# full paths.
+	python3 <<-EOF
+		TEE = '├──'
+		paths = set()
+		path = []
+		with open("$1.txt") as fh:
+		    for l in fh:
+		        if TEE not in l:
+		            continue
+		        fields = l.split()
+		        # We rely on the number of │ to figure out at what level the file
+		        # name applies.
+		        if fields[-2:-1] == [TEE]:
+		            path[len(fields) - 2:] = [fields[-1]]
+		        else:
+		            # Align path length to match the number of │
+		            path.append(None)
+		        path_ = [p for p in path if p]
+		        full_path = '/'.join(path_)
+		        parent_path = '/'.join(path_[:-1])
+		        if parent_path in paths:
+		            paths.remove(parent_path)
+		        if full_path:
+		            paths.add(full_path)
+
+		for p in sorted(paths):
+		    print('TEST-UNEXPECTED-FAIL | {} differs. See the $1.html or $1.txt artifact'.format(p))
+	EOF
+}
+
+# Builds are 99% of the time differing in some small ways, so it's not
+# really useful to report a failure (at least not until we actually
+# care about the builds being 100% identical).
+POST=true
+
+fail() {
+	exit 1
+}
+
+for option; do
+	case "$option" in
+	--unpack)
+		CURDIR=$PWD
+		for dir in a b; do
+			# Need to run mach python from inside the gecko source.
+			# See bug #1533642.
+			(cd $GECKO_PATH && ./mach python --no-virtualenv toolkit/mozapps/installer/unpack.py --omnijar $OMNIJAR $CURDIR/$dir)
+		done
+		;;
+	--fail)
+		POST="fail"
+		;;
+	*)
+		echo "Unsupported option: $option" >&2
+		exit 1
+	esac
+done
+
+if [ -n "$PRE_DIFF" ]; then
+	eval $PRE_DIFF
+fi
+
+if diffoscope \
+	--html diff.html \
+	--text diff.txt \
+	--progress \
+	$DIFFOSCOPE_ARGS \
+	a b
+then
+	# Ok
+	:
+else
+	report_error diff
+	$POST
+fi
diff --git a/taskcluster/docker/diffoscope/readelf b/taskcluster/docker/diffoscope/readelf
new file mode 100644
index 0000000000..6b864171d7
--- /dev/null
+++ b/taskcluster/docker/diffoscope/readelf
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+case "$1 $2" in
+"--wide --symbols")
+    # When called with --wide --symbols, we remove the first column (which
+    # is essentially a line number that is not very useful), and then sort,
+    # which will order symbols by address, making a diff more useful.
+    /usr/bin/readelf "$@" | awk -F: '{print $2}' | sort
+    ;;
+*)
+    exec /usr/bin/readelf "$@"
+    ;;
+esac