BUG/MEDIUM: compression: Rewrite strong ETags
RFC 7232 section 2.3.3 states:
> Note: Content codings are a property of the representation data,
> so a strong entity-tag for a content-encoded representation has to
> be distinct from the entity tag of an unencoded representation to
> prevent potential conflicts during cache updates and range
> requests. In contrast, transfer codings (Section 4 of [RFC7230])
> apply only during message transfer and do not result in distinct
> entity-tags.
Thus a strong ETag must be changed when compressing. Usually this is done
by converting it into a weak ETag, which represents a semantically, but not
byte-by-byte identical response. A conversion to a weak ETag still allows
If-None-Match to work.
This should be backported to 1.9 and might be backported to every supported
branch with compression.
diff --git a/reg-tests/compression/h00001.vtc b/reg-tests/compression/h00001.vtc
new file mode 100644
index 0000000..0cae679
--- /dev/null
+++ b/reg-tests/compression/h00001.vtc
@@ -0,0 +1,226 @@
+varnishtest "Compression converts strong ETags to weak ETags"
+
+#REQUIRE_VERSION=1.9
+#REQUIRE_OPTION=ZLIB|SLZ
+
+feature ignore_unknown_macro
+
+server s1 {
+ rxreq
+ expect req.url == "/strong"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: \"123\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/weak"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: W/\"456\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/weak-incorrect-quoting"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: \"W/789\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/empty-strong"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: \"\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/empty-weak"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: W/\"\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/invalid1"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: \"invalid" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/invalid2"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: invalid\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/invalid3"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: invalid" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/invalid4"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: W/\"invalid" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/invalid5"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: W/invalid\"" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/invalid6"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: W/invalid" \
+ -bodylen 100
+
+ rxreq
+ expect req.url == "/multiple"
+ expect req.http.accept-encoding == "gzip"
+ txresp \
+ -hdr "Content-Type: text/plain" \
+ -hdr "ETag: \"one\"" \
+ -hdr "ETag: \"two\"" \
+ -bodylen 100
+} -start
+
+
+haproxy h1 -conf {
+ defaults
+ mode http
+ ${no-htx} option http-use-htx
+ timeout connect 1s
+ timeout client 1s
+ timeout server 1s
+
+ frontend fe-gzip
+ bind "fd@${fe_gzip}"
+ default_backend be-gzip
+
+ backend be-gzip
+ compression algo gzip
+ compression type text/html text/plain
+ server www ${s1_addr}:${s1_port}
+} -start
+
+client c1 -connect ${h1_fe_gzip_sock} {
+ txreq -url "/strong" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "gzip"
+ expect resp.http.etag == "W/\"123\""
+ gunzip
+ expect resp.bodylen == 100
+
+ txreq -url "/weak" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "gzip"
+ expect resp.http.etag == "W/\"456\""
+ gunzip
+ expect resp.bodylen == 100
+
+ txreq -url "/weak-incorrect-quoting" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "gzip"
+ expect resp.http.etag == "W/\"W/789\""
+ gunzip
+ expect resp.bodylen == 100
+
+ txreq -url "/empty-strong" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "gzip"
+ expect resp.http.etag == "W/\"\""
+ gunzip
+ expect resp.bodylen == 100
+
+ txreq -url "/empty-weak" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "gzip"
+ expect resp.http.etag == "W/\"\""
+ gunzip
+ expect resp.bodylen == 100
+
+ txreq -url "/invalid1" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.http.etag == "\"invalid"
+ expect resp.bodylen == 100
+
+ txreq -url "/invalid2" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.http.etag == "invalid\""
+ expect resp.bodylen == 100
+
+ txreq -url "/invalid3" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.http.etag == "invalid"
+ expect resp.bodylen == 100
+
+ txreq -url "/invalid4" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.http.etag == "W/\"invalid"
+ expect resp.bodylen == 100
+
+ txreq -url "/invalid5" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.http.etag == "W/invalid\""
+ expect resp.bodylen == 100
+
+ txreq -url "/invalid6" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.http.etag == "W/invalid"
+ expect resp.bodylen == 100
+
+ txreq -url "/multiple" \
+ -hdr "Accept-Encoding: gzip"
+ rxresp
+ expect resp.status == 200
+ expect resp.http.content-encoding == "<undef>"
+ expect resp.bodylen == 100
+} -run