From 7e669f18120751e8f711731081ce51ae095986d8 Mon Sep 17 00:00:00 2001
From: Sascha Brawer <sascha@brawer.ch>
Date: Fri, 2 Mar 2018 11:29:40 +0100
Subject: [PATCH] Add tests on SSML parsing

A corpus of SSML input files will also be needed for fuzzing.
https://github.com/espeak-ng/espeak-ng/issues/407
---
 .gitignore                          |  1 +
 Makefile.am                         |  1 +
 tests/ssml.test                     | 13 +++++++++++++
 tests/ssml/billion-laughs.expected  |  3 +++
 tests/ssml/billion-laughs.ssml      | 20 ++++++++++++++++++++
 tests/ssml/external-entity.expected |  9 +++++++++
 tests/ssml/external-entity.ssml     | 12 ++++++++++++
 tests/ssml/language-switch.expected | 10 ++++++++++
 tests/ssml/language-switch.ssml     |  7 +++++++
 tests/ssml/spec-example-1.expected  | 12 ++++++++++++
 tests/ssml/spec-example-1.ssml      | 18 ++++++++++++++++++
 11 files changed, 106 insertions(+)
 create mode 100755 tests/ssml.test
 create mode 100644 tests/ssml/billion-laughs.expected
 create mode 100644 tests/ssml/billion-laughs.ssml
 create mode 100644 tests/ssml/external-entity.expected
 create mode 100644 tests/ssml/external-entity.ssml
 create mode 100644 tests/ssml/language-switch.expected
 create mode 100644 tests/ssml/language-switch.ssml
 create mode 100644 tests/ssml/spec-example-1.expected
 create mode 100644 tests/ssml/spec-example-1.ssml

diff --git a/.gitignore b/.gitignore
index 145e4cf1..617d6282 100644
--- a/.gitignore
+++ b/.gitignore
@@ -100,6 +100,7 @@ tests/*.test
 !tests/languages.test
 !tests/phoneme-output.test
 !tests/phoneme-audio.test
+!tests/ssml.test
 
 espeak-ng.pc
 
diff --git a/Makefile.am b/Makefile.am
index c2d5851b..28c39f87 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -252,6 +252,7 @@ check:	tests/encoding.check \
 	tests/api.check \
 	tests/phoneme-output.check \
 	tests/phoneme-audio.check \
+	tests/ssml.check \
 	tests/languages.check
 
 ##### phoneme data:
diff --git a/tests/ssml.test b/tests/ssml.test
new file mode 100755
index 00000000..37504383
--- /dev/null
+++ b/tests/ssml.test
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+test_ssml() {
+	INPUT=$1
+	echo "testing ${INPUT}"
+	cp $(dirname $INPUT)/$(basename ${INPUT%.*}).expected expected.txt
+	ESPEAK_DATA_PATH=`pwd` LD_LIBRARY_PATH=src:${LD_LIBRARY_PATH} \
+		src/espeak-ng -m -v en-US --ipa=2 -f ${INPUT} > actual.txt
+	diff expected.txt actual.txt || exit 1
+}
+
+for i in `ls tests/ssml/*.ssml` ; do test_ssml $i ; done
+
diff --git a/tests/ssml/billion-laughs.expected b/tests/ssml/billion-laughs.expected
new file mode 100644
index 00000000..18c59b5b
--- /dev/null
+++ b/tests/ssml/billion-laughs.expected
@@ -0,0 +1,3 @@
+ lˈɑːl ænd lˈɑːl nˈa͡ɪn
+ lˈɑːl
+
diff --git a/tests/ssml/billion-laughs.ssml b/tests/ssml/billion-laughs.ssml
new file mode 100644
index 00000000..a0b374ea
--- /dev/null
+++ b/tests/ssml/billion-laughs.ssml
@@ -0,0 +1,20 @@
+<?xml version="1.0"?>
+<!-- SSML version of the “Billion laughs” attack
+     https://en.wikipedia.org/wiki/Billion_laughs_attack
+-->
+<!DOCTYPE lolz [
+ <!ENTITY lol "lol">
+ <!ELEMENT lolz (#PCDATA)>
+ <!ENTITY lol1 "&lol;&lol;&lol;&lol;&lol;&lol;&lol;&lol;&lol;&lol;">
+ <!ENTITY lol2 "&lol1;&lol1;&lol1;&lol1;&lol1;&lol1;&lol1;&lol1;&lol1;&lol1;">
+ <!ENTITY lol3 "&lol2;&lol2;&lol2;&lol2;&lol2;&lol2;&lol2;&lol2;&lol2;&lol2;">
+ <!ENTITY lol4 "&lol3;&lol3;&lol3;&lol3;&lol3;&lol3;&lol3;&lol3;&lol3;&lol3;">
+ <!ENTITY lol5 "&lol4;&lol4;&lol4;&lol4;&lol4;&lol4;&lol4;&lol4;&lol4;&lol4;">
+ <!ENTITY lol6 "&lol5;&lol5;&lol5;&lol5;&lol5;&lol5;&lol5;&lol5;&lol5;&lol5;">
+ <!ENTITY lol7 "&lol6;&lol6;&lol6;&lol6;&lol6;&lol6;&lol6;&lol6;&lol6;&lol6;">
+ <!ENTITY lol8 "&lol7;&lol7;&lol7;&lol7;&lol7;&lol7;&lol7;&lol7;&lol7;&lol7;">
+ <!ENTITY lol9 "&lol8;&lol8;&lol8;&lol8;&lol8;&lol8;&lol8;&lol8;&lol8;&lol8;">
+]>
+<speak version="1.1" xmlns="http://www.w3.org/2001/10/synthesis">
+  lol &lol9; lol
+</speak>
diff --git a/tests/ssml/external-entity.expected b/tests/ssml/external-entity.expected
new file mode 100644
index 00000000..8d8cef37
--- /dev/null
+++ b/tests/ssml/external-entity.expected
@@ -0,0 +1,9 @@
+
+
+ hˈɪɹ ɑː͡ɹ ðə pˈæswɜːdz
+
+
+ ænd ˈɛkszˈiːɾæk
+
+
+
diff --git a/tests/ssml/external-entity.ssml b/tests/ssml/external-entity.ssml
new file mode 100644
index 00000000..8ceb6058
--- /dev/null
+++ b/tests/ssml/external-entity.ssml
@@ -0,0 +1,12 @@
+<?xml version="1.0"?>
+<!-- SSML version of the “XML External Entity” attack
+     https://en.wikipedia.org/wiki/XML_external_entity_attack
+-->
+<!DOCTYPE test [ 
+    <!ENTITY xxeattack SYSTEM "file:///etc/passwd"> 
+]>
+<speak version="1.1" xmlns="http://www.w3.org/2001/10/synthesis"
+       xml:lang="en-US">
+  <p>Here are the passwords:</p>
+  <p>&xxeattack;</p>
+</speak>
diff --git a/tests/ssml/language-switch.expected b/tests/ssml/language-switch.expected
new file mode 100644
index 00000000..5b65ee9b
--- /dev/null
+++ b/tests/ssml/language-switch.expected
@@ -0,0 +1,10 @@
+
+ bˈuːhuː
+
+ bˈoːhuː
+
+ booˈɔo
+
+ bˌooˈoo
+
+
diff --git a/tests/ssml/language-switch.ssml b/tests/ssml/language-switch.ssml
new file mode 100644
index 00000000..8772babb
--- /dev/null
+++ b/tests/ssml/language-switch.ssml
@@ -0,0 +1,7 @@
+<?xml version="1.0"?>
+<speak version="1.1" xmlns="http://www.w3.org/2001/10/synthesis">
+  <p xml:lang="en">Boohoo</p>
+  <p xml:lang="de">Boohoo</p>
+  <p xml:lang="it">Boohoo</p>
+  <p xml:lang="es">Boohoo</p>
+</speak>
diff --git a/tests/ssml/spec-example-1.expected b/tests/ssml/spec-example-1.expected
new file mode 100644
index 00000000..6760149b
--- /dev/null
+++ b/tests/ssml/spec-example-1.expected
@@ -0,0 +1,12 @@
+
+
+
+ juː hæv fˈoː͡ɹ nˈuː mˈɛsɪd͡ʒᵻz
+
+ ðə fˈɜːst ɪz fɹʌm stˈɛfəni wˈɪljəmz ænd ɚɹˈa͡ɪvd æt
+ θɹˈiː fˈoː͡ɹɾi fˈa͡ɪv pˌiːˈɛm
+
+ ðə sˈʌbd͡ʒɛkt ɪz skˈiː tɹˈɪp
+
+
+
diff --git a/tests/ssml/spec-example-1.ssml b/tests/ssml/spec-example-1.ssml
new file mode 100644
index 00000000..e46706d5
--- /dev/null
+++ b/tests/ssml/spec-example-1.ssml
@@ -0,0 +1,18 @@
+<?xml version="1.0"?>
+<!-- Example from SSML 1.1 specification, Appendix E
+     https://www.w3.org/TR/speech-synthesis11/#AppE
+-->
+<speak version="1.1"
+       xmlns="http://www.w3.org/2001/10/synthesis"
+       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+       xsi:schemaLocation="http://www.w3.org/2001/10/synthesis
+                   http://www.w3.org/TR/speech-synthesis11/synthesis.xsd"
+       xml:lang="en-US">
+  <p>
+    <s>You have 4 new messages.</s>
+    <s>The first is from Stephanie Williams and arrived at <break/> 3:45pm.</s>
+    <s>
+      The subject is <prosody rate="-20%">ski trip</prosody>
+    </s>
+  </p>
+</speak>