Add more test-cases for UTF-8 parser, including test-cases for TCL_UTF_MAX=4 or TCL_UTF_MAX=6

1 files changed, 32 insertions, 6 deletions

diff --git a/tests/utf.test b/tests/utf.test
index a03dd6c..28981d6 100644
--- a/tests/utf.test
+++ b/tests/utf.test
@@ -20,6 +20,9 @@ testConstraint testbytestring [llength [info commands testbytestring]]
 
 catch {unset x}
 
+# Some tests require support for 4-byte UTF-8 sequences
+testConstraint fullutf [expr {[format %c 0x010000] != "\ufffd"}]
+
 test utf-1.1 {Tcl_UniCharToUtf: 1 byte sequences} testbytestring {
     expr {"\x01" eq [testbytestring "\x01"]}
 } 1
@@ -38,6 +41,9 @@ test utf-1.5 {Tcl_UniCharToUtf: overflowed Tcl_UniChar} testbytestring {
 test utf-1.6 {Tcl_UniCharToUtf: negative Tcl_UniChar} testbytestring {
     expr {[format %c -1] eq [testbytestring "\xef\xbf\xbd"]}
 } 1
+test utf-1.7 {Tcl_UniCharToUtf: 4 byte sequences} -constraints {fullutf testbytestring} -body {
+    expr {"\U014e4e" eq [testbytestring "\xf0\x94\xb9\x8e"]}
+} -result 1
 
 test utf-2.1 {Tcl_UtfToUniChar: low ascii} {
     string length "abc"
@@ -60,9 +66,21 @@ test utf-2.6 {Tcl_UtfToUniChar: lead (3-byte) followed by 1 trail} testbytestrin
 test utf-2.7 {Tcl_UtfToUniChar: lead (3-byte) followed by 2 trail} testbytestring {
     string length [testbytestring "\xE4\xb9\x8e"]
 } {1}
-test utf-2.8 {Tcl_UtfToUniChar: longer UTF sequences not supported} testbytestring {
-    string length [testbytestring "\xF4\xA2\xA2\xA2"]
+test utf-2.8 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail} -constraints {fullutf testbytestring} -body {
+    string length [testbytestring "\xF0\x90\x80\x80"]
+} -result {1}
+test utf-2.9 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail} -constraints {fullutf testbytestring} -body {
+    string length [testbytestring "\xF4\x8F\xBF\xBF"]
+} -result {1}
+test utf-2.10 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail, underflow} testbytestring {
+    string length [testbytestring "\xF0\x8F\xBF\xBF"]
+} {4}
+test utf-2.11 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail, overflow} testbytestring {
+    string length [testbytestring "\xF4\x90\x80\x80"]
 } {4}
+test utf-2.12 {Tcl_UtfToUniChar: longer UTF sequences not supported} testbytestring {
+    string length [testbytestring "\xF8\xA2\xA2\xA2\xA2"]
+} {5}
 
 test utf-3.1 {Tcl_UtfCharComplete} {
 } {}
@@ -195,8 +213,16 @@ bsCheck \Ua1	161
 bsCheck \U4e21	20001
 bsCheck \U004e21	20001
 bsCheck \U00004e21	20001
-bsCheck \U00110000	65533
-bsCheck \Uffffffff	65533
+bsCheck \U0000004e21	78
+if {[testConstraint fullutf]} {
+    bsCheck \U00110000	69632
+    bsCheck \U01100000	69632
+    bsCheck \U11000000	69632
+    bsCheck \U0010FFFF	1114111
+    bsCheck \U010FFFF0	1114111
+    bsCheck \U10FFFF00	1114111
+    bsCheck \UFFFFFFFF	1048575
+}
 
 test utf-11.1 {Tcl_UtfToUpper} {
     string toupper {}
@@ -264,8 +290,8 @@ test utf-16.1 {Tcl_UniCharToLower, negative delta} {
     string tolower aA
 } aa
 test utf-16.2 {Tcl_UniCharToLower, positive delta} {
-    string tolower \u0178\u00ff\uA78D\u01c5
-} \u00ff\u00ff\u0265\u01c6
+    string tolower \u0178\u00ff\uA78D\u01c5\U10400
+} \u00ff\u00ff\u0265\u01c6\U10428
 
 test utf-17.1 {Tcl_UniCharToLower, no delta} {
     string tolower !