Merge 8.6. And add more documentation and test-cases regarding the behavior of Tcl_UniCharToUtf()

2 files changed, 16 insertions, 4 deletions

diff --git a/tests/pkgIndex.tcl b/tests/pkgIndex.tcl
index 48ab71b..0feb0eb 100644
--- a/tests/pkgIndex.tcl
+++ b/tests/pkgIndex.tcl
@@ -1,6 +1,6 @@
 #! /usr/bin/env tclsh
 
-package ifneeded tcltests 0.1 {
-	source [file dirname [file dirname [file normalize [info script]/...]]]/tcltests.tcl
-	package provide tcltests 0.1
-}
+package ifneeded tcltests 0.1 "
+    source [list $dir]/tcltests.tcl
+    package provide tcltests 0.1
+"
diff --git a/tests/utf.test b/tests/utf.test
index 9dd8017..67a6778 100644
--- a/tests/utf.test
+++ b/tests/utf.test
@@ -44,6 +44,18 @@ test utf-1.6 {Tcl_UniCharToUtf: negative Tcl_UniChar} testbytestring {
 test utf-1.7 {Tcl_UniCharToUtf: 4 byte sequences} -constraints testbytestring -body {
     expr {"\U014e4e" eq [testbytestring "\xf0\x94\xb9\x8e"]}
 } -result 1
+test utf-1.8 {Tcl_UniCharToUtf: 3 byte sequence, upper surrogate} testbytestring {
+    expr {"\ud842" eq [testbytestring "\xed\xa1\x82"]}
+} 1
+test utf-1.9 {Tcl_UniCharToUtf: 3 byte sequence, lower surrogate} testbytestring {
+    expr {"\udc42" eq [testbytestring "\xed\xb1\x82"]}
+} 1
+test utf-1.10 {Tcl_UniCharToUtf: 3 byte sequence, upper surrogate} testbytestring {
+    expr {[format %c 0xd842] eq [testbytestring "\xed\xa1\x82"]}
+} 1
+test utf-1.11 {Tcl_UniCharToUtf: 3 byte sequence, lower surrogate} testbytestring {
+    expr {[format %c 0xdc42] eq [testbytestring "\xed\xb1\x82"]}
+} 1
 
 test utf-2.1 {Tcl_UtfToUniChar: low ascii} {
     string length "abc"