TIP #601: Make "encoding convertto/convertfrom" throw exceptions

8 files changed, 75 insertions, 37 deletions

diff --git a/tests/chanio.test b/tests/chanio.test
index 2d26ac9..578dc9f 100644
--- a/tests/chanio.test
+++ b/tests/chanio.test
@@ -249,7 +249,7 @@ test chan-io-3.3 {WriteChars: compatibility with WriteBytes: flush on line} -bod
 } -cleanup {
     chan close $f
 } -result "\r\n12"
-test chan-io-3.4 {WriteChars: loop over stage buffer} {
+test chan-io-3.4 {WriteChars: loop over stage buffer} deprecated {
     # stage buffer maps to more than can be queued at once.
     set f [open $path(test1) w]
     chan configure $f -encoding jis0208 -buffersize 16
@@ -258,7 +258,7 @@ test chan-io-3.4 {WriteChars: loop over stage buffer} {
     chan close $f
     lappend x [contents $path(test1)]
 } [list "!)!)!)!)!)!)!)!)" "!)!)!)!)!)!)!)!)!)!)!)!)!)!)!)"]
-test chan-io-3.5 {WriteChars: saved != 0} {
+test chan-io-3.5 {WriteChars: saved != 0} deprecated {
     # Bytes produced by UtfToExternal from end of last channel buffer had to
     # be moved to beginning of next channel buffer to preserve requested
     # buffersize.
@@ -285,7 +285,7 @@ test chan-io-3.6 {WriteChars: (stageRead + dstWrote == 0)} {
     chan close $f
     lappend x [contents $path(test1)]
 } [list "12345678901234\x82\x60" "12345678901234\x82\x60\x82\x61"]
-test chan-io-3.7 {WriteChars: (bufPtr->nextAdded > bufPtr->length)} {
+test chan-io-3.7 {WriteChars: (bufPtr->nextAdded > bufPtr->length)} deprecated {
     # When translating UTF-8 to external, the produced bytes went past end of
     # the channel buffer. This is done on purpose - we then truncate the bytes
     # at the end of the partial character to preserve the requested blocksize
diff --git a/tests/cmdAH.test b/tests/cmdAH.test
index 5fefbeb..d787c7f 100644
--- a/tests/cmdAH.test
+++ b/tests/cmdAH.test
@@ -178,7 +178,7 @@ test cmdAH-4.2 {Tcl_EncodingObjCmd} -returnCodes error -body {
 } -result {unknown or ambiguous subcommand "foo": must be convertfrom, convertto, dirs, names, or system}
 test cmdAH-4.3 {Tcl_EncodingObjCmd} -returnCodes error -body {
     encoding convertto
-} -result {wrong # args: should be "encoding convertto ?encoding? data"}
+} -result {wrong # args: should be "encoding convertto ?-nocomplain? ?encoding? data"}
 test cmdAH-4.4 {Tcl_EncodingObjCmd} -returnCodes error -body {
     encoding convertto foo bar
 } -result {unknown encoding "foo"}
@@ -200,7 +200,7 @@ test cmdAH-4.6 {Tcl_EncodingObjCmd} -setup {
 } -result 8C
 test cmdAH-4.7 {Tcl_EncodingObjCmd} -returnCodes error -body {
     encoding convertfrom
-} -result {wrong # args: should be "encoding convertfrom ?encoding? data"}
+} -result {wrong # args: should be "encoding convertfrom ?-nocomplain? ?encoding? data"}
 test cmdAH-4.8 {Tcl_EncodingObjCmd} -returnCodes error -body {
     encoding convertfrom foo bar
 } -result {unknown encoding "foo"}
diff --git a/tests/encoding.test b/tests/encoding.test
index 75e0dcc..bf82493 100644
--- a/tests/encoding.test
+++ b/tests/encoding.test
@@ -22,6 +22,8 @@ catch {
     package require -exact tcl::test [info patchlevel]
 }
 
+testConstraint deprecated [expr {![info exists tcl_precision]}]
+
 proc toutf {args} {
     variable x
     lappend x "toutf $args"
@@ -297,7 +299,7 @@ test encoding-11.11 {encoding: extended Unicode UTF-32} {
 
 test encoding-12.1 {LoadTableEncoding: normal encoding} {
     set x [encoding convertto iso8859-3 Ġ]
-    append x [encoding convertto iso8859-3 Õ]
+    append x [encoding convertto -nocomplain iso8859-3 Õ]
     append x [encoding convertfrom iso8859-3 Õ]
 } "Õ?Ġ"
 test encoding-12.2 {LoadTableEncoding: single-byte encoding} {
@@ -346,67 +348,67 @@ test encoding-15.5 {UtfToUtfProc emoji character input} {
 } "4 😂"
 test encoding-15.6 {UtfToUtfProc emoji character output} {
     set x \uDE02\uD83D\uDE02\uD83D
-    set y [encoding convertto utf-8 \uDE02\uD83D\uDE02\uD83D]
+    set y [encoding convertto -nocomplain utf-8 \uDE02\uD83D\uDE02\uD83D]
     binary scan $y H* z
     list [string length $y] $z
 } {10 edb882f09f9882eda0bd}
 test encoding-15.7 {UtfToUtfProc emoji character output} {
     set x \uDE02\uD83D\uD83D
-    set y [encoding convertto utf-8 \uDE02\uD83D\uD83D]
+    set y [encoding convertto -nocomplain utf-8 \uDE02\uD83D\uD83D]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {3 9 edb882eda0bdeda0bd}
 test encoding-15.8 {UtfToUtfProc emoji character output} {
     set x \uDE02\uD83Dé
-    set y [encoding convertto utf-8 \uDE02\uD83Dé]
+    set y [encoding convertto -nocomplain utf-8 \uDE02\uD83Dé]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {3 8 edb882eda0bdc3a9}
 test encoding-15.9 {UtfToUtfProc emoji character output} {
     set x \uDE02\uD83DX
-    set y [encoding convertto utf-8 \uDE02\uD83DX]
+    set y [encoding convertto -nocomplain utf-8 \uDE02\uD83DX]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {3 7 edb882eda0bd58}
 test encoding-15.10 {UtfToUtfProc high surrogate character output} {
     set x \uDE02é
-    set y [encoding convertto utf-8 \uDE02é]
+    set y [encoding convertto -nocomplain utf-8 \uDE02é]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {2 5 edb882c3a9}
 test encoding-15.11 {UtfToUtfProc low surrogate character output} {
     set x \uDA02é
-    set y [encoding convertto utf-8 \uDA02é]
+    set y [encoding convertto -nocomplain utf-8 \uDA02é]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {2 5 eda882c3a9}
 test encoding-15.12 {UtfToUtfProc high surrogate character output} {
     set x \uDE02Y
-    set y [encoding convertto utf-8 \uDE02Y]
+    set y [encoding convertto -nocomplain utf-8 \uDE02Y]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {2 4 edb88259}
 test encoding-15.13 {UtfToUtfProc low surrogate character output} {
     set x \uDA02Y
-    set y [encoding convertto utf-8 \uDA02Y]
+    set y [encoding convertto -nocomplain utf-8 \uDA02Y]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {2 4 eda88259}
 test encoding-15.14 {UtfToUtfProc high surrogate character output} {
     set x \uDE02
-    set y [encoding convertto utf-8 \uDE02]
+    set y [encoding convertto -nocomplain utf-8 \uDE02]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {1 3 edb882}
 test encoding-15.15 {UtfToUtfProc low surrogate character output} {
     set x \uDA02
-    set y [encoding convertto utf-8 \uDA02]
+    set y [encoding convertto -nocomplain utf-8 \uDA02]
     binary scan $y H* z
     list [string length $x] [string length $y] $z
 } {1 3 eda882}
 test encoding-15.16 {UtfToUtfProc: Invalid 4-byte UTF-8, see [ed29806ba]} {
     set x \xF0\xA0\xA1\xC2
-    set y [encoding convertfrom utf-8 \xF0\xA0\xA1\xC2]
+    set y [encoding convertfrom -nocomplain utf-8 \xF0\xA0\xA1\xC2]
     list [string length $x] $y
 } "4 \xF0\xA0\xA1\xC2"
 test encoding-15.17 {UtfToUtfProc emoji character output} {
@@ -487,10 +489,10 @@ test encoding-17.2 {UtfToUcs2Proc} -body {
     encoding convertfrom utf-16 [encoding convertto ucs-2 "\U460DC"]
 } -result "\uFFFD"
 test encoding-17.3 {UtfToUtf16Proc} -body {
-    encoding convertto utf-16be "\uDCDC"
+    encoding convertto -nocomplain utf-16be "\uDCDC"
 } -result "\xFF\xFD"
 test encoding-17.4 {UtfToUtf16Proc} -body {
-    encoding convertto utf-16le "\uD8D8"
+    encoding convertto -nocomplain utf-16le "\uD8D8"
 } -result "\xFD\xFF"
 test encoding-17.5 {UtfToUtf16Proc} -body {
     encoding convertto utf-32le "\U460DC"
@@ -615,26 +617,62 @@ test encoding-24.4 {Parse valid or invalid utf-8} {
     string length [encoding convertfrom utf-8 "\xC0\x80"]
 } 1
 test encoding-24.5 {Parse valid or invalid utf-8} {
-    string length [encoding convertfrom utf-8 "\xC0\x81"]
+    string length [encoding convertfrom -nocomplain utf-8 "\xC0\x81"]
 } 2
 test encoding-24.6 {Parse valid or invalid utf-8} {
-    string length [encoding convertfrom utf-8 "\xC1\xBF"]
+    string length [encoding convertfrom -nocomplain utf-8 "\xC1\xBF"]
 } 2
 test encoding-24.7 {Parse valid or invalid utf-8} {
     string length [encoding convertfrom utf-8 "\xC2\x80"]
 } 1
 test encoding-24.8 {Parse valid or invalid utf-8} {
-    string length [encoding convertfrom utf-8 "\xE0\x80\x80"]
+    string length [encoding convertfrom -nocomplain utf-8 "\xE0\x80\x80"]
 } 3
 test encoding-24.9 {Parse valid or invalid utf-8} {
-    string length [encoding convertfrom utf-8 "\xE0\x9F\xBF"]
+    string length [encoding convertfrom -nocomplain utf-8 "\xE0\x9F\xBF"]
 } 3
 test encoding-24.10 {Parse valid or invalid utf-8} {
     string length [encoding convertfrom utf-8 "\xE0\xA0\x80"]
 } 1
 test encoding-24.11 {Parse valid or invalid utf-8} {
-    string length [encoding convertfrom utf-8 "\xEF\xBF\xBF"]
+    string length [encoding convertfrom -nocomplain utf-8 "\xEF\xBF\xBF"]
+} 1
+test encoding-24.12 {Parse valid or invalid utf-8} -constraints deprecated -body {
+    encoding convertfrom utf-8 "\xC0\x81"
+} -returnCodes 1 -result {unexpected byte sequence starting at index 0: '\xC0'}
+test encoding-24.13 {Parse valid or invalid utf-8} -constraints deprecated -body {
+    encoding convertfrom utf-8 "\xC1\xBF"
+} -returnCodes 1 -result {unexpected byte sequence starting at index 0: '\xC1'}
+test encoding-24.14 {Parse valid or invalid utf-8} {
+    string length [encoding convertfrom utf-8 "\xC2\x80"]
+} 1
+test encoding-24.15 {Parse valid or invalid utf-8} -constraints deprecated -body {
+    encoding convertfrom utf-8 "Z\xE0\x80"
+} -returnCodes 1 -result {unexpected byte sequence starting at index 1: '\xE0'}
+test encoding-24.16 {Parse valid or invalid utf-8} -constraints {testbytestring deprecated} -body {
+    encoding convertto utf-8 [testbytestring "Z\u4343\x80"]
+} -returnCodes 1 -result {expected byte sequence but character 1 was '䍃€' (U+004343)}
+test encoding-24.17 {Parse valid or invalid utf-8} -constraints {testbytestring deprecated} -body {
+    encoding convertto utf-8 [testbytestring "Z\xE0\x80"]
+} -result "Z\xC3\xA0\xE2\x82\xAC"
+test encoding-24.18 {Parse valid or invalid utf-8} -constraints {testbytestring deprecated} -body {
+    encoding convertto utf-8 [testbytestring "Z\xE0\x80xxxxxx"]
+} -result "Z\xC3\xA0\xE2\x82\xACxxxxxx"
+test encoding-24.19 {Parse valid or invalid utf-8} -constraints deprecated -body {
+    encoding convertto utf-8 "ZX\uD800"
+} -returnCodes 1 -match glob -result "unexpected character at index 2: 'U+00D800'"
+test encoding-24.20 {Parse with -nocomplain but without providing encoding} {
+    string length [encoding convertfrom -nocomplain "\x20"]
+} 1
+test encoding-24.21 {Parse with -nocomplain but without providing encoding} {
+    string length [encoding convertto -nocomplain "\x20"]
 } 1
+test encoding-24.22 {Syntax error, two encodings} -body {
+    encoding convertfrom iso8859-1 utf-8 "ZX\uD800"
+} -returnCodes 1 -result {wrong # args: should be "::tcl::encoding::convertfrom ?-nocomplain? ?encoding? data"}
+test encoding-24.23 {Syntax error, two encodings} -body {
+    encoding convertto iso8859-1 utf-8 "ZX\uD800"
+} -returnCodes 1 -result {wrong # args: should be "::tcl::encoding::convertto ?-nocomplain? ?encoding? data"}
 
 file delete [file join [temporaryDirectory] iso2022.txt]
 
@@ -790,7 +828,7 @@ test encoding-28.0 {all encodings load} -body {
 	set string hello
 	foreach name [encoding names] {
 		incr count
-		encoding convertto $name $string
+		encoding convertto -nocomplain $name $string
 
 		# discard the cached internal representation of Tcl_Encoding
 		# Unfortunately, without this, encoding 2-1 fails.
diff --git a/tests/http.test b/tests/http.test
index 2fd5af4..93998fe 100644
--- a/tests/http.test
+++ b/tests/http.test
@@ -661,7 +661,7 @@ test http-7.3 {http::formatQuery} -setup {
 } -cleanup {
     http::config -urlencoding $enc
 } -result "can't read \"formMap(∈)\": no such element in array"
-test http-7.4 {http::formatQuery} -setup {
+test http-7.4 {http::formatQuery} -constraints deprecated -setup {
     set enc [http::config -urlencoding]
 } -body {
     # this would be reverting to http <=2.4 behavior w/o errors
diff --git a/tests/io.test b/tests/io.test
index 0ef3422..821b11e 100644
--- a/tests/io.test
+++ b/tests/io.test
@@ -268,7 +268,7 @@ test io-3.3 {WriteChars: compatibility with WriteBytes: flush on line} {
     close $f
     set x
 } "\r\n12"
-test io-3.4 {WriteChars: loop over stage buffer} {
+test io-3.4 {WriteChars: loop over stage buffer} deprecated {
     # stage buffer maps to more than can be queued at once.
 
     set f [open $path(test1) w]
@@ -278,7 +278,7 @@ test io-3.4 {WriteChars: loop over stage buffer} {
     close $f
     lappend x [contents $path(test1)]
 } [list "!)!)!)!)!)!)!)!)" "!)!)!)!)!)!)!)!)!)!)!)!)!)!)!)"]
-test io-3.5 {WriteChars: saved != 0} {
+test io-3.5 {WriteChars: saved != 0} deprecated {
     # Bytes produced by UtfToExternal from end of last channel buffer
     # had to be moved to beginning of next channel buffer to preserve
     # requested buffersize.
@@ -307,7 +307,7 @@ test io-3.6 {WriteChars: (stageRead + dstWrote == 0)} {
     close $f
     lappend x [contents $path(test1)]
 } [list "12345678901234\x82\x60" "12345678901234\x82\x60\x82\x61"]
-test io-3.7 {WriteChars: (bufPtr->nextAdded > bufPtr->length)} {
+test io-3.7 {WriteChars: (bufPtr->nextAdded > bufPtr->length)} deprecated {
     # When translating UTF-8 to external, the produced bytes went past end
     # of the channel buffer.  This is done purpose -- we then truncate the
     # bytes at the end of the partial character to preserve the requested
@@ -1532,7 +1532,7 @@ test io-12.8 {ReadChars: multibyte chars split} {
     close $f
     scan [string index $in end] %c
 } 160
-test io-12.9 {ReadChars: multibyte chars split} {
+test io-12.9 {ReadChars: multibyte chars split} deprecated {
     set f [open $path(test1) w]
     fconfigure $f -translation binary
     puts -nonewline $f [string repeat a 9]\xC2
@@ -1543,7 +1543,7 @@ test io-12.9 {ReadChars: multibyte chars split} {
     close $f
     scan [string index $in end] %c
 } 194
-test io-12.10 {ReadChars: multibyte chars split} {
+test io-12.10 {ReadChars: multibyte chars split} deprecated {
     set f [open $path(test1) w]
     fconfigure $f -translation binary
     puts -nonewline $f [string repeat a 9]\xC2
diff --git a/tests/main.test b/tests/main.test
index 2d3f63c..ef58b6f 100644
--- a/tests/main.test
+++ b/tests/main.test
@@ -143,7 +143,7 @@ namespace eval ::tcl::test::main {
     test Tcl_Main-1.8 {
 	Tcl_Main: startup script - -encoding option - mismatched encodings
     } -constraints {
-	stdio
+	stdio deprecated
     } -setup {
 	set script [makeFile {} script]
 	file delete $script
diff --git a/tests/safe.test b/tests/safe.test
index 773b16f..5f3eae8 100644
--- a/tests/safe.test
+++ b/tests/safe.test
@@ -1269,7 +1269,7 @@ test safe-11.7 {testing safe encoding} -setup {
     interp eval $i encoding convertfrom
 } -returnCodes error -cleanup {
     safe::interpDelete $i
-} -result {wrong # args: should be "encoding convertfrom ?encoding? data"}
+} -result {wrong # args: should be "encoding convertfrom ?-nocomplain? ?encoding? data"}
 test safe-11.7.1 {testing safe encoding} -setup {
     set i [safe::interpCreate]
 } -body {
@@ -1278,7 +1278,7 @@ test safe-11.7.1 {testing safe encoding} -setup {
 } -returnCodes ok -match glob -cleanup {
     unset -nocomplain m o
     safe::interpDelete $i
-} -result {wrong # args: should be "encoding convertfrom ?encoding? data"
+} -result {wrong # args: should be "encoding convertfrom ?-nocomplain? ?encoding? data"
     while executing
 "encoding convertfrom"
     invoked from within
@@ -1291,7 +1291,7 @@ test safe-11.8 {testing safe encoding} -setup {
     interp eval $i encoding convertto
 } -returnCodes error -cleanup {
     safe::interpDelete $i
-} -result {wrong # args: should be "encoding convertto ?encoding? data"}
+} -result {wrong # args: should be "encoding convertto ?-nocomplain? ?encoding? data"}
 test safe-11.8.1 {testing safe encoding} -setup {
     set i [safe::interpCreate]
 } -body {
@@ -1300,7 +1300,7 @@ test safe-11.8.1 {testing safe encoding} -setup {
 } -returnCodes ok -match glob -cleanup {
     unset -nocomplain m o
     safe::interpDelete $i
-} -result {wrong # args: should be "encoding convertto ?encoding? data"
+} -result {wrong # args: should be "encoding convertto ?-nocomplain? ?encoding? data"
     while executing
 "encoding convertto"
     invoked from within
diff --git a/tests/source.test b/tests/source.test
index eee03ec..d67bec5 100644
--- a/tests/source.test
+++ b/tests/source.test
@@ -275,7 +275,7 @@ test source-7.5 {source -encoding: correct operation} -setup {
     removeFile source.file
     rename € {}
 } -result foo
-test source-7.6 {source -encoding: mismatch encoding error} -setup {
+test source-7.6 {source -encoding: mismatch encoding error} -constraints deprecated -setup {
     set sourcefile [makeFile {} source.file]
     file delete $sourcefile
     set f [open $sourcefile w]