Fix for [a7a89d422a4f5dd3], Under strict encoding, [gets] returns an error instead of returning the second line.

Also, ensure that position in file does not change if [gets] return an error.

2 files changed, 20 insertions, 15 deletions

diff --git a/generic/tclIO.c b/generic/tclIO.c
index aa82f89..3fd8d48 100644
--- a/generic/tclIO.c
+++ b/generic/tclIO.c
@@ -4616,6 +4616,7 @@ Tcl_GetsObj(
 
     if (GotFlag(statePtr, CHANNEL_ENCODING_ERROR)) {
 	UpdateInterest(chanPtr);
+	ResetFlag(statePtr, CHANNEL_ENCODING_ERROR);
 	Tcl_SetErrno(EILSEQ);
 	return TCL_INDEX_NONE;
     }
@@ -4632,7 +4633,7 @@ Tcl_GetsObj(
     if (GotFlag(statePtr, CHANNEL_STICKY_EOF)) {
 	SetFlag(statePtr, CHANNEL_EOF);
 	assert(statePtr->inputEncodingFlags & TCL_ENCODING_END);
-	assert(!GotFlag(statePtr, CHANNEL_BLOCKED|INPUT_SAW_CR));
+assert(!GotFlag(statePtr, CHANNEL_BLOCKED|INPUT_SAW_CR));
 
 	/* TODO: Do we need this? */
 	UpdateInterest(chanPtr);
@@ -4998,9 +4999,11 @@ Tcl_GetsObj(
     UpdateInterest(chanPtr);
     TclChannelRelease((Tcl_Channel)chanPtr);
     if (GotFlag(statePtr, CHANNEL_ENCODING_ERROR) && gs.bytesWrote == 0) {
+	bufPtr->nextRemoved = oldRemoved;
 	Tcl_SetErrno(EILSEQ);
 	copiedTotal = -1;
     }
+    ResetFlag(statePtr, CHANNEL_ENCODING_ERROR);
     return copiedTotal;
 }
 
@@ -5462,6 +5465,8 @@ FilterInputBytes(
 
 	if (result == TCL_CONVERT_UNKNOWN || result == TCL_CONVERT_SYNTAX) {
 	    SetFlag(statePtr, CHANNEL_ENCODING_ERROR);
+	    ResetFlag(statePtr, CHANNEL_STICKY_EOF);
+	    ResetFlag(statePtr, CHANNEL_EOF);
 	    result = TCL_OK;
 	}
 
@@ -7589,7 +7594,7 @@ Tcl_InputBuffered(
     }
 
     /*
-     * Don't forget the bytes in the topmost pushback area.
+     * Remember the bytes in the topmost pushback area.
      */
 
     for (bufPtr = statePtr->topChanPtr->inQueueHead; bufPtr != NULL;
diff --git a/tests/io.test b/tests/io.test
index 286a136..7a35e0e 100644
--- a/tests/io.test
+++ b/tests/io.test
@@ -9266,13 +9266,14 @@ test io-75.6 {invalid utf-8 encoding, gets is not ignored (-profile strict)} -se
     set fn [makeFile {} io-75.6]
     set f [open $fn w+]
     fconfigure $f -encoding binary
-    # \x81 is invalid in utf-8
+    # \x81 is an incomplete byte sequence in utf-8
     puts -nonewline $f A\x81
     flush $f
     seek $f 0
     fconfigure $f -encoding utf-8 -buffering none -eofchar {} \
 	-translation lf -profile strict
 } -body {
+after 1
     gets $f
 } -cleanup {
     close $f
@@ -9452,13 +9453,14 @@ test io-75.14 {
     set chan  [file tempfile]
     fconfigure $f -encoding binary
     # \xc0\n is an invalid utf-8 sequence
-    puts -nonewline $f a\nb\xc0\nc\n
+    puts -nonewline $f a\nb\nc\xc0\nd\n
     flush $f
     seek $f 0
     fconfigure $f -encoding utf-8 -buffering none -eofchar {} \
 	-translation lf -profile strict
 } -body {
     lappend res [gets $f]
+    lappend res [gets $f]
     set status [catch {gets $f} cres copts]
     lappend res $status $cres
     chan configure $f -profile tcl8 
@@ -9466,36 +9468,34 @@ test io-75.14 {
     lappend res [gets $f]
     close $f
     return $res
-} -match glob -result {a 1 {error reading "*":\
-    invalid or incomplete multibyte or wide character} bÀ c}
+} -match glob -result {a b 1 {error reading "*":\
+    invalid or incomplete multibyte or wide character} cÀ d}
 
 test io-75.15 {
-    invalid utf-8 encoding strict gets does not hang
-    after error reconfigure and read
+    invalid utf-8 encoding strict
+    gets does not hang
+    gets succeeds for the first two lines
 } -setup {
     set res {}
     set chan [file tempfile]
     fconfigure $chan -encoding binary
     # \xc0\x40 is an invalid utf-8 sequence
-    puts $chan hello\nAB\xc0\x40CD\nEFG
+    puts $chan hello\nAB\nCD\xc0\x40EF\nGHI
 	seek $chan 0
 } -body {
     #Now try to read it with [gets]
     fconfigure $chan -encoding utf-8 -profile strict
     lappend res [gets $chan]
+    lappend res [gets $chan]
     set status [catch {gets $chan} cres copts]
     lappend res $status $cres
     set status [catch {gets $chan} cres copts]
     lappend res $status $cres
-    chan configure $chan -encoding binary
-    foreach char [split [read $chan 6] {}] {
-	lappend res [format %x [scan $char %c]]
-    }
     return $res
 } -cleanup {
     close $chan
-} -match glob -result {hello 1 {error reading "*": invalid or incomplete multibyte or wide character}\
-    1 {error reading "*": invalid or incomplete multibyte or wide character} 41 42 c0 40 43 44}
+} -match glob -result {hello AB 1 {error reading "*": invalid or incomplete multibyte or wide character}\
+    1 {error reading "*": invalid or incomplete multibyte or wide character}}
 
 # ### ### ### ######### ######### #########