From 92eea63a7df5071fddef9b89b9122a5761670908 Mon Sep 17 00:00:00 2001 From: dkf Date: Sat, 14 Sep 2013 07:07:03 +0000 Subject: [2152292] Corrected implementation of uuencoding. --- doc/binary.n | 17 +++---- generic/tclBinary.c | 125 ++++++++++++++++++++++++++++++++++++++++++++++++++-- tests/binary.test | 43 ++++++++++++------ 3 files changed, 158 insertions(+), 27 deletions(-) diff --git a/doc/binary.n b/doc/binary.n index a40afe6..95be36e 100644 --- a/doc/binary.n +++ b/doc/binary.n @@ -96,25 +96,22 @@ largely superseded by the \fBbase64\fR binary encoding. .RS .PP During encoding, the following options are supported: -'\" This is wrong! The uuencode format had more complexity than this! .TP \fB\-maxlen \fIlength\fR . Indicates that the output should be split into lines of no more than -\fIlength\fR characters. By default, lines are not split. -.TP -\fB\-wrapchar \fIcharacter\fR -. -Indicates that, when lines are split because of the \fB\-maxlen\fR option, -\fIcharacter\fR should be used to separate lines. By default, this is a -newline character, -.QW \en . +\fIlength\fR characters. By default, lines are split every 61 characters, and +this must be in the range 3 to 85 due to limitations in the encoding. .PP During decoding, the following options are supported: .TP \fB\-strict\fR . -Instructs the decoder to throw an error if it encounters whitespace characters. Otherwise it ignores them. +Instructs the decoder to throw an error if it encounters whitespace +characters. Otherwise it ignores them. +.PP +Note that neither the encoder nor the decoder handle the header and footer of +the uuencode format. .RE .VE 8.6 .SH "BINARY FORMAT" diff --git a/generic/tclBinary.c b/generic/tclBinary.c index 7625d39..71da309 100644 --- a/generic/tclBinary.c +++ b/generic/tclBinary.c @@ -87,10 +87,13 @@ static int BinaryDecodeHex(ClientData clientData, static int BinaryEncode64(ClientData clientData, Tcl_Interp *interp, int objc, Tcl_Obj *const objv[]); -static int BinaryDecodeUu(ClientData clientData, +static int BinaryDecode64(ClientData clientData, Tcl_Interp *interp, int objc, Tcl_Obj *const objv[]); -static int BinaryDecode64(ClientData clientData, +static int BinaryEncodeUu(ClientData clientData, + Tcl_Interp *interp, int objc, + Tcl_Obj *const objv[]); +static int BinaryDecodeUu(ClientData clientData, Tcl_Interp *interp, int objc, Tcl_Obj *const objv[]); @@ -140,7 +143,7 @@ static const EnsembleImplMap binaryMap[] = { }; static const EnsembleImplMap encodeMap[] = { { "hex", BinaryEncodeHex, TclCompileBasic1ArgCmd, NULL, (ClientData)HexDigits, 0 }, - { "uuencode", BinaryEncode64, NULL, NULL, (ClientData)UueDigits, 0 }, + { "uuencode", BinaryEncodeUu, NULL, NULL, NULL, 0 }, { "base64", BinaryEncode64, NULL, NULL, (ClientData)B64Digits, 0 }, { NULL, NULL, NULL, NULL, NULL, 0 } }; @@ -2439,7 +2442,7 @@ BinaryDecodeHex( * This implements a generic 6 bit binary encoding. Input is broken into * 6 bit chunks and a lookup table passed in via clientData is used to * turn these values into output characters. This is used to implement - * base64 and uuencode binary encodings. + * base64 binary encodings. * * Results: * Interp result set to an encoded byte array object @@ -2498,6 +2501,12 @@ BinaryEncode64( if (Tcl_GetIntFromObj(interp, objv[i+1], &maxlen) != TCL_OK) { return TCL_ERROR; } + if (maxlen < 0) { + Tcl_SetResult(interp, "line length out of range", TCL_STATIC); + Tcl_SetErrorCode(interp, "TCL", "BINARY", "ENCODE", + "LINE_LENGTH", NULL); + return TCL_ERROR; + } break; case OPT_WRAPCHAR: wrapchar = Tcl_GetStringFromObj(objv[i+1], &wrapcharlen); @@ -2550,6 +2559,114 @@ BinaryEncode64( /* *---------------------------------------------------------------------- * + * BinaryEncodeUu -- + * + * This implements the uuencode binary encoding. Input is broken into 6 + * bit chunks and a lookup table is used to turn these values into output + * characters. This differs from the generic code above in that line + * lengths are also encoded. + * + * Results: + * Interp result set to an encoded byte array object + * + * Side effects: + * None + * + *---------------------------------------------------------------------- + */ + +static int +BinaryEncodeUu( + ClientData clientData, + Tcl_Interp *interp, + int objc, + Tcl_Obj *const objv[]) +{ + Tcl_Obj *resultObj; + unsigned char *data, *start, *cursor; + int offset, count, rawLength, n, i, bits, index; + int lineLength = 61; + enum {OPT_MAXLEN}; + static const char *const optStrings[] = { "-maxlen", NULL }; + + if (objc < 2 || objc%2 != 0) { + Tcl_WrongNumArgs(interp, 1, objv, "?-maxlen len? data"); + return TCL_ERROR; + } + for (i = 1; i < objc-1; i += 2) { + if (Tcl_GetIndexFromObj(interp, objv[i], optStrings, "option", + TCL_EXACT, &index) != TCL_OK) { + return TCL_ERROR; + } + switch (index) { + case OPT_MAXLEN: + if (Tcl_GetIntFromObj(interp, objv[i+1], &lineLength) != TCL_OK) { + return TCL_ERROR; + } + if (lineLength < 3 || lineLength > 85) { + Tcl_SetResult(interp, "line length out of range", TCL_STATIC); + Tcl_SetErrorCode(interp, "TCL", "BINARY", "ENCODE", + "LINE_LENGTH", NULL); + return TCL_ERROR; + } + break; + } + } + + /* + * Allocate the buffer. This is a little bit too long, but is "good + * enough". + */ + + resultObj = Tcl_NewObj(); + offset = 0; + data = Tcl_GetByteArrayFromObj(objv[objc-1], &count); + rawLength = (lineLength - 1) * 3 / 4; + start = cursor = Tcl_SetByteArrayLength(resultObj, + (lineLength + 1) * ((count + (rawLength - 1)) / rawLength)); + n = bits = 0; + + /* + * Encode the data. Each output line first has the length of raw data + * encoded by the output line described in it by one encoded byte, then + * the encoded data follows (encoding each 6 bits as one character). + * Encoded lines are always terminated by a newline. + */ + + while (offset < count) { + int lineLen = count - offset; + + if (lineLen > rawLength) { + lineLen = rawLength; + } + *cursor++ = UueDigits[lineLen]; + for (i=0 ; i 6 ; bits -= 6) { + *cursor++ = UueDigits[(n >> (bits-6)) & 0x3f]; + } + } + if (bits > 0) { + n <<= 8; + *cursor++ = UueDigits[(n >> (bits + 2)) & 0x3f]; + bits = 0; + } + *cursor++ = '\n'; + } + + /* + * Fix the length of the output bytearray. + */ + + Tcl_SetByteArrayLength(resultObj, cursor-start); + Tcl_SetObjResult(interp, resultObj); + return TCL_OK; +} + +/* + *---------------------------------------------------------------------- + * * BinaryDecodeUu -- * * Decode a uuencoded string. diff --git a/tests/binary.test b/tests/binary.test index d424837..4b71b77 100644 --- a/tests/binary.test +++ b/tests/binary.test @@ -2705,34 +2705,51 @@ test binary-74.1 {binary encode uuencode} -body { } -returnCodes error -match glob -result "wrong # args: *" test binary-74.2 {binary encode uuencode} -body { binary encode uuencode abc -} -result {86)C} +} -result {#86)C +} test binary-74.3 {binary encode uuencode} -body { binary encode uuencode {} } -result {} test binary-74.4 {binary encode uuencode} -body { binary encode uuencode [string repeat abc 20] -} -result [string repeat 86)C 20] +} -result "M[string repeat 86)C 15]\n/[string repeat 86)C 5]\n" test binary-74.5 {binary encode uuencode} -body { binary encode uuencode \0\1\2\3\4\0\1\2\3 -} -result "``\$\"`P0``0(#" +} -result ")``\$\"`P0``0(#\n" test binary-74.6 {binary encode uuencode} -body { binary encode uuencode \0 -} -result {````} +} -result {!`` +} test binary-74.7 {binary encode uuencode} -body { binary encode uuencode \0\0 -} -result {````} +} -result "\"``` +" test binary-74.8 {binary encode uuencode} -body { binary encode uuencode \0\0\0 -} -result {````} +} -result {#```` +} test binary-74.9 {binary encode uuencode} -body { binary encode uuencode \0\0\0\0 -} -result {````````} -test binary-74.10 {binary encode uuencode} -body { - binary encode uuencode -maxlen 0 -wrapchar | abcabcabc -} -result {86)C86)C86)C} -test binary-74.11 {binary encode uuencode} -body { - binary encode uuencode -maxlen 1 -wrapchar | abcabcabc -} -result {8|6|)|C|8|6|)|C|8|6|)|C} +} -result {$`````` +} +test binary-74.10 {binary encode uuencode} -returnCodes error -body { + binary encode uuencode -maxlen 30 -wrapchar | abcabcabc +} -result {bad option "-wrapchar": must be -maxlen} +test binary-74.11 {binary encode uuencode} -returnCodes error -body { + binary encode uuencode -maxlen 1 abcabcabc +} -result {line length out of range} +test binary-74.12 {binary encode uuencode} -body { + binary encode uuencode -maxlen 3 abcabcabc +} -result {!80 +!8@ +!8P +!80 +!8@ +!8P +!80 +!8@ +!8P +} test binary-75.1 {binary decode uuencode} -body { binary decode uuencode -- cgit v0.12