Index: Configure.pl =================================================================== RCS file: /cvs/public/parrot/Configure.pl,v retrieving revision 1.139 diff -u -b -r1.139 Configure.pl --- Configure.pl 9 Apr 2004 20:31:51 -0000 1.139 +++ Configure.pl 13 Apr 2004 08:16:17 -0000 @@ -54,7 +54,7 @@ You can add and remove option values with C<< :rem{} >> and C<< :add{} >>. For example: - perl Configure.pl --ccflags="rem{-g} :add{-O2}" + perl Configure.pl --ccflags=":rem{-g} :add{-O2}" =over @@ -167,6 +167,10 @@ be one of: C, C, C or C. The default is C. +=item C<--icudatadir=(path)> + +Use the given directory to locate ICU's data file(s) + =back Other Options (may not be implemented) @@ -274,6 +278,8 @@ --execcapable Use JIT to emit a native executable --gc=(type) Determine the type of garbage collection type=(gc|libc|malloc|malloc-trace) default is gc + + --icudatadir=(path) Use the given directory to locate ICU's data file(s) Other Options (may not be implemented): Index: config/gen/icu.pl =================================================================== RCS file: /cvs/public/parrot/config/gen/icu.pl,v retrieving revision 1.7 diff -u -b -r1.7 icu.pl --- config/gen/icu.pl 12 Apr 2004 20:36:36 -0000 1.7 +++ config/gen/icu.pl 13 Apr 2004 08:16:17 -0000 @@ -20,10 +20,17 @@ $description="Configuring ICU if requested..."; -@args=qw(buildicu verbose); +@args=qw(buildicu verbose icudatadir); sub runstep { - my ($buildicu, $verbose) = @_; + my ($buildicu, $verbose, $icudatadir) = @_; + + if( !defined $icudatadir ) + { + $icudatadir = 'blib/lib/icu/2.6.1'; + } + + Configure::Data->set( icudatadir => $icudatadir ); # unless ($buildicu) { # print " [Skipped] " if $verbose; Index: config/gen/config_h/config_h.in =================================================================== RCS file: /cvs/public/parrot/config/gen/config_h/config_h.in,v retrieving revision 1.21 diff -u -b -r1.21 config_h.in --- config/gen/config_h/config_h.in 10 Apr 2004 09:49:15 -0000 1.21 +++ config/gen/config_h/config_h.in 13 Apr 2004 08:16:17 -0000 @@ -134,6 +134,8 @@ #define PARROT_CORE_CG_OPLIB_INIT Parrot_DynOp_core_cg_${MAJOR}_${MINOR}_${PATCH} #define PARROT_CORE_CGP_OPLIB_INIT Parrot_DynOp_core_cgp_${MAJOR}_${MINOR}_${PATCH} +#define DEFAULT_ICU_DATA_DIR "${icudatadir}" + #define INTVAL_FMT "${intvalfmt}" #define FLOATVAL_FMT "${floatvalfmt}" Index: include/parrot/string_primitives.h =================================================================== RCS file: /cvs/public/parrot/include/parrot/string_primitives.h,v retrieving revision 1.1 diff -u -b -r1.1 string_primitives.h --- include/parrot/string_primitives.h 9 Apr 2004 20:32:24 -0000 1.1 +++ include/parrot/string_primitives.h 13 Apr 2004 08:16:18 -0000 @@ -23,6 +23,10 @@ void string_fill_from_buffer(struct Parrot_Interp *interpreter, const void *buffer, UINTVAL len, const char *encoding_name, STRING *s); +/* Utility method which knows how to uwind a single escape sequence */ +Parrot_UInt4 string_unescape_single_sequence( Parrot_UInt4 *offset, + Parrot_UInt4 input_length, char *cstring); + UINTVAL Parrot_char_digit_value(struct Parrot_Interp *interpreter, UINTVAL character); Index: src/string.c =================================================================== RCS file: /cvs/public/parrot/src/string.c,v retrieving revision 1.185 diff -u -b -r1.185 string.c --- src/string.c 11 Apr 2004 13:14:07 -0000 1.185 +++ src/string.c 13 Apr 2004 08:16:24 -0000 @@ -243,8 +243,9 @@ void string_init(void) { -/* XXXX: pull out into a config */ - string_set_data_directory("blib/lib/icu/2.6.1"); +/* DEFAULT_ICU_DATA_DIR is configured at build time. Need a way to + specify this at runtime as well. */ + string_set_data_directory(DEFAULT_ICU_DATA_DIR); /* encoding_init(); chartype_init(); @@ -3053,6 +3054,7 @@ */ +#if 0 STRING * string_unescape_cstring(struct Parrot_Interp * interpreter, char *cstring, char delimiter) @@ -3166,6 +3168,85 @@ result->bufused = p - (char *)result->strstart; string_compute_strlen(interpreter, result); + + return string_constant_copy(interpreter, result); +} +#endif + +STRING * +string_unescape_cstring(struct Parrot_Interp *interpreter, + char *cstring, char delimiter) +{ + char *p, *string; + STRING *result; + size_t clength = strlen(cstring); /* is this ok if cstring is NULL? */ + + if( !cstring || !clength ) return NULL; + + result = string_make(interpreter, cstring, clength, "iso-8859-1", 0); + + for (p = (char *)result->strstart, string = cstring ; *string; ) + { + if (*string == '\\' && string[1]) + { + Parrot_UInt4 c32; + Parrot_UInt4 lenParsed = 0; + + ++string; /* get past the backslash */ + c32 = string_unescape_single_sequence(&lenParsed, + cstring + clength - string, string); + + if( lenParsed == 0 ) + { + /* error */ + } + else + { + string += lenParsed; /* this had better not go past the terminating null */ + + if( result->representation == enum_stringrep_one ) + { + if( c32 <= 0xFF ) + { + *p++ = c32; + } + else + { + /* finish up the string so far */ + result->bufused = p - (char *)result->strstart; + string_compute_strlen(interpreter, result); + + string_append_chr(interpreter, result, c32); + } + } + else + { + string_append_chr(interpreter, result, c32); + } + } + } + else if (*string == delimiter) + { + break; + } + else + { + if( result->representation == enum_stringrep_one ) + { + *p++ = *string++; + } + else + { + string_append_chr(interpreter, result, *string++); + } + } + } + + if( result->representation == enum_stringrep_one ) + { + result->bufused = p - (char *)result->strstart; + string_compute_strlen(interpreter, result); + } return string_constant_copy(interpreter, result); } Index: src/string_primitives.c =================================================================== RCS file: /cvs/public/parrot/src/string_primitives.c,v retrieving revision 1.4 diff -u -b -r1.4 string_primitives.c --- src/string_primitives.c 11 Apr 2004 13:14:07 -0000 1.4 +++ src/string_primitives.c 13 Apr 2004 08:16:25 -0000 @@ -23,6 +23,7 @@ #include #include #include +#include #include /* @@ -41,6 +42,19 @@ string_set_data_directory(const char *dir) { u_setDataDirectory(dir); + + /* Since u_setDataDirectory doesn't have a result code, we'll spot + check that everything is okay by making sure that '9' had decimal + value 9. Using 57 rather than '9' so that the encoding of this + source code file isn't an issue.... (Don't want to get bitten by + EBCDIC.) */ + + if( !u_isdigit(57) || (u_charDigitValue(57) != 9) ) + { + internal_exception(ICU_ERROR, + "string_set_data_directory: ICU data files not found" + "(apparently) for directory [%s]", dir); + } } /* @@ -125,82 +139,22 @@ string_compute_strlen(interpreter, s); } -/* -{ - - UErrorCode myError = U_ZERO_ERROR; - UConverter *conv = ucnv_open("ISO-8859-1", &myError); - - NSData *data = [NSData dataWithContentsOfFile:@"/var/tmp/ja.txt"]; - - - UConverter *conv2 = ucnv_open("UTF-8", &myError); - UChar *outputBuffer = malloc(50*sizeof(UChar)); - UChar *start = outputBuffer; - char *source = [data bytes]; - UErrorCode convError = 0; - - ucnv_toUnicode(conv2, &outputBuffer, &(outputBuffer[50]), &source, source + [data length], NULL, TRUE, &convError); - - NSLog(@"conv error = %i", convError); - NSLog(@"conv length = %i", outputBuffer - start); - - - UConverter *conv3 = ucnv_open("UCS-2", &myError); //ISO-8859-1" - char *outbuffer = malloc(50); - char *outstart = outbuffer; - UErrorCode redoErr = 0; - - NSLog(@"conv3 = %p", conv3); - NSLog(@"%s", ucnv_getName(conv3, &redoErr)); - - UErrorCode callbackStatus = 0; - - ucnv_setFromUCallBack(conv3, - UCNV_FROM_U_CALLBACK_STOP, //UCNV_FROM_U_CALLBACK_ESCAPE - NULL, - NULL, - NULL, - &callbackStatus); - - ucnv_fromUnicode(conv3, &outbuffer, outbuffer + 50, &start, outputBuffer, NULL, TRUE, &redoErr); -// ucnv_fromUnicode(conv3, NULL, NULL, &start, outputBuffer, NULL, TRUE, &redoErr); - - NSLog(@"redo error = %i", redoErr); - NSLog(@"%s", u_errorName(redoErr)); - NSLog(@"redo length = %i", outbuffer - outstart); - -NSLog(@"%x", *outstart); -NSLog(@"%d", *outstart); -NSLog(@"%x", outstart[1]); - NSLog(@"%@", [[NSString alloc] initWithData:[NSData dataWithBytes:outstart length:(outbuffer - outstart)] encoding:NSISOLatin1StringEncoding]); +static Parrot_UInt2 +_string_char_at_callback(Parrot_Int4 offset, void *context); +static Parrot_UInt2 +_string_char_at_callback(Parrot_Int4 offset, void *context) { - UConverter *jisConv = ucnv_open("shift_jis", &myError); - USet *set = uset_open(0, 0); - UErrorCode setError = U_ZERO_ERROR; - - ucnv_getUnicodeSet( jisConv, set, UCNV_ROUNDTRIP_SET, &setError); - - NSLog(@"set size = %d", uset_size(set)); - NSLog(@"contains range = %d", uset_containsRange(set, 0x10000, 0x10ffff)); - NSLog(@"uset_containsNoneOfRange = %d", uset_containsNoneOfRange(set, 0x10000, 0x10ffff)); + return *((char*)context + offset); } + +Parrot_UInt4 +string_unescape_single_sequence( Parrot_UInt4 *offset, + Parrot_UInt4 input_length, char *cstring) { - UConverter *utf8Conv = ucnv_open("UTF-8", &myError); - USet *set = uset_open(0, 0); - UErrorCode setError = U_ZERO_ERROR; - - ucnv_getUnicodeSet( utf8Conv, set, UCNV_ROUNDTRIP_SET, &setError); - - NSLog(@"set size = %d", uset_size(set)); - NSLog(@"contains range = %d", uset_containsRange(set, 0x10000, 0x10ffff)); - NSLog(@"uset_containsNoneOfRange = %d", uset_containsNoneOfRange(set, 0x10000, 0x10ffff)); -} - [pool release]; - return 0; + return u_unescapeAt(_string_char_at_callback, offset, input_length, + cstring); } -*/ /*