/*
*******************************************************************************
*
- * Copyright (C) 2003, International Business Machines
+ * Copyright (C) 2003-2004, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
#include "unicode/uidna.h"
#include "unicode/ustring.h"
-#include "strprep.h"
+#include "unicode/usprep.h"
#include "punycode.h"
#include "ustr_imp.h"
#include "cmemory.h"
#define CAPITAL_Z 0x005A
#define LOWER_CASE_DELTA 0x0020
#define FULL_STOP 0x002E
+#define DATA_FILE_NAME "uidna"
inline static UChar
toASCIILower(UChar ch){
}
-U_CAPI int32_t U_EXPORT2
-uidna_toASCII(const UChar* src, int32_t srcLength,
- UChar* dest, int32_t destCapacity,
- int32_t options,
- UParseError* parseError,
- UErrorCode* status){
-
- if(status == NULL || U_FAILURE(*status)){
- return 0;
+/**
+ * Ascertain if the given code point is a label separator as
+ * defined by the IDNA RFC
+ *
+ * @param ch The code point to be ascertained
+ * @return true if the char is a label separator
+ * @draft ICU 2.8
+ */
+static inline UBool isLabelSeparator(UChar ch){
+ switch(ch){
+ case 0x002e:
+ case 0x3002:
+ case 0xFF0E:
+ case 0xFF61:
+ return TRUE;
+ default:
+ return FALSE;
}
- if((src==NULL) || (srcLength < -1) || (destCapacity<0) || (!dest && destCapacity > 0)){
- *status = U_ILLEGAL_ARGUMENT_ERROR;
- return 0;
+}
+
+// returns the length of the label excluding the separator
+// if *limit == separator then the length returned does not include
+// the separtor.
+static inline int32_t
+getNextSeparator(UChar *src,int32_t srcLength,
+ UChar **limit,
+ UBool *done){
+ if(srcLength == -1){
+ int32_t i;
+ for(i=0 ; ;i++){
+ if(src[i] == 0){
+ *limit = src + i; // point to null
+ *done = TRUE;
+ return i;
+ }
+ if(isLabelSeparator(src[i])){
+ *limit = src + (i+1); // go past the delimiter
+ return i;
+
+ }
+ }
+ }else{
+ int32_t i;
+ for(i=0;i<srcLength;i++){
+ if(isLabelSeparator(src[i])){
+ *limit = src + (i+1); // go past the delimiter
+ return i;
+ }
+ }
+ // we have not found the delimiter
+ // if(i==srcLength)
+ *limit = src+srcLength;
+ *done = TRUE;
+
+ return i;
}
+}
+static inline UBool isLDHChar(UChar ch){
+ // high runner case
+ if(ch>0x007A){
+ return FALSE;
+ }
+ //[\\u002D \\u0030-\\u0039 \\u0041-\\u005A \\u0061-\\u007A]
+ if( (ch==0x002D) ||
+ (0x0030 <= ch && ch <= 0x0039) ||
+ (0x0041 <= ch && ch <= 0x005A) ||
+ (0x0061 <= ch && ch <= 0x007A)
+ ){
+ return TRUE;
+ }
+ return FALSE;
+}
+
+static int32_t
+_internal_toASCII(const UChar* src, int32_t srcLength,
+ UChar* dest, int32_t destCapacity,
+ int32_t options,
+ UStringPrepProfile* nameprep,
+ UParseError* parseError,
+ UErrorCode* status){
+
UChar b1Stack[MAX_LABEL_BUFFER_SIZE], b2Stack[MAX_LABEL_BUFFER_SIZE];
//initialize pointers to stack buffers
UChar *b1 = b1Stack, *b2 = b2Stack;
b2Capacity = MAX_LABEL_BUFFER_SIZE ,
reqLength=0;
-
+ int32_t namePrepOptions = ((options & UIDNA_ALLOW_UNASSIGNED) != 0) ? USPREP_ALLOW_UNASSIGNED: 0;
UBool* caseFlags = NULL;
// the source contains all ascii codepoints
int32_t j=0;
//get the options
- UBool allowUnassigned = (UBool)((options & UIDNA_ALLOW_UNASSIGNED) != 0);
UBool useSTD3ASCIIRules = (UBool)((options & UIDNA_USE_STD3_RULES) != 0);
-
+
int32_t failPos = -1;
- // step 2
- StringPrep* prep = StringPrep::createNameprepInstance(*status);
- if(U_FAILURE(*status)){
- goto CLEANUP;
- }
-
- b1Len = prep->process(src,srcLength,b1, b1Capacity,allowUnassigned, parseError, *status);
+ // step 2
+ b1Len = usprep_prepare(nameprep, src, srcLength, b1, b1Capacity, namePrepOptions, parseError, status);
if(*status == U_BUFFER_OVERFLOW_ERROR){
// redo processing of string
*status = U_ZERO_ERROR; // reset error
- b1Len = prep->process(src,srcLength,b1, b1Len,allowUnassigned, parseError, *status);
+ b1Len = usprep_prepare(nameprep, src, srcLength, b1, b1Len, namePrepOptions, parseError, status);
}
// error bail out
if(U_FAILURE(*status)){
for( j=0;j<b1Len;j++){
if(b1[j] > 0x7F){
srcIsASCII = FALSE;
- }
- // here we do not assemble surrogates
- // since we know that LDH code points
- // are in the ASCII range only
- if(prep->isLDHChar(b1[j])==FALSE){
+ }else if(isLDHChar(b1[j])==FALSE){ // if the char is in ASCII range verify that it is an LDH character
srcIsLDH = FALSE;
failPos = j;
}
if(useSTD3ASCIIRules == TRUE){
// verify 3a and 3b
- if( srcIsLDH == FALSE /* source contains some non-LDH characters */
+ // 3(a) Verify the absence of non-LDH ASCII code points; that is, the
+ // absence of 0..2C, 2E..2F, 3A..40, 5B..60, and 7B..7F.
+ // 3(b) Verify the absence of leading and trailing hyphen-minus; that
+ // is, the absence of U+002D at the beginning and end of the
+ // sequence.
+ if( srcIsLDH == FALSE /* source at this point should not contain anyLDH characters */
|| b1[0] == HYPHEN || b1[b1Len-1] == HYPHEN){
*status = U_IDNA_STD3_ASCII_RULES_ERROR;
}
uprv_free(caseFlags);
- delete prep;
-
return u_terminateUChars(dest, destCapacity, reqLength, status);
}
-
-U_CAPI int32_t U_EXPORT2
-uidna_toUnicode(const UChar* src, int32_t srcLength,
- UChar* dest, int32_t destCapacity,
- int32_t options,
- UParseError* parseError,
- UErrorCode* status){
-
- if(status == NULL || U_FAILURE(*status)){
- return 0;
- }
- if( (src==NULL) || (srcLength < -1) || (destCapacity<0) || (!dest && destCapacity > 0)){
- *status = U_ILLEGAL_ARGUMENT_ERROR;
- return 0;
- }
+static int32_t
+_internal_toUnicode(const UChar* src, int32_t srcLength,
+ UChar* dest, int32_t destCapacity,
+ int32_t options,
+ UStringPrepProfile* nameprep,
+ UParseError* parseError,
+ UErrorCode* status){
//get the options
- UBool allowUnassigned = (UBool)((options & UIDNA_ALLOW_UNASSIGNED) != 0);
UBool useSTD3ASCIIRules = (UBool)((options & UIDNA_USE_STD3_RULES) != 0);
+ int32_t namePrepOptions = ((options & UIDNA_ALLOW_UNASSIGNED) != 0) ? USPREP_ALLOW_UNASSIGNED: 0;
UChar b1Stack[MAX_LABEL_BUFFER_SIZE], b2Stack[MAX_LABEL_BUFFER_SIZE], b3Stack[MAX_LABEL_BUFFER_SIZE];
b2Capacity = MAX_LABEL_BUFFER_SIZE,
b3Capacity = MAX_LABEL_BUFFER_SIZE,
reqLength=0;
-
- StringPrep* prep = StringPrep::createNameprepInstance(*status);
+
b1Len = 0;
UBool* caseFlags = NULL;
UBool srcIsLDH = TRUE;
int32_t failPos =0;
- if(U_FAILURE(*status)){
- goto CLEANUP;
- }
-
// step 1: find out if all the codepoints in src are ASCII
if(srcLength==-1){
srcLength = 0;
for(;src[srcLength]!=0;){
if(src[srcLength]> 0x7f){
srcIsASCII = FALSE;
- }
- // here we do not assemble surrogates
- // since we know that LDH code points
- // are in the ASCII range only
- if(prep->isLDHChar(src[srcLength])==FALSE){
+ }else if(isLDHChar(src[srcLength])==FALSE){
+ // here we do not assemble surrogates
+ // since we know that LDH code points
+ // are in the ASCII range only
srcIsLDH = FALSE;
failPos = srcLength;
}
srcLength++;
}
- }else{
+ }else if(srcLength > 0){
for(int32_t j=0; j<srcLength; j++){
if(src[j]> 0x7f){
srcIsASCII = FALSE;
- }
- // here we do not assemble surrogates
- // since we know that LDH code points
- // are in the ASCII range only
- if(prep->isLDHChar(src[j])==FALSE){
+ }else if(isLDHChar(src[j])==FALSE){
+ // here we do not assemble surrogates
+ // since we know that LDH code points
+ // are in the ASCII range only
srcIsLDH = FALSE;
failPos = j;
}
}
+ }else{
+ return 0;
}
if(srcIsASCII == FALSE){
// step 2: process the string
- b1Len = prep->process(src,srcLength,b1,b1Capacity,allowUnassigned, parseError, *status);
+ b1Len = usprep_prepare(nameprep, src, srcLength, b1, b1Capacity, namePrepOptions, parseError, status);
if(*status == U_BUFFER_OVERFLOW_ERROR){
// redo processing of string
/* we do not have enough room so grow the buffer*/
*status = U_ZERO_ERROR; // reset error
- b1Len = prep->process(src,srcLength,b1, b1Len,allowUnassigned, parseError, *status);
+ b1Len = usprep_prepare(nameprep, src, srcLength, b1, b1Len, namePrepOptions, parseError, status);
}
//bail out on error
if(U_FAILURE(*status)){
}
uprv_free(caseFlags);
- delete prep;
-
+
// The RFC states that
// <quote>
// ToUnicode never fails. If any step fails, then the original input
return u_terminateUChars(dest, destCapacity, reqLength, status);
}
-// returns the length of the label excluding the separator
-// if *limit == separator then the length returned does not include
-// the separtor.
-static int32_t
-getNextSeparator(UChar *src,int32_t srcLength,StringPrep* prep,
- UChar **limit,
- UBool *done,
- UErrorCode *status){
- if(srcLength == -1){
- int32_t i;
- for(i=0 ; ;i++){
- if(src[i] == 0){
- *limit = src + i; // point to null
- *done = TRUE;
- return i;
- }
- if(prep->isLabelSeparator(src[i],*status)){
- *limit = src + (i+1); // go past the delimiter
- return i;
-
- }
- }
- }else{
- int32_t i;
- for(i=0;i<srcLength;i++){
- if(prep->isLabelSeparator(src[i],*status)){
- *limit = src + (i+1); // go past the delimiter
- return i;
- }
- }
- // we have not found the delimiter
- // if(i==srcLength)
- *limit = src+srcLength;
- *done = TRUE;
+U_CAPI int32_t U_EXPORT2
+uidna_toASCII(const UChar* src, int32_t srcLength,
+ UChar* dest, int32_t destCapacity,
+ int32_t options,
+ UParseError* parseError,
+ UErrorCode* status){
+
+ if(status == NULL || U_FAILURE(*status)){
+ return 0;
+ }
+ if((src==NULL) || (srcLength < -1) || (destCapacity<0) || (!dest && destCapacity > 0)){
+ *status = U_ILLEGAL_ARGUMENT_ERROR;
+ return 0;
+ }
- return i;
+ UStringPrepProfile* nameprep = usprep_open(NULL,DATA_FILE_NAME, status);
+
+ if(U_FAILURE(*status)){
+ return -1;
}
+
+ int32_t retLen = _internal_toASCII(src, srcLength, dest, destCapacity, options, nameprep, parseError, status);
+
+ /* close the profile*/
+ usprep_close(nameprep);
+
+ return retLen;
}
+U_CAPI int32_t U_EXPORT2
+uidna_toUnicode(const UChar* src, int32_t srcLength,
+ UChar* dest, int32_t destCapacity,
+ int32_t options,
+ UParseError* parseError,
+ UErrorCode* status){
+
+ if(status == NULL || U_FAILURE(*status)){
+ return 0;
+ }
+ if( (src==NULL) || (srcLength < -1) || (destCapacity<0) || (!dest && destCapacity > 0)){
+ *status = U_ILLEGAL_ARGUMENT_ERROR;
+ return 0;
+ }
+
+ UStringPrepProfile* nameprep = usprep_open(NULL, DATA_FILE_NAME, status);
+
+ if(U_FAILURE(*status)){
+ return -1;
+ }
+
+ int32_t retLen = _internal_toUnicode(src, srcLength, dest, destCapacity, options, nameprep, parseError, status);
+
+ usprep_close(nameprep);
+
+ return retLen;
+}
+
+
U_CAPI int32_t U_EXPORT2
uidna_IDNToASCII( const UChar *src, int32_t srcLength,
UChar* dest, int32_t destCapacity,
int32_t reqLength = 0;
- StringPrep* prep = StringPrep::createNameprepInstance(*status);
+ UStringPrepProfile* nameprep = usprep_open(NULL, DATA_FILE_NAME, status);
if(U_FAILURE(*status)){
return 0;
for(;;){
- labelLen = getNextSeparator(labelStart,remainingLen, prep, &delimiter,&done, status);
+ labelLen = getNextSeparator(labelStart,remainingLen, &delimiter,&done);
- labelReqLength = uidna_toASCII( labelStart, labelLen,
- currentDest, remainingDestCapacity,
- options, parseError, status);
+ labelReqLength = _internal_toASCII( labelStart, labelLen,
+ currentDest, remainingDestCapacity,
+ options, nameprep,
+ parseError, status);
if(*status == U_BUFFER_OVERFLOW_ERROR){
}
- delete prep;
+ usprep_close(nameprep);
return u_terminateUChars(dest, destCapacity, reqLength, status);
}
int32_t reqLength = 0;
- StringPrep* prep = StringPrep::createNameprepInstance(*status);
+ UStringPrepProfile* nameprep = usprep_open(NULL, DATA_FILE_NAME, status);
if(U_FAILURE(*status)){
return 0;
for(;;){
- labelLen = getNextSeparator(labelStart,remainingLen, prep, &delimiter,&done, status);
+ labelLen = getNextSeparator(labelStart,remainingLen, &delimiter,&done);
- labelReqLength = uidna_toUnicode(labelStart, labelLen,
- currentDest, remainingDestCapacity,
- options, parseError, status);
+ labelReqLength = _internal_toUnicode(labelStart, labelLen,
+ currentDest, remainingDestCapacity,
+ options, nameprep,
+ parseError, status);
if(*status == U_BUFFER_OVERFLOW_ERROR){
}
- delete prep;
+ usprep_close(nameprep);
return u_terminateUChars(dest, destCapacity, reqLength, status);
}
projects / apple / icu.git / blobdiff