U_CFUNC void U_CALLCONV
derivedPropsLineFn(void *context,
char *fields[][2], int32_t /* fieldCount */,
UErrorCode *pErrorCode)
{
UnicodeTest *me=(UnicodeTest *)context;
uint32_t start, end;
int32_t i;
u_parseCodePointRange(fields[0][0], &start, &end, pErrorCode);
if(U_FAILURE(*pErrorCode)) {
me->errln("UnicodeTest: syntax error in DerivedCoreProperties.txt or DerivedNormalizationProps.txt field 0 at %s\n", fields[0][0]);
return;
}
/* parse derived binary property name, ignore unknown names */
i=getTokenIndex(derivedPropsNames, LENGTHOF(derivedPropsNames), fields[1][0]);
if(i<0) {
UnicodeString propName(fields[1][0], (int32_t)(fields[1][1]-fields[1][0]));
propName.trim();
if(me->unknownPropertyNames->find(propName)==NULL) {
UErrorCode errorCode=U_ZERO_ERROR;
me->unknownPropertyNames->puti(propName, 1, errorCode);
me->errln("UnicodeTest warning: unknown property name '%s' in DerivedCoreProperties.txt or DerivedNormalizationProps.txt\n", fields[1][0]);
}
return;
}
me->derivedProps[i].add(start, end);
}
SourceRangeContainer FollowSymbol::followSymbol(CXTranslationUnit tu,
const Cursor &fullCursor,
uint line,
uint column)
{
std::unique_ptr<Tokens> tokens(new Tokens(fullCursor));
if (!tokens->tokenCount)
tokens.reset(new Tokens(tu));
if (!tokens->tokenCount)
return SourceRangeContainer();
QVector<CXCursor> cursors(static_cast<int>(tokens->tokenCount));
clang_annotateTokens(tu, tokens->data, tokens->tokenCount, cursors.data());
int tokenIndex = getTokenIndex(tu, *tokens, line, column);
QTC_ASSERT(tokenIndex >= 0, return SourceRangeContainer());
const Utf8String tokenSpelling = ClangString(
clang_getTokenSpelling(tu, tokens->data[tokenIndex]));
if (tokenSpelling.isEmpty())
return SourceRangeContainer();
Cursor cursor{cursors[tokenIndex]};
if (cursor.kind() == CXCursor_InclusionDirective) {
CXFile file = clang_getIncludedFile(cursors[tokenIndex]);
const ClangString filename(clang_getFileName(file));
const SourceLocation loc(tu, filename, 1, 1);
return SourceRange(loc, loc);
}
// For definitions we can always find a declaration in current TU
if (cursor.isDefinition())
return extractMatchingTokenRange(cursor.canonical(), tokenSpelling);
if (!cursor.isDeclaration()) {
// This is the symbol usage
// We want to return definition
cursor = cursor.referenced();
if (cursor.isNull() || !cursor.isDefinition()) {
// We can't find definition in this TU
return SourceRangeContainer();
}
return extractMatchingTokenRange(cursor, tokenSpelling);
}
cursor = cursor.definition();
// If we are able to find a definition in current TU
if (!cursor.isNull())
return extractMatchingTokenRange(cursor, tokenSpelling);
return SourceRangeContainer();
}
std::wstring CommonToken::toString() {
std::wstring channelStr = L"";
if (channel > 0) {
channelStr = std::wstring(L",channel=") + std::to_wstring(channel);
}
std::wstring txt = getText();
if (txt != L"") {
replaceAll(txt, L"\n",L"\\n");
replaceAll(txt, L"\r",L"\\r");
replaceAll(txt, L"\t",L"\\t");
} else {
txt = L"<no text>";
}
return std::wstring(L"[@") + std::to_wstring(getTokenIndex()) + std::wstring(L",") + std::to_wstring(start) + std::wstring(L":") + std::to_wstring(stop) + std::wstring(L"='") + txt + std::wstring(L"',<") + std::to_wstring(type) + std::wstring(L">") + channelStr + std::wstring(L",") + std::to_wstring(line) + std::wstring(L":") + std::to_wstring(getCharPositionInLine()) + std::wstring(L"]");
}
U_CFUNC void U_CALLCONV
derivedCorePropsLineFn(void *context,
char *fields[][2], int32_t /* fieldCount */,
UErrorCode *pErrorCode)
{
UnicodeTest *me=(UnicodeTest *)context;
uint32_t start, end;
int32_t i;
u_parseCodePointRange(fields[0][0], &start, &end, pErrorCode);
if(U_FAILURE(*pErrorCode)) {
me->errln("UnicodeTest: syntax error in DerivedCoreProperties.txt field 0 at %s\n", fields[0][0]);
return;
}
/* parse derived binary property name, ignore unknown names */
i=getTokenIndex(derivedCorePropsNames, LENGTHOF(derivedCorePropsNames), fields[1][0]);
if(i<0) {
me->errln("UnicodeTest warning: unknown property name '%s' in \n", fields[1][0]);
return;
}
me->derivedCoreProps[i].add(start, end);
}
static void U_CALLCONV
unicodeDataLineFn(void *context,
char *fields[][2], int32_t fieldCount,
UErrorCode *pErrorCode) {
Props p;
char *end;
static UChar32 prevCode=0;
UChar32 value;
int32_t i;
/* reset the properties */
uprv_memset(&p, 0, sizeof(Props));
/* get the character code, field 0 */
p.code=(UChar32)uprv_strtoul(fields[0][0], &end, 16);
if(end<=fields[0][0] || end!=fields[0][1]) {
fprintf(stderr, "gencase: syntax error in field 0 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
/* get general category, field 2 */
i=getTokenIndex(genCategoryNames, U_CHAR_CATEGORY_COUNT, fields[2][0]);
if(i>=0) {
p.gc=(uint8_t)i;
} else {
fprintf(stderr, "gencase: unknown general category \"%s\" at code 0x%lx\n",
fields[2][0], (unsigned long)p.code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
/* get canonical combining class, field 3 */
value=(UChar32)uprv_strtoul(fields[3][0], &end, 10);
if(end<=fields[3][0] || end!=fields[3][1] || value>0xff) {
fprintf(stderr, "gencase: syntax error in field 3 at %s\n", fields[0][0]);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
p.cc=(uint8_t)value;
/* get uppercase mapping, field 12 */
value=(UChar32)uprv_strtoul(fields[12][0], &end, 16);
if(end!=fields[12][1]) {
fprintf(stderr, "gencase: syntax error in field 12 at code 0x%lx\n",
(unsigned long)p.code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(value!=0 && value!=p.code) {
p.upperCase=value;
uset_add(caseSensitive, p.code);
uset_add(caseSensitive, value);
}
/* get lowercase value, field 13 */
value=(UChar32)uprv_strtoul(fields[13][0], &end, 16);
if(end!=fields[13][1]) {
fprintf(stderr, "gencase: syntax error in field 13 at code 0x%lx\n",
(unsigned long)p.code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(value!=0 && value!=p.code) {
p.lowerCase=value;
uset_add(caseSensitive, p.code);
uset_add(caseSensitive, value);
}
/* get titlecase value, field 14 */
value=(UChar32)uprv_strtoul(fields[14][0], &end, 16);
if(end!=fields[14][1]) {
fprintf(stderr, "gencase: syntax error in field 14 at code 0x%lx\n",
(unsigned long)p.code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
if(value!=0 && value!=p.code) {
p.titleCase=value;
uset_add(caseSensitive, p.code);
uset_add(caseSensitive, value);
}
/* set additional properties from previously parsed files */
if(specialCasingIndex<specialCasingCount && p.code==specialCasings[specialCasingIndex].code) {
p.specialCasing=specialCasings+specialCasingIndex++;
} else {
p.specialCasing=NULL;
}
if(caseFoldingIndex<caseFoldingCount && p.code==caseFoldings[caseFoldingIndex].code) {
p.caseFolding=caseFoldings+caseFoldingIndex++;
/* ignore "Common" mappings (simple==full) that map to the same code point as the regular lowercase mapping */
if( p.caseFolding->status=='C' &&
p.caseFolding->simple==p.lowerCase
) {
p.caseFolding=NULL;
}
} else {
p.caseFolding=NULL;
}
/* check for non-character code points */
if((p.code&0xfffe)==0xfffe || (uint32_t)(p.code-0xfdd0)<0x20) {
fprintf(stderr, "gencase: error - properties for non-character code point U+%04lx\n",
(unsigned long)p.code);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
/* check that the code points (p.code) are in ascending order */
if(p.code<=prevCode && p.code>0) {
fprintf(stderr, "gencase: error - UnicodeData entries out of order, U+%04lx after U+%04lx\n",
(unsigned long)p.code, (unsigned long)prevCode);
*pErrorCode=U_PARSE_ERROR;
exit(U_PARSE_ERROR);
}
/* properties for a single code point */
setProps(&p);
prevCode=p.code;
}
