CSVMap::CSVMap(const string &file_path, bool& header)
{
this->file_path_ = file_path;
this->header_ = header;
ConstructMap();
}
bool
RE_Class::ConstructL (unsigned depth, const uni_char *source, unsigned &index, unsigned length, bool extended)
{
int range_first = -1;
for (unsigned i = 0; i < BITMAP_ELEMENTS; ++i)
bitmap[i] = 0;
if (source[index] == '^')
{
inverted = true;
++index;
}
else
inverted = false;
#ifdef RE_FEATURE__CHARACTER_CLASS_INTERSECTION
if (depth++ >= RE_FEATURE__CHARACTER_CLASS_INTERSECTION_MAX_DEPTH)
return false;
#endif // RE_FEATURE__CHARACTER_CLASS_INTERSECTION
#ifdef ES_FEATURE__ERROR_MESSAGES
unsigned range_first_index, range_second_index;
#endif /* ES_FEATURE__ERROR_MESSAGES */
RE_SKIP_WHITESPACE ();
RE_ExcludeIncludeRange *eir = OP_NEW_L (RE_ExcludeIncludeRange, (false, 0, INT_MAX));
OpStackAutoPtr<RE_ExcludeIncludeRange> eir_anchor (eir);
#define SetCharacter(ch) do { SetCharacter (ch); if (!(ch < BITMAP_RANGE)) eir = RE_ExcludeIncludeRange::Include (eir, ch, ch); } while (0)
#define SetRange(first, last) do { SetRange (first, last); if (last >= BITMAP_RANGE) eir = RE_ExcludeIncludeRange::Include (eir, MAX (first, BITMAP_RANGE), last); } while (0)
while (index < length)
{
#ifdef ES_FEATURE__ERROR_MESSAGES
if (range_first == -1)
range_first_index = index;
#endif /* ES_FEATURE__ERROR_MESSAGES */
int character = source[index++];
if (character == ']')
{
if (range_first >= 0)
{
SetCharacter (range_first);
SetCharacter ('-');
}
--index;
break;
}
else if (character == '\\')
{
if (index == length)
return false;
character = -2;
bool builtin_class_handled = false;
switch (source[index++])
{
case '0':
if (index == length || !RE_Compiler::IsOctalDigit (source[index]))
{
character = 0;
break;
}
++index;
case '1': case '2': case '3': case '4': case '5': case'6': case '7':
character = 0;
--index;
while (index < length && RE_Compiler::IsOctalDigit (source[index]))
{
int next_character = character * 8 + (source[index] - '0');
if (next_character > 255)
break;
character = next_character;
++index;
}
break;
case 'b':
character = 8;
break;
case 't':
character = 9;
break;
case 'n':
character = 10;
break;
case 'v':
character = 11;
break;
case 'f':
character = 12;
break;
case 'r':
character = 13;
break;
case 'd':
if (builtin_class == BUILTIN_EMPTY || builtin_class == BUILTIN_DIGIT)
{
builtin_class = BUILTIN_DIGIT;
builtin_class_handled = true;
}
SetRange ('0', '9');
break;
case 'D':
if (builtin_class == BUILTIN_EMPTY || builtin_class == BUILTIN_NON_DIGIT)
{
builtin_class = BUILTIN_NON_DIGIT;
builtin_class_handled = true;
}
SetRange (0, '0' - 1);
SetRange ('9' + 1, INT_MAX - 1);
break;
case 's':
if (builtin_class == BUILTIN_EMPTY || builtin_class == BUILTIN_WHITESPACE)
{
builtin_class = BUILTIN_WHITESPACE;
builtin_class_handled = true;
}
SetRange (9, 13);
SetCharacter (32);
SetCharacter (160);
SetCharacter (0x1680);
SetCharacter (0x180e);
SetRange (0x2000, 0x200b);
SetCharacter (0x2028);
SetCharacter (0x2029);
SetCharacter (0x202f);
SetCharacter (0x205f);
SetCharacter (0x3000);
SetCharacter (0xfeff);
break;
case 'S':
if (builtin_class == BUILTIN_EMPTY || builtin_class == BUILTIN_NON_WHITESPACE)
{
builtin_class = BUILTIN_NON_WHITESPACE;
builtin_class_handled = true;
}
SetRange (0, 8);
SetRange (14, 31);
SetRange (33, 159);
SetRange (161, 0x1680 - 1);
SetRange (0x1680 + 1, 0x180e - 1);
SetRange (0x180e + 1, 0x2000 - 1);
SetRange (0x200b + 1, 0x2028 - 1);
SetRange (0x2029 + 1, 0x202f - 1);
SetRange (0x202f + 1, 0x205f - 1);
SetRange (0x205f + 1, 0x3000 - 1);
SetRange (0x3000 + 1, 0xfeff - 1);
SetRange (0xfeff + 1, INT_MAX - 1);
break;
case 'w':
if (builtin_class == BUILTIN_EMPTY || builtin_class == BUILTIN_WORD_CHARACTER)
{
builtin_class = BUILTIN_WORD_CHARACTER;
builtin_class_handled = true;
}
SetRange ('0', '9');
SetRange ('A', 'Z');
SetCharacter ('_');
SetRange ('a', 'z');
break;
case 'W':
if (builtin_class == BUILTIN_EMPTY || builtin_class == BUILTIN_NON_WORD_CHARACTER)
{
builtin_class = BUILTIN_NON_WORD_CHARACTER;
builtin_class_handled = true;
}
SetRange (0, '0' - 1);
SetRange ('9' + 1, 'A' - 1);
SetRange ('Z' + 1, '_' - 1);
SetCharacter ('_' + 1);
SetRange ('z' + 1, INT_MAX - 1);
break;
case 'c':
if (index - length < 2)
continue;
if (!RE_Compiler::IsLetter (source[index]))
continue;
character = source[index++] % 32;
break;
case 'x':
case 'u':
if (index - length < 3)
continue;
#ifdef RE_FEATURE__BRACED_HEXADECIMAL_ESCAPES
if (source[index] == '{' && RE_Compiler::IsHexDigit (source[index + 1]))
{
unsigned index0 = index + 1;
character = 0;
while (index0 < length && source[index0] != '}' && RE_Compiler::IsHexDigit (source[index0]))
character = (character << 4) | RE_Compiler::HexToCharacter (source[index0++]);
if (index0 < length && source[index0] == '}')
{
index = index0 + 1;
break;
}
character = -2;
}
#endif // RE_FEATURE__BRACED_HEXADECIMAL_ESCAPES
if (!RE_Compiler::IsHexDigit (source[index]) || !RE_Compiler::IsHexDigit (source[index + 1]))
continue;
if (source[index - 1] == 'x')
{
character = RE_Compiler::HexToCharacter (source[index + 1], source[index]);
index += 2;
}
else
{
if (index - length < 5)
continue;
if (!RE_Compiler::IsHexDigit (source[index + 2]) || !RE_Compiler::IsHexDigit (source[index + 3]))
continue;
character = RE_Compiler::HexToCharacter (source[index + 3], source[index + 2], source[index + 1], source[index]);
index += 4;
}
break;
default:
character = source[index - 1];
break;
#if 0
#ifdef ES_FEATURE__ERROR_MESSAGES
error_string = ES_ERROR ("invalid escape sequence.");
--index;
#endif /* ES_FEATURE__ERROR_MESSAGES */
return false;
#endif // 0
}
if (!builtin_class_handled)
builtin_class = BUILTIN_NONE;
}
#ifdef RE_FEATURE__CHARACTER_CLASS_INTERSECTION
else if (character == '&' && length - index >= 3 && source[index] == '&' && source[index + 1] == '[')
{
unsigned index0 = index + 2;
RE_Class *iw = OP_NEW_L (RE_Class, (case_insensitive));
iw->intersect_next = intersect_with;
intersect_with = iw;
if (!iw->ConstructL (depth, source, index0, length, extended))
return false;
index = index0 + 1;
}
#endif // RE_FEATURE__CHARACTER_CLASS_INTERSECTION
if (character == -2)
range_first = -1;
else if (range_first == -1)
{
RE_SKIP_WHITESPACE ();
if (index < length)
{
int dash = source[index];
if (dash == '-')
{
range_first = character;
++index;
#ifdef ES_FEATURE__ERROR_MESSAGES
range_second_index = index;
#endif /* ES_FEATURE__ERROR_MESSAGES */
continue;
}
}
if (character != -2)
{
SetCharacter (character);
builtin_class = BUILTIN_NONE;
}
}
else if (range_first == -2 || range_first > character)
{
#ifdef ES_FEATURE__ERROR_MESSAGES
if (range_first == -2 || character == -2)
{
error_string = ES_ERROR ("invalid component in range.");
if (range_first == -2)
index = range_first_index;
else
index = range_second_index;
}
else
{
error_string = ES_ERROR ("negative range.");
index = range_first_index;
}
#endif /* ES_FEATURE__ERROR_MESSAGES */
return false;
}
else
{
SetRange (range_first, character);
builtin_class = BUILTIN_NONE;
range_first = -1;
}
RE_SKIP_WHITESPACE ();
}
if (index == length)
return false;
#ifdef RE_FEATURE__CHARACTER_CLASS_INTERSECTION
RE_Class **iwp = &intersect_with;
while (RE_Class *iw = *iwp)
{
for (unsigned i = 0; i < BITMAP_ELEMENTS; ++i)
#ifdef RE_COMPACT_CLASS_BITMAP
bitmap[i] &= iw->bitmap[i];
#else // RE_COMPACT_CLASS_BITMAP
bitmap[i] &= iw->bitmap[i];
#endif // RE_COMPACT_CLASS_BITMAP
if (!iw->map)
{
/* Discard ranges that only use the bitmap; they are fully
merged with the main set in the step above. */
*iwp = iw->intersect_next;
OP_DELETE(iw);
}
else
iwp = &iw->intersect_next;
}
#endif // RE_FEATURE__CHARACTER_CLASS_INTERSECTION
if (case_insensitive)
{
#ifdef RE_COMPACT_CLASS_BITMAP
bitmap[3] |= bitmap[2] & 0x7ffffe;
bitmap[2] |= bitmap[3] & 0x7ffffe;
#else // RE_COMPACT_CLASS_BITMAP
unsigned b;
for (b = 'A'; b <= 'Z'; ++b)
if (bitmap[b])
bitmap[b + 32] = 1;
else if (bitmap[b + 32])
bitmap[b] = 1;
#endif // RE_COMPACT_CLASS_BITMAP
for (int character = 128; character < 256; ++character)
if (Match (character))
if (uni_islower (character))
SetCharacter (uni_toupper (character));
else if (uni_isupper (character))
SetCharacter (uni_tolower (character));
}
if (inverted)
for (unsigned idx = 0; idx < BITMAP_ELEMENTS; ++idx)
#ifdef RE_COMPACT_CLASS_BITMAP
bitmap[idx] ^= ~0u;
#else // RE_COMPACT_CLASS_BITMAP
bitmap[idx] = !bitmap[idx];
#endif // RE_COMPACT_CLASS_BITMAP
#undef SetCharacter
#undef SetRange
ConstructMap (eir);
return true;
}
MapParameterStorage(It begin, It end, const Value& value, const string& default_value) : storage_(ConstructMap(begin, end, value)), default_value_(default_value) {
}