CFX_ByteStringC CPDF_SimpleParser::GetWord() {
const uint8_t* pStart;
FX_DWORD dwSize;
int type;
ParseWord(pStart, dwSize, type);
if (dwSize == 1 && pStart[0] == '<') {
while (m_dwCurPos < m_dwSize && m_pData[m_dwCurPos] != '>') {
m_dwCurPos++;
}
if (m_dwCurPos < m_dwSize) {
m_dwCurPos++;
}
return CFX_ByteStringC(pStart,
(FX_STRSIZE)(m_dwCurPos - (pStart - m_pData)));
}
if (dwSize == 1 && pStart[0] == '(') {
int level = 1;
while (m_dwCurPos < m_dwSize) {
if (m_pData[m_dwCurPos] == ')') {
level--;
if (level == 0) {
break;
}
}
if (m_pData[m_dwCurPos] == '\\') {
if (m_dwSize <= m_dwCurPos) {
break;
}
m_dwCurPos++;
} else if (m_pData[m_dwCurPos] == '(') {
level++;
}
if (m_dwSize <= m_dwCurPos) {
break;
}
m_dwCurPos++;
}
if (m_dwCurPos < m_dwSize) {
m_dwCurPos++;
}
return CFX_ByteStringC(pStart,
(FX_STRSIZE)(m_dwCurPos - (pStart - m_pData)));
}
return CFX_ByteStringC(pStart, dwSize);
}
UWORD ParseCPPLines
(
struct SyntaxHandle *handle,
LineDef** lines,
BSTR textData,
UWORD startoffset,
UWORD maxlength,
CArray <SyntaxChunk, SyntaxChunk>& SyntaxArr
)
{
SyntaxArr.SetSize(maxlength); // hm.. TODO, add instead
SyntaxChunk* syntax = SyntaxArr.GetData();
int ChunkUsed = 0;
UWORD parsedlength = 0;
BOOL donext;
int nline = 0;
if (maxlength == 0) return 0;
int offset = 0;
do
{
// register UWORD len = linenode->LineLen;
UBYTE el = 0;
BOOL cont = FALSE;
if (offset == 0) /* If first line in the doc */
el = 0;
else
{
// while (!prevline->IsParsed
el = 0;//syntax[-1].Element;
}
// linenode->BegEl = el; /* Attr at beginning of line taken from the previous line's EndAttr */
// if (linenode->Syntax) free(linenode->Syntax);
// linenode->Syntax = (SyntaxChunk*)malloc(sizeof(struct SyntaxChunk)*(linenode->LineLen+1));
lines[nline]->m_ckOffset = ChunkUsed;
syntax->Length = 0;
syntax->Element = el;
syntax->Flags = ElementFlags[el];
ChunkUsed += 1;
int linelen = lines[nline]->m_lineLength;
WCHAR* lineData = &textData[offset];
UWORD col = 0;
while (col < linelen)
{
WCHAR *c = &lineData[col];
UWORD ellen = 1; /* Assume */
UBYTE sel = 255;
BOOL off = FALSE;
if (!(ctype[*c] & C))
{
if (el == 0) ellen = ParseWord(lineData, col, linelen, &sel);
}
else
{
if (c[0] == '/') /* This could be the start of a comment */
{
if ((el == 0) || (el == EL_PREPROCESSOR)) /* Can have comments on same line as preprocessor */
{
if (c[1] == '/') /* CPP comment, set rest of line to that */
{
el = EL_CPPCOMMENT;
ellen = linelen-col;
}
else if (c[1] == '*') /* ANSI comment */
{
el = EL_COMMENT;
ellen = 2;
}
else
{
if (el != EL_PREPROCESSOR) /* If it wasn't a comment and we're not */
sel = EL_SYMBOL; /* in preprocessor make it a symbol */
}
}
}
else if (c[0] == '*') /* This could be the end of a comment */
{
if ((el == EL_COMMENT) && (c[1] == '/'))
{
off = TRUE;
el = 0;
ellen = 2;
}
else
{
if (el == 0) /* It was just a normal symbol */
{
sel = EL_SYMBOL;
}
}
}
else if (c[0] == '\"') /* Either beginning or ending a string quote */
{
if (el == 0) /* Beginning */
{
el = EL_STRING;
}
else if (el == EL_STRING) // Maybe it's ending */
{
/* Here we check if it's a \", then we check if it's a \\" */
if (((col > 0) && (c[-1] != '\\')) || /* it's not a '\"' */
((col > 1) && (c[-2] == '\\'))) /* it's a '\\"' */
{
off = TRUE;
el = 0;
}
}
}
else if (c[0] == '\'') /* Either beginning or ending a char quote */
{
if (el == 0) /* Beginning */
{
el = EL_CHARACTER;
}
else if (el == EL_CHARACTER) /* Maybe it's ending */
{
/* Here we check if it's a \', then we check if it's a \\' */
if (((col > 0) && (c[-1] != '\\')) || /* it's not a \' */
((col > 1) && (c[-2] == '\\'))) /* it's a \\' */
{
off = TRUE;
el = 0;
}
}
}
else if (c[0] == '#') /* If first non-blank char on line, this is preprocessor statement */
{
if (el == 0)
{
/* TODO: Check if first non-blank */
el = EL_PREPROCESSOR;
}
}
else if (c[0] == '\\') /* If last on line, this is a continuation on the next line */
{
if ((col == linelen-1) && /* last on line */
(el != EL_CPPCOMMENT))
{
if (el) cont = TRUE; /* Only need to continue if there was some previous el */
}
}
}
if (sel == 255) sel = el;
if (off) syntax->Length += ellen;
/* If the element has changed goto next syntax chunk */
if (sel != syntax->Element)
{
// Only create new if the previous syntax chunk contains something
// if not, just overwrite that one
if (syntax->Length > 0)
{
ChunkUsed++;
syntax++;
}
syntax->Length = 0;
syntax->Element = sel;
syntax->Flags = ElementFlags[sel];
}
if (!off) syntax->Length += ellen;
// CString msg;
// msg.Format("%d", syntax->Length);
// AfxMessageBox(msg/*(LPCTSTR)(&lineData[col])*/);
col += ellen;
}
// msg.Format("%d, %d, %d", linelen, offset, maxlength);
// AfxMessageBox(msg);
offset += linelen+1;
syntax++; // Always new on each line
nline++;
#if 0
linenode->Syntax = realloc(linenode->Syntax, linenode->ChunkUsed*sizeof(struct SyntaxChunk));
/* numlines is used later to know the number of lines to visually update
folded lines shouldn't be included in this count */
if (!linenode->Folded) parsedlines++;
/* Set this line's EndEl */
if ((el == EL_COMMENT) || cont) /* This continues over several lines */
linenode->EndEl= el;
else
linenode->EndEl = 0;
donext = FALSE; /* Assume that we're not gonna parse the next line */
if (--maxlines)
{
/* Check if Next line is candidate for scanning
First check if there IS a next line */
if ((nextline = linenode->Succ)->Succ)
{
/* If the next line's BegEl is different than the line we just
parsed's EndAttr then parse the next one too */
if (nextline->BegEl != linenode->EndEl)
{
linenode = nextline;
donext = TRUE;
}
}
}
#endif
}
while (offset < maxlength/*-1*/);
// while (donext);
SyntaxArr.SetSize(ChunkUsed);
return parsedlength; /* Number of lines that were parsed */
}
void FileStringReader::HandlePragma(char *str)
{
if (!memcmp(str, "id ", 3)) {
this->data.next_string_id = strtoul(str + 3, NULL, 0);
} else if (!memcmp(str, "name ", 5)) {
strecpy(_lang.name, str + 5, lastof(_lang.name));
} else if (!memcmp(str, "ownname ", 8)) {
strecpy(_lang.own_name, str + 8, lastof(_lang.own_name));
} else if (!memcmp(str, "isocode ", 8)) {
strecpy(_lang.isocode, str + 8, lastof(_lang.isocode));
} else if (!memcmp(str, "textdir ", 8)) {
if (!memcmp(str + 8, "ltr", 3)) {
_lang.text_dir = TD_LTR;
} else if (!memcmp(str + 8, "rtl", 3)) {
_lang.text_dir = TD_RTL;
} else {
error("Invalid textdir %s", str + 8);
}
} else if (!memcmp(str, "digitsep ", 9)) {
str += 9;
strecpy(_lang.digit_group_separator, strcmp(str, "{NBSP}") == 0 ? NBSP : str, lastof(_lang.digit_group_separator));
} else if (!memcmp(str, "digitsepcur ", 12)) {
str += 12;
strecpy(_lang.digit_group_separator_currency, strcmp(str, "{NBSP}") == 0 ? NBSP : str, lastof(_lang.digit_group_separator_currency));
} else if (!memcmp(str, "decimalsep ", 11)) {
str += 11;
strecpy(_lang.digit_decimal_separator, strcmp(str, "{NBSP}") == 0 ? NBSP : str, lastof(_lang.digit_decimal_separator));
} else if (!memcmp(str, "winlangid ", 10)) {
const char *buf = str + 10;
long langid = strtol(buf, NULL, 16);
if (langid > (long)UINT16_MAX || langid < 0) {
error("Invalid winlangid %s", buf);
}
_lang.winlangid = (uint16)langid;
} else if (!memcmp(str, "grflangid ", 10)) {
const char *buf = str + 10;
long langid = strtol(buf, NULL, 16);
if (langid >= 0x7F || langid < 0) {
error("Invalid grflangid %s", buf);
}
_lang.newgrflangid = (uint8)langid;
} else if (!memcmp(str, "gender ", 7)) {
if (this->master) error("Genders are not allowed in the base translation.");
char *buf = str + 7;
for (;;) {
const char *s = ParseWord(&buf);
if (s == NULL) break;
if (_lang.num_genders >= MAX_NUM_GENDERS) error("Too many genders, max %d", MAX_NUM_GENDERS);
strecpy(_lang.genders[_lang.num_genders], s, lastof(_lang.genders[_lang.num_genders]));
_lang.num_genders++;
}
} else if (!memcmp(str, "case ", 5)) {
if (this->master) error("Cases are not allowed in the base translation.");
char *buf = str + 5;
for (;;) {
const char *s = ParseWord(&buf);
if (s == NULL) break;
if (_lang.num_cases >= MAX_NUM_CASES) error("Too many cases, max %d", MAX_NUM_CASES);
strecpy(_lang.cases[_lang.num_cases], s, lastof(_lang.cases[_lang.num_cases]));
_lang.num_cases++;
}
} else {
StringReader::HandlePragma(str);
}
}