/**
* Count the number of characters in a word.
* The first character is already valid for a keyword
*
* @param pc The structure to update, str is an input.
* @return Whether a word was parsed (always true)
*/
bool parse_word(tok_ctx &ctx, chunk_t &pc, bool skipcheck)
{
int ch;
static unc_text intr_txt("@interface");
/* The first character is already valid */
pc.str.clear();
pc.str.append(ctx.get());
while (ctx.more())
{
ch = ctx.peek();
if (CharTable::IsKw2(ch))
{
pc.str.append(ctx.get());
}
else if ((ch == '\\') && (unc_tolower(ctx.peek(1)) == 'u'))
{
pc.str.append(ctx.get());
pc.str.append(ctx.get());
skipcheck = true;
}
else
{
break;
}
/* HACK: Non-ASCII character are only allowed in identifiers */
if (ch > 0x7f)
{
skipcheck = true;
}
}
pc.type = CT_WORD;
if (skipcheck)
{
return(true);
}
/* Detect pre-processor functions now */
if ((cpd.in_preproc == CT_PP_DEFINE) &&
(cpd.preproc_ncnl_count == 1))
{
if (ctx.peek() == '(')
{
pc.type = CT_MACRO_FUNC;
}
else
{
pc.type = CT_MACRO;
}
}
else
{
/* '@interface' is reserved, not an interface itself */
if ((cpd.lang_flags & LANG_JAVA) && pc.str.startswith("@") &&
!pc.str.equals(intr_txt))
{
pc.type = CT_ANNOTATION;
}
else
{
/* Turn it into a keyword now */
pc.type = find_keyword_type(pc.text(), pc.str.size());
}
}
return(true);
} // parse_word
/**
* Skips the next bit of whatever and returns the type of block.
*
* pc.str is the input text.
* pc.len in the output length.
* pc.type is the output type
* pc.column is output column
*
* @param pc The structure to update, str is an input.
* @return true/false - whether anything was parsed
*/
static bool parse_next(tok_ctx& ctx, chunk_t& pc)
{
const chunk_tag_t *punc;
int ch, ch1;
if (!ctx.more())
{
//fprintf(stderr, "All done!\n");
return(false);
}
/* Save off the current column */
pc.orig_line = ctx.c.row;
pc.column = ctx.c.col;
pc.orig_col = ctx.c.col;
pc.type = CT_NONE;
pc.nl_count = 0;
pc.flags = 0;
/* If it is turned off, we put everything except newlines into CT_UNKNOWN */
if (cpd.unc_off)
{
if (parse_ignored(ctx, pc))
{
return(true);
}
}
/**
* Parse whitespace
*/
if (parse_whitespace(ctx, pc))
{
return(true);
}
/**
* Handle unknown/unhandled preprocessors
*/
if ((cpd.in_preproc > CT_PP_BODYCHUNK) &&
(cpd.in_preproc <= CT_PP_OTHER))
{
pc.str.clear();
tok_info ss;
ctx.save(ss);
/* Chunk to a newline or comment */
pc.type = CT_PREPROC_BODY;
int last = 0;
while (ctx.more())
{
int ch = ctx.peek();
if ((ch == '\n') || (ch == '\r'))
{
/* Back off if this is an escaped newline */
if (last == '\\')
{
ctx.restore(ss);
pc.str.pop_back();
}
break;
}
/* Quit on a C++ comment start */
if ((ch == '/') && (ctx.peek(1) == '/'))
{
break;
}
last = ch;
ctx.save(ss);
pc.str.append(ctx.get());
}
if (pc.str.size() > 0)
{
return(true);
}
}
/**
* Detect backslash-newline
*/
if ((ctx.peek() == '\\') && parse_bs_newline(ctx, pc))
{
return(true);
}
/**
* Parse comments
*/
if (parse_comment(ctx, pc))
{
return(true);
}
/* Parse code placeholders */
if (parse_code_placeholder(ctx, pc))
{
return(true);
}
/* Check for C# literal strings, ie @"hello" and identifiers @for*/
if ((cpd.lang_flags & LANG_CS) && (ctx.peek() == '@'))
{
if (ctx.peek(1) == '"')
{
parse_cs_string(ctx, pc);
return(true);
}
/* check for non-keyword identifiers such as @if @switch, etc */
if (CharTable::IsKw1(ctx.peek(1)))
{
parse_word(ctx, pc, true);
return(true);
}
}
/* Check for C# Interpolated strings */
if ((cpd.lang_flags & LANG_CS) && (ctx.peek() == '$') && (ctx.peek(1) == '"'))
{
parse_cs_interpolated_string(ctx, pc);
return(true);
}
/* handle VALA """ strings """ */
if ((cpd.lang_flags & LANG_VALA) &&
(ctx.peek() == '"') &&
(ctx.peek(1) == '"') &&
(ctx.peek(2) == '"'))
{
parse_verbatim_string(ctx, pc);
return(true);
}
/* handle C++0x strings u8"x" u"x" U"x" R"x" u8R"XXX(I'm a "raw UTF-8" string.)XXX" */
ch = ctx.peek();
if ((cpd.lang_flags & LANG_CPP) &&
((ch == 'u') || (ch == 'U') || (ch == 'R')))
{
int idx = 0;
bool is_real = false;
if ((ch == 'u') && (ctx.peek(1) == '8'))
{
idx = 2;
}
else if (unc_tolower(ch) == 'u')
{
idx++;
}
if (ctx.peek(idx) == 'R')
{
idx++;
is_real = true;
}
if (ctx.peek(idx) == '"')
{
if (is_real)
{
if (parse_cr_string(ctx, pc, idx))
{
return(true);
}
}
else
{
if (parse_string(ctx, pc, idx, true))
{
parse_suffix(ctx, pc, true);
return(true);
}
}
}
}
/* PAWN specific stuff */
if (cpd.lang_flags & LANG_PAWN)
{
if ((cpd.preproc_ncnl_count == 1) &&
((cpd.in_preproc == CT_PP_DEFINE) ||
(cpd.in_preproc == CT_PP_EMIT)))
{
parse_pawn_pattern(ctx, pc, CT_MACRO);
return(true);
}
/* Check for PAWN strings: \"hi" or !"hi" or !\"hi" or \!"hi" */
if ((ctx.peek() == '\\') || (ctx.peek() == '!'))
{
if (ctx.peek(1) == '"')
{
parse_string(ctx, pc, 1, (ctx.peek() == '!'));
return(true);
}
else if (((ctx.peek(1) == '\\') || (ctx.peek(1) == '!')) &&
(ctx.peek(2) == '"'))
{
parse_string(ctx, pc, 2, false);
return(true);
}
}
/* handle PAWN preprocessor args %0 .. %9 */
if ((cpd.in_preproc == CT_PP_DEFINE) &&
(ctx.peek() == '%') &&
unc_isdigit(ctx.peek(1)))
{
pc.str.clear();
pc.str.append(ctx.get());
pc.str.append(ctx.get());
pc.type = CT_WORD;
return(true);
}
}
/**
* Parse strings and character constants
*/
if (parse_number(ctx, pc))
{
return(true);
}
if (cpd.lang_flags & LANG_D)
{
/* D specific stuff */
if (d_parse_string(ctx, pc))
{
return(true);
}
}
else
{
/* Not D stuff */
/* Check for L'a', L"abc", 'a', "abc", <abc> strings */
ch = ctx.peek();
ch1 = ctx.peek(1);
if ((((ch == 'L') || (ch == 'S')) &&
((ch1 == '"') || (ch1 == '\''))) ||
(ch == '"') ||
(ch == '\'') ||
((ch == '<') && (cpd.in_preproc == CT_PP_INCLUDE)))
{
parse_string(ctx, pc, unc_isalpha(ch) ? 1 : 0, true);
return(true);
}
if ((ch == '<') && (cpd.in_preproc == CT_PP_DEFINE))
{
if (chunk_get_tail()->type == CT_MACRO)
{
/* We have "#define XXX <", assume '<' starts an include string */
parse_string(ctx, pc, 0, false);
return(true);
}
}
}
/* Check for Objective C literals and VALA identifiers ('@1', '@if')*/
if ((cpd.lang_flags & (LANG_OC | LANG_VALA)) && (ctx.peek() == '@'))
{
int nc = ctx.peek(1);
if ((nc == '"') || (nc == '\''))
{
/* literal string */
parse_string(ctx, pc, 1, true);
return(true);
}
else if ((nc >= '0') && (nc <= '9'))
{
/* literal number */
pc.str.append(ctx.get()); /* store the '@' */
parse_number(ctx, pc);
return(true);
}
}
/* Check for pawn/ObjectiveC/Java and normal identifiers */
if (CharTable::IsKw1(ctx.peek()) ||
((ctx.peek() == '@') && CharTable::IsKw1(ctx.peek(1))))
{
parse_word(ctx, pc, false);
return(true);
}
/* see if we have a punctuator */
char punc_txt[4];
punc_txt[0] = ctx.peek();
punc_txt[1] = ctx.peek(1);
punc_txt[2] = ctx.peek(2);
punc_txt[3] = ctx.peek(3);
if ((punc = find_punctuator(punc_txt, cpd.lang_flags)) != NULL)
{
int cnt = strlen(punc->tag);
while (cnt--)
{
pc.str.append(ctx.get());
}
pc.type = punc->type;
pc.flags |= PCF_PUNCTUATOR;
return(true);
}
/* throw away this character */
pc.type = CT_UNKNOWN;
pc.str.append(ctx.get());
LOG_FMT(LWARN, "%s:%d Garbage in col %d: %x\n",
cpd.filename, pc.orig_line, (int)ctx.c.col, pc.str[0]);
cpd.error_count++;
return(true);
} // parse_next
/**
* Check the backup-md5 file and copy the input file to a backup if needed.
*/
int backup_copy_file(const char *filename, const vector<UINT8> &data)
{
char newpath[1024];
char buffer[128];
char md5_str_in[33];
char md5_str[33];
UINT8 dig[16];
md5_str_in[0] = 0;
MD5::Calc(&data[0], data.size(), dig);
snprintf(md5_str, sizeof(md5_str),
"%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x\n",
dig[0], dig[1], dig[2], dig[3],
dig[4], dig[5], dig[6], dig[7],
dig[8], dig[9], dig[10], dig[11],
dig[12], dig[13], dig[14], dig[15]);
/* Create the backup-md5 filename, open it and read the md5 */
snprintf(newpath, sizeof(newpath), "%s%s", filename, UNC_BACKUP_MD5_SUFFIX);
FILE *thefile = fopen(newpath, "rb");
if (thefile != NULL)
{
if (fgets(buffer, sizeof(buffer), thefile) != NULL)
{
for (int i = 0; buffer[i] != 0; i++)
{
if (unc_isxdigit(buffer[i]))
{
md5_str_in[i] = unc_tolower(buffer[i]);
}
else
{
md5_str_in[i] = 0;
break;
}
}
}
fclose(thefile);
}
/* if the MD5s match, then there is no need to back up the file */
if (memcmp(md5_str, md5_str_in, 32) == 0)
{
LOG_FMT(LNOTE, "%s: MD5 match for %s\n", __func__, filename);
return(SUCCESS);
}
LOG_FMT(LNOTE, "%s: MD5 mismatch - backing up %s\n", __func__, filename);
/* Create the backup file */
snprintf(newpath, sizeof(newpath), "%s%s", filename, UNC_BACKUP_SUFFIX);
thefile = fopen(newpath, "wb");
if (thefile != NULL)
{
int retval = fwrite(&data[0], data.size(), 1, thefile);
int my_errno = errno;
fclose(thefile);
if (retval == 1)
{
return(SUCCESS);
}
LOG_FMT(LERR, "fwrite(%s) failed: %s (%d)\n",
newpath, strerror(my_errno), my_errno);
cpd.error_count++;
}
else
{
LOG_FMT(LERR, "fopen(%s) failed: %s (%d)\n",
newpath, strerror(errno), errno);
cpd.error_count++;
}
return(FAILURE);
} // backup_copy_file