why the c/c++ parser use the parser2?

[ollydbg]

Hi, from the source code c.c, I found that there are code like:

extern parserDefinition* CParser (void)
{
    static const char *const extensions [] = { "c", NULL };
    parserDefinition* def = parserNew ("C");
    def->kinds      = CKinds;
    def->kindCount  = KIND_COUNT (CKinds);
    def->extensions = extensions;
    def->parser2    = findCTags;
    def->initialize = initializeCParser;
    return def;
}
extern parserDefinition* CppParser (void)
{
    static const char *const extensions [] = {
        "c++", "cc", "cp", "cpp", "cxx", "h", "h++", "hh", "hp", "hpp", "hxx",
#ifndef CASE_INSENSITIVE_FILENAMES
        "C", "H",
#endif
        NULL
    };
    parserDefinition* def = parserNew ("C++");
    def->kinds      = CKinds;
    def->kindCount  = KIND_COUNT (CKinds);
    def->extensions = extensions;
    def->parser2    = findCTags;
    def->initialize = initializeCppParser;
    return def;
}

I just wondering that why the parser2 is used? it seems the parser field is left, it is default set to NULL?

thanks.

[ollydbg]

By the way, from the
http://ctags.sourceforge.net/EXTENDING.html

it seems that the "parser" field should be filled like:

/* Create parser definition stucture */
extern parserDefinition* SwineParser (void)
{
    static const char *const extensions [] = { "swn", NULL };
    parserDefinition* def = parserNew ("Swine");
    def->kinds      = SwineKinds;
    def->kindCount  = KIND_COUNT (SwineKinds);
    def->extensions = extensions;
    def->parser     = findSwineTags;
    return def;
}

So, why c/c++ parser use "parser2" field???

thanks.

[Steve]

A parserDefinition may use either parser or parser2, never both (and not neither). The difference is whether the parser is a "simple parser" or "rescanning parser". The findCTags parser can rescan with a different brace format if the first, normal, pass fails.