On Sun, Apr 14, 2019 at 5:40 AM x <tam118...@hotmail.com> wrote:
> On second thoughts JD, can’t use strlen or sqlite3_value_bytes in case
> values(1) contains more than a single unicode character. This looks OK.
>
> Bytes are what you need though; it doesn't matter how big the buffer is,
as long as you have all of it.
As long as you use _value_bytes after _text you're fine... so if any
conversion did take place the value will be right of the last returned
string type.
> # define CHARLEN(x) !(x & 128) ? 1 : (x & 16 ? 4 : (x & 32 ? 3 : 2))
>
> char *c = (char *)sqlite3_value_text(values[0]);
> char *Sep = (char *)sqlite3_value_text(values[1]);
> int Count=0, Len, SepLen = CHARLEN(*Sep);
>
> while (*c)
> {
> if ((Len = CHARLEN(*c)) == SepLen && memcmp(c, Sep, Len)==0)
> Count++; // at start of Sep
> c += Len;
> }
> sqlite3_result_int(ctx, Count);
>
> _______________________________________________
> sqlite-users mailing list
> sqlite-users@mailinglists.sqlite.org
> http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
>
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
