class String

Public Instance Methods

fast_uxs_cgi() click to toggle source

Unescapes CGI, converting plus bytes '+' to space ' '

static VALUE fast_uxs_cgi(VALUE self)
{
        return _uxs_uri(self, 1);
}
fast_xs() click to toggle source

escapes strings for XML The double-quote (“) character is translated to ”"“

static VALUE fast_xs(VALUE self)
{
        long i;
        VALUE array;
        char *c;
        size_t s_len;
        VALUE *tmp;
        VALUE rv;

        array = rb_rescue(unpack_utf8, self, unpack_uchar, self);

        for (tmp = RARRAY_PTR(array), s_len = i = RARRAY_LEN(array);
             --i >= 0;
             tmp++) {
                int n = NUM2INT(*tmp);
                if (likely(n < 128)) {
                        if (unlikely(n == '"'))
                                s_len += (sizeof("&quot;") - 2);
                        if (unlikely(n == '&'))
                                s_len += (sizeof("&amp;") - 2);
                        if (unlikely(n == '>' || n == '<'))
                                s_len += (sizeof("&gt;") - 2);
                        continue;
                }

                CP_1252_ESCAPE(n);

                if (VALID_VALUE(n))
                        s_len += bytes_for(n) - 1;
        }

        rv = fast_xs_buf_new(self, s_len);
        c = RSTRING_PTR(rv);

        for (tmp = RARRAY_PTR(array), i = RARRAY_LEN(array); --i >= 0; tmp++)
                c += escape(c, NUM2INT(*tmp));

        return rv;
}
fast_xs_cgi() click to toggle source

Compatible with CGI::escape(), this iterates through each byte, so multibyte character sets may not supported (but UTF-8 should be).

static VALUE fast_xs_cgi(VALUE self)
{
        return _xs_uri_encode(self, 1);
}
fast_xs_html() click to toggle source

This is coding agnostic, and works on each byte, so some multibyte character sets may not be fully supported (but UTF-8 should be). This is meant to be 100% compatible with the ERB::Util::escape_html and CGI::escapeHTML methods

static VALUE fast_xs_html(VALUE self)
{
        long i;
        char *s;
        size_t new_len = RSTRING_LEN(self);
        char *new_str;
        VALUE rv;

        for (s = RSTRING_PTR(self), i = RSTRING_LEN(self); --i >= 0; ++s) {
                if (unlikely(*s == '&'))
                        new_len += (sizeof("&amp;") - 2);
                else if (unlikely(*s == '<' || *s == '>'))
                        new_len += (sizeof("&gt;") - 2);
                else if (unlikely(*s == '"'))
                        new_len += (sizeof("&quot;") - 2);
        }

        rv = fast_xs_buf_new(self, new_len);
        new_str = RSTRING_PTR(rv);

        for (s = RSTRING_PTR(self), i = RSTRING_LEN(self); --i >= 0; ++s) {
                if (unlikely(*s == '&'))
                        APPEND_CONST(new_str, "&amp;");
                else if (unlikely(*s == '<'))
                        APPEND_CONST(new_str, "&lt;");
                else if (unlikely(*s == '>'))
                        APPEND_CONST(new_str, "&gt;");
                else if (unlikely(*s == '"'))
                        APPEND_CONST(new_str, "&quot;");
                else
                        *new_str++ = *s;
        }

        return rv;
}
fast_xs_url() click to toggle source

Compatible with ERB::Util::url_encode / ERB::Util::u, this iterates through each byte, so multibyte character sets may not supported (but UTF-8 should be).

static VALUE fast_xs_url(VALUE self)
{
        return _xs_uri_encode(self, 0);
}