--- In [email protected], "entropyreduction"
<alancampbelllists+ya...@...> wrote:
>
>
> Try
>
> regexPlugin090824.zip in
> http://tech.groups.yahoo.com/group/power-pro/files/0_TEMP_/AlansPluginProvisional/
>
> In forgot to alter the pattern you provided for matching
> replacement patterns, now
>
> LPSTR pszPatternCaptures =
>
> "(?s)\\\\\\\\|\\\\\\$|\\\\[0-9]|\\$(?:U|u|L|l|T|t|X|x|Y|y|Z|z)?\\d{1,3}|\\$(?:U|u|L|l|T|t|X|x|Y|y|Z|z)?<[A-Za-z_1-9]{1,32}>|\\$\\#";
>
>
Lets try it this way, it should be a bit more efficient:
"(?si)\\\\\\\\|\\\\\\$|\\\\[0-9]|\\$[LUTXYZ]?(?:\\d{1,3}|<[A-Z_1-9]{1,32}>)|\\$\\#";
Also, please advance the plugin version and date.
> Script below does something, not sure if what you want.
The new letters are not being invalidated in the absence of utf8 plugin option.
I think that's ok; as is they work even when the new internal pattern utf8
option is used. It will be the user's responsibiltity to ensure L,U,T and X,Y,Z
are used only where appropriate. If either is used on incompatible subject
data, it can result in corrupted output.
In initial tests they seem to be working fine. Also, the utf8 plugin option is
now working as a multipurpose option (compile and accept-unicode-handles).
