CodeLobster wrote:would you mind in elaborating your test over diacritics a bit more, stefan?
it seems it will scan systems codepage am i right?
Your system use an codepage to translate keyboard scancodes (for an key) to an glyph,
in the paste there was only 7 bit codepage = 128 signs (ASCII), then there was an extended 8-bit table (256 signs) (extended ascii)
now there are unicode codepage which can hold thousands of different items.
But the base, the first 128 signs are still the same as the old standard ascii table.
The decimal code 48 till 57 stands for digits 0 till 9 (48=0, 49=1, 50=2,... 57=9)
65 till 90 are A-Z
97 till 122 are a-z
What else digits are from interest for you depends on your used codepage.
See your start menu > programs > accessories > system apps > code table (or smtg like that)
Or use this code to see the mapping digit<>glyph
Code: Select all
$i=33; $arr="";
while($i < 1640){ $arr = $arr . "$i=" . chr($i) . "<crlf>"; $i++;} text $arr;
The max number depends on your used codepage. I guess asians use higher number then european.
Now you can check which signs are from interest for you and use an code to check the asc() of each token of an string to see if it is an digit, an char or an sign.
Wasn't there an command to change diacritics to standard english chars, which could be used firstly?