Hi,
I would like pd to analyse a .txt file
for example, i have a .txt file with a load of words in, written in sentences etc. and i would like pd to go through the file and recognise when it reads an 'a', 'b', 'c' etc. if it reads an 'a' it would play a sound for example.
Ive done it with numbers before, but even then the numbers need to be a column:

1
2
3
4
5
6
7

not 1234567

is it possible for pd to do this? if not, does anyone know of some software / script that will analyse text files and format them into something pd will read?

thanks

james

[stripfilename] from iemlib looks like it will do the job

(or at least part of the job)

This simple command will do the job*:

sed s/./\\0\;/g <input.txt >output.txt

What it does is replace each character from input.txt with that character followed by a semi-colon. Then you can load output.txt with [textfile] and step through it one character at a time.

*It doesn't do the whole job - because some characters are special (like "{" "}" "\" ";" "," " " and probably some more). It would be easy to strip these characters out completely, but if you need to analyse them with Pd you could replace them with words like "bra" "ket" "backslash" etc

Or you could go the "human-unfriendly" route and replace all characters by their decimal values (again followed by a semicolon so that you can load it with [textfile]:

hexdump -v -e '1/1 "%d ;\n"' input.txt >output.txt

Both sed and hexdump are standard on unix systems like Linux and Mac OS X, but for Windows you can look into MinGW or CygWin.

Personally I would use the hexdump method, as numbers are generally a lot nicer to deal with than symbols, because there are objects like [moses] for numbers that don't exist for symbols...