read text file, problem

n_wattam

i'm reading in a text file and found a major problem, it seems that the txt file is saved as unicode, its a chat log which i want to read in and display.

However because its saved as this format, there is a space after every character, you only see this though when doing source code view.. but appears normal on the webpage..

Is there a way to read the file in another format.. etc..

bradgrafelman

'Fraid I can't reproduce the error.. I read in a Unicode BOM file using [man]file_get_contents/man and it worked fine.

Are you sure there isn't actually an error in the way the log was produced?

EDIT: Actually, I did have some problems when the .txt file was encoded with "Unicode Big Endian" (Unicode BE BOM) format, though the malformed characters only appeared where line breaks occured.

Weedpacket

n_wattam wrote:
However because its saved as this format, there is a space after every character, you only see this though when doing source code view.. but appears normal on the webpage..

Those wouldn't be spaces, they're NUL characters (because every byte of the original is being saved as two bytes). NUL is chr(0), so you can just strip those out. I'm assuming that the original was plain ASCII.