Format transformation in .txt

Good afternoon,

I made a MATLAB program that enters a ".txt" file, makes some changes on it, and outputs the same ".txt" file with the changes.

The problem is that the IMPUT ".txt" uses a conversion UNIX(LF) and a unicode transformation format UTF-16 LE. But the OUTPUT ".txt" is created by MATLAB with a conversion WINDOWS(CRLF) and an unicode transformation format UTF-8. That makes some symbols or the newline parameter not to be read properly.

How could I fix it?

Thanks in advanced

Sergio

[Walter Roberson]

If this is the same as your earlier question, the only input character that was non-ASCII was the paragraph symbol, char(182) . UTF encoding is not necessary in that case. Would it be acceptable to write it out without UTF encoding?

[Sergio Vez]

Not the same. I need the output file to have the same UNIX(LF) and UTF-16 LE like the imput file. I send you the program I have.

• In order to do proper changes I first delete the paragraph sign in MATLAB
• But then, I need to put it again but it only recognizes it as a symbol and not as a paragraph sign
• So then when I use the output file in another software it doesn't read it completly
• I guess it has something to do with the difference between UNIX(LF)/UTF-16 (input) vs. WINDOWS(CRLF)/UTF-8 (output).

Any idea on how to solve this?

Big thanks!

[Sergio Vez]

Forgot to attach

[Sergio Vez]

Any idea?

Thanks!

[Walter Roberson]

The file is not UTF16-LE . It is also not provably UTF-8 but it might be. It is CR+LF delimited (not LF only) lines with a maximum byte value of 182, which is the paragraph symbol.

[Sergio Vez]

https://drive.google.com/file/d/1QO_oBll_X2VL47FvcpCukV4PpUalbgs0/view?usp=sharing

[Sergio Vez]

This is the real imput file. In (.Z14). I made a mistake sending it in (.txt)

[Walter Roberson]

fileID = fopen('ENTRADA.Z14','rt','n','UTF16-LE');

and on output,

fileID = fopen('OUTPUT.txt', 'wt', 'n', 'UTF16-LE');

ignore the warnings about UTF16-LE not being supported.

There is a way to avoid having the warning generated (rather than ignored) but it involves doing unicode to native transformation. Not impossible by any means, but a bit of a nuisance.

[Sergio Vez]

Thank you so much! It worked!