[solved] csv output file in UTF-8 with BOM encoding

Boris_R

Hi all,

Is it possible to save csv output file (the one, which saved with the logger) in UTF-8 with BOM encoding?
By default it is saved as an UTF-8 without BOM.

Thank you and best regards,
Boris.

sebastiaan

Hi Boris,

With a small hack you can do this, yes. If you place the following script in the prepare phase of an inline_script at the beginning of your experiment, you will have a logfile with utf_8_sig encoding, which I believe corresponds to UTF8 with BOM.

# Close and re-open the logfile with a different codec.
import codecs
exp._log.close()
exp._log = codecs.open(self.get('logfile'), 'w', encoding='utf_8_sig')

You can use any standard codec, by the way, in case utf_8_sig is not the correct one:

• http://docs.python.org/2/library/codecs.html#standard-encodings

Just as a small aside, a BOM header is not necessary for UTF8 encoding. The reason is that UTF8 uses bytes as low-level units, instead of integers or some other unit that takes more than one byte. The byte order is therefore irrelevant, kind of like deciding whether you will read a single letter from left to right or right to left. But then again, you may have your reasons ;-)

Cheers,
Sebastiaan

Boris_R

Thanks a lot!