When using ‘web_intelligent_plain_text_to_html’ from portal_transforms, make sure your input is decoded to utf-8 and the output is unicode.
Creating the option for users to create text in a textarea and save that on a object will always encounter encoding problems you did not test. Users will use charaters like é or ö in some kind of sentences. When displaying this message using web_intelligent_plain_text_to_html, to auto convert url’s or newlines you have to make sure you can handel them all.
Here we will have a tal snippet to display contect.getMessage:
<span tal:content="structure python: unicode(portal_transforms.convert( 'web_intelligent_plain_text_to_html', context.getMessage().decode('utf-8')).getData(), 'utf-8')"> The message </span>
This way getMessage may return éëê like characters and will still convert all urls to clickable url’s. web_intelligent_plain_text_to_html works pretty nice by default, and will convert all links to email or urls to clickable links and convert newlines to html br tags.
Nothing worse than getting:
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 284: ordinal not in range(128)