Decode 'htmlsrc' to string before re.sub() #14
Conversation
Python 3 has distinct types for a byte array and a string, and requires explicit conversion between them. In order to make the script to work with Python 3, decode 'htmlsrc' to string before using the re.sub() API. Before writing the attachment to a file, encode it back to a byte array. Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@iki.fi>
|
Thanks! I guess I haven't been using viewhtmlmail much lately and hadn't realized it wasn't updated for python3. |
|
NP, Otherwise seems to be working :-) Thanks for the great script! Have been using it for some (quite many) years as part of my mutt setup. |
|
While fixing another issue (issue #15), I realized that it would be better to decode into the default (system) charset instead of forcing UTF-8, so the latest viewhtmlmail makes that change. I think it should still work the same on any system that uses UTF-8 by default (it did in the tests I tried), but if this causes any problems for you, let me know, reopen this issue or file a new one. I also made a simple unit test for viewhtmlmail, so any charset-related problems anyone encounters from now on can be added to the unit test so they won't regress. |
|
Does not cause problems, makes perfect sense. BTW, it might sense to encapsulate just this script to its own repository. There is a rise of new text based email clients like https://aerc-mail.org/. There was also one implemented in Rust of which name I cannot recall ATM. |
Python 3 has distinct types for a byte array and a string, and requires
explicit conversion between them. In order to make the script to work with
Python 3, decode 'htmlsrc' to string before applying using the re.sub() API.
Before writing the attachment to a file, encode it back to as a byte array.
Signed-off-by: Jarkko Sakkinen jarkko.sakkinen@iki.fi