Open gerth2 opened 11 months ago
Why are you encoding it in utf-16? This definitely feels like a "don't do this" type of thing.
Also, does pip normally accept utf-16? Maybe on Windows, but probably not on Linux.
I don't have a driving reason to require this, though it may need to be documented since the failure mode is unexpected, especially for a new team.
I did a bit more investigation this morning. I'm not 100% sure how I created this issue on my end to start. However, on windows in powershell, running pip freeze > requirements.txt
will produce a utf-16 LE file. The same command in cmd
produces utf-8.
At least in most vsCode installs on windows, powershell is the default terminal.
I think simply leaving this issue around as docs would be sufficient, and just something to be on the lookout for in support for this coming year?
I may have some time this week to do more investigation. If it's a one-liner fix to handle serving all encodings as utf-8 it might save a few support calls. But certanly if it's more than that, probably not worth it.
I think it's best to force utf8 lf wherever possible.
cp1252 and utf 16 will undoubtedly lead to more issues in the future.
Problem description
Looks like extra 0x00 shows up on the RIO side when the local requirements.txt isn't utf-8
Operating System
Windows
Installed Python Packages
Reproducible example code