Added setting to hide the header #614
Conversation
pudb/debugger.py
Outdated
| self.top._w.header = self.header if override else None | ||
| self.top._w.header = self.header if not CONFIG["hide_header"] else None |
There was a problem hiding this comment.
Is there an else missing here? Otherwise, it would seem the override logic would be ineffective. (Why is it needed anyhow?)
There was a problem hiding this comment.
Actually, there's a return missing here between these lines. Good catch. I knew something was off with it.
The override is just a glorified show_header function that I squished into a single function. In hindsight there's no reason to not just have a separate function for it. The purpose was for showing exceptions since the warning for it occurs in the header and it needs to be displayed even if the header is hidden.
I'll break it up into two clearer functions and you can see how you feel about it
inducer
force-pushed
the
thomryan/header_toggle
branch
from
f54b1eb to
d35e640
Compare
| elif exc_tuple is not None: | ||
| caption.extend([ | ||
| (None, " "), | ||
| ("header warning", "[PROCESSING EXCEPTION - hit 'e' to examine]") | ||
| ]) | ||
| self.show_header() | ||
| CONFIG["hide_header"] = False |
There was a problem hiding this comment.
This is a little weird; at this point the on-disk configuration will start disagreeing from the in-memory one. I'm guessing if one opens the preferences dialog, this changed config will then be persisted to disk, but not otherwise, which may be confusing.
Overall, I find this code a bit too stateful. I would like it better if there were simply one place that decides whether the header is shown (if configured so, or if an there's an exception), and then all the places that may cause state changes affecting this decision call that central place.
Now you see it
Now you don't
In reference to this issue
#542