Aarni Koskela

This sounds like the file is corrupted. Since `wallet.dat` is a BerkeleyDB file, you could attempt using the [`db_dump`](https://www.manpagez.com/man/1/db_dump/) utility's `-r` option for recovery; something like ``` cp wallet.dat wallet.dat.bak...

The fix in `django-user-sessions` is to simply derive DUS's middleware from the built-in middleware instead of just `MiddlewareMixin`. For a quick workaround before that's implemented, as per my reply in...

@rbaron Commit 044eab6 makes `PRINTER_MODE` an unused variable, and thus `dark_mode` an unused parameter, and thus the `--darker` command line argument a no-op.

Hmm, since Travis passes mostly fine (and it builds using clang), I think this might be a gcc-only problem?

Actually, no, `clang++` fails the same way. :( I'm building 3fecdfa88189ba00d4ac6a2600822c5ad3f3f5d5 from source in a Docker container. I just wonder how Travis seems to pass alright with the current `master`...

Ah, this might be the thing: > If building against an external Mapnik please know that Mapnik Vector Tile does not currently support Mapnik 3.1.x. I'm currently on Mapnik `master`...

I bumped into this too – trying to e.g. ctrl+c out of py.test requires a couple of tries because `KeyboardInterrupt` (a `BaseException`) is caught where it really must not be.

@chayim Now that you mention it, perhaps 5.0.0 should split the nonfree modules out, á la the discussion in https://github.com/redis/redis-py/issues/1715?

Sounds like these speedups should be shipped as a separate opt-in module like `hiredis` is a separate opt-in parser for the protocol... (Even better, maybe this should be part of...

No, and it shouldn't. It looks like fastBPE isn't Windows compatible at this point.