• configuration doc: convert tabs to spaces

better commit message

Caveat: this does much more than just moving the -C option up one level, so please use a more appropriate commit message before landing.

Also: what are the cons of adding a umount command with just wraps fusermount -u?

We can't expect all users to read the tutorial, that just doesn't happen in the real world :-)
If we don't add an explicit umount command, we should at least document how to unmount also in the doc of the mount command

In T2702#50857, @haltode wrote:

To unmount you can use fusermount -u /mntdir

LGTM. (Please squash the 2 commits before merging though; 2 commits for a one-liner seem too much ;-))

I confirm this fixes the issue. LGTM, thanks !

While we wait for a proper solution, I think there is nothing stopping swh fuse from loading a configuration file by default if it exists and -C is not passed on the CLI.
We should just move -C up one level (from swh fuse mount to swh fuse) and do something similar to what swh-scanner does.
The location of the file might change in the future as a consequence of T1410, but meh, it's not a big deal.
We can stay on the safe side for now and only support ~/.config/swh/fuse/fuse.yml. And we will worry about merging with a global config file later.

please just fix the inconsistency in quoting in the example before landing, it is not clear to me why url is quoted whereas path is not

• can you please add info about where the configuration file is located? (both by default and w.r.t. XDG_CONFIG_HOME, if applicable)
• as people might not be familiar with YAML quirks, having an example of a skeleton config file, in full concrete syntax, would be also useful

In D4193#104293, @acezar wrote:

Changed --ignore-dir by --exclude

It supports glob, but had to copy extract_regex_objs from swh.scan.

It'd be useful to have a post-mortem on this, to make sure it doesn't happen again in the future.
Why wasn't @rdicosmo able to login? What has changed? The login URL? (if so, is the old URL still linked from anywhere?) Or were the old credentials temporarily unavailable? etc.

Thanks, even though this is a little bit disturbing discrepancy wrt swh-scanner exclusion mechanism, it's probably for the best, and arguably we should have something this simple also there.

Closed by @vsellier with rCJSWH5ade7a89a66b0ac6f12c00601f7e996f49e703e3

is this actually proposed for merging now? If so, please remove "WIP" from the diff title

In T2656#50089, @olasd wrote:

I've thought about it, but I don't know which component to mark as "degraded" currently. If anything, user-facing features like save code now and the vault should be going faster than usual :-)

When this is done (I guess in large part it is already, judging from the ingestion dashboard), please also update https://status.softwareheritage.org/ with a suitable message.

maybe I'm doing something wrong, but with the current version of this diff applied locally I get the following traceback when trying to FUSE-mount the archive:

$ mkdir tmp-archive
$ swh fuse mount tmp-archive
Traceback (most recent call last):
  File "/home/zack/.virtualenvs/swh/bin/swh", line 33, in <module>
    sys.exit(load_entry_point('swh.core', 'console_scripts', 'swh')())
  File "/home/zack/dati/projects/sw-heritage/git/swh-environment/swh-core/swh/core/cli/__init__.py", line 135, in main
    return swh(auto_envvar_prefix="SWH")
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/core.py", line 829, in __call__
    return self.main(*args, **kwargs)
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/core.py", line 782, in main
    rv = self.invoke(ctx)
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/core.py", line 1259, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/core.py", line 1066, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/core.py", line 610, in invoke
    return callback(*args, **kwargs)
  File "/home/zack/.virtualenvs/swh/lib/python3.8/site-packages/click/decorators.py", line 21, in new_func
    return f(get_current_context(), *args, **kwargs)
  File "/home/zack/dati/projects/sw-heritage/git/swh-environment/swh-fuse/swh/fuse/cli.py", line 120, in mount
    stack.enter(DaemonContext())
AttributeError: 'ExitStack' object has no attribute 'enter'

tested and confirmed—yes, I like it! thanks

Just a couple of minor things for me. I'll delegate to @seirl the remaining checks.
Good job !