| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| ... | |
| | | |
| | |
| | |
| | |
| | | |
This addresses the TODO that resulted from this thread
https://gitlab.torproject.org/tpo/core/arti/-/merge_requests/1677#note_2955700
|
| | | |
| | |
| | |
| | |
| | |
| | | |
(The actual false positive rate for these cases is very small, since
the filters are very much not full. I haven't been able to hit one
in hours of looping.)
|
| | | | |
|
| | | | |
|
| | | |
| | |
| | |
| | |
| | | |
Right now, this is always an ephemeral log, which isn't right: There
are TODO comments here explaining what to do instead.
|
| | | |
| | |
| | |
| | | |
See comments for design notes.
|
| |\ \ \
| |/ /
|/| |
| | |
| | | |
HSS misc tweaks
See merge request tpo/core/arti!1718
|
| | | | |
|
| | | | |
|
| |\ \ \
| | | |
| | | |
| | | |
| | | | |
tor-keymgr: Fix a broken docs link
See merge request tpo/core/arti!1721
|
| | |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Fixes:
cargo doc --workspace $f --document-private-items
1 | //! [`KeySpecifier`] implementations for hidden service keys.
| ^^^^^^^^^^^^ no item named `KeySpecifier` in scope
8 | /// A helper for defining service [`KeySpecifier`]s.
| ^^^^^^^^^^^^ no item named `KeySpecifier` in scope
|
| |/ /
| |
| |
| |
| | |
`cargo clippy --all-features --tests` is reporting warnings about
the unnecessary cloning.
|
| |\ \
| | |
| | |
| | |
| | |
| | |
| | | |
tor-keymgr: Add macros for defining key specifiers.
Closes #1069
See merge request tpo/core/arti!1710
|
| | | | |
|
| | | | |
|
| | | |
| | |
| | |
| | | |
Closes #1069
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | |
| | |
| | |
| | |
| | | |
The trait is no longer sealed because users should be allowed to define
their own key denotators.
|
| | | |
| | |
| | |
| | |
| | |
| | | |
I think retaining this lint even in tests is probably reasonable.
IptsPublisherView has interior mutability, so with this function
taking one, it won't ever need mut.
|
| |\ \ \
| | | |
| | | |
| | | |
| | | | |
timeout_track improvements
See merge request tpo/core/arti!1711
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
We have a profusion of similar methods with similar names and similar
documentation. Centralising the documentation in a trait is helpful.
This also means that callers can just naturally call `.update` rather
than needing to spell out `.update_abs` vs `.update_rel` etc.
|
| | |/ /
| | |
| | |
| | |
| | | |
Clarify this in the update functions, and in the general discussion.
We don't want zero timeouts.
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Representing the supported HTYPEs as `HandshakeType`s instead of `u32`s
makes it more difficult to pass in wrong/invalid values to
`HsDescBuilder::create2_formats`.
This also fixes a descriptor publisher bug spotted by @jnewsome, where
the advertised CREATE2 HTYPEs included HTYPE `1`, which is actually
supposed to be a reserved value. The publisher now only advertises the
`NTOR` HTYPE (just like C Tor).
|
| | |
| |
| |
| |
| |
| | |
`Bug` wasn't necessarily the right error type here. Plus, with the new
error type adding new errors (i.e. `CertEncodeError` variants), is not a
breaking change.
|
| | |
| |
| |
| | |
blinded_id.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
argument.
This will enable us to (eventually) the load the descriptor signing key
cert from the keystore (as opposed to always recomputing it when
building the `HsDesc`).
Part of #1048
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Enough of them have turned out to need to be async so far that I
don't think it's a great idea to commit to making any of them
not-async forever. Additionally, we've written the code that uses
these, and having them be async didn't seem to cause any major
trouble.
By the same "we may want to make these more complicated underneath"
argument, we should IMO leave them as fallible.
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| | |
This particular public key is not stored in the keystore (though maybe
it should be).
|
| | |
| |
| |
| |
| |
| | |
The pattern was only supposed to match the 3 components of the time
period, separated by `_`, but not the leading `_`, which separates the
key name from its denotators.
|
| | |
| |
| |
| |
| |
| | |
We don't need to be logging the entire descriptor for each upload
attempt (logging descriptors immediately after they are generated should
be sufficient).
|
| | |
| |
| |
| |
| |
| | |
The message is supposed to print how many HSDirs the descriptor was
uploaded to (previously it was showing the
`<success_count>/<failure_count>` which is pretty unintuitive).
|
| | | |
|
| |/ |
|
| |\
| |
| |
| |
| | |
Fix rustdoc link warnings/errors.
See merge request tpo/core/arti!1690
|
| | | |
|
| |\ \
| | |
| | |
| | |
| | | |
tor-keymgr: Auto-generate missing keys in descriptor publisher
See merge request tpo/core/arti!1688
|
| | | |
| | |
| | |
| | |
| | | |
This is not an error, it just means we need to wait until some IPTs are
established.
|
| | | | |
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
We don't need this helper anymore (we no longer map `None` to
`Err(MissingKey)`, because the new get-or-create functions don't return
an `Option`).
Also, the key lookups are going to look less uniform from now on
(because some will be auto-generated with `get_or_generate`, and others
with `get_or_generate_with_derived`).
|
