ddd323c6ff
tor-hsservice: Fix an erroneous reference to an hs-client feature
2023-02-28 11:10:17 +00:00
f63c7c48af
Rename onion-* cargo features to hs-* in tor-* crates
...
Fixes #756
2023-02-28 11:10:17 +00:00
fcbeceeb61
Merge branch 'cmd_checker' into 'main'
...
tor-proto: Introduce CmdChecker, and use it to enforce correctness for our streams.
Closes #774 and #769
See merge request tpo/core/arti!1026
2023-02-27 16:43:34 +00:00
07473ed1fc
proto: Tweak documentation and names around CmdChecker.
2023-02-27 08:14:15 -05:00
d09273a6e3
Apply a missing cfg_attr(docsrs...) in netdoc
2023-02-21 15:18:46 +00:00
c1fbcb34f6
Rearrange cfg_attr in relaycell::msg
...
docsrs wants to find its `cfg_attr(docsrs...)` line after the
`cfg()` line.
2023-02-21 15:18:46 +00:00
750c512264
Merge branch 'unparsed_linkspec' into 'main'
...
HsDesc: Use a new UnparsedLinkSpecifier to avoid leaking which linkspec types we know
See merge request tpo/core/arti!1029
2023-02-21 15:16:45 +00:00
67234568bc
Merge branch 'hsdir_remove_comment' into 'main'
...
netdoc: Remove a TODO hs comment.
See merge request tpo/core/arti!1028
2023-02-21 15:15:10 +00:00
3233c25943
tor-netdoc: Use UnparsedLinkSpec in IntroPointDesc.
2023-02-17 15:29:37 -05:00
bc68974212
tor-linkspec: New UnparsedLinkspec type.
...
Unlike linkspec, this doesn't validate the actual contents of the
specifiers. We'll use this so we can handle the linkspec list for an
introduction point in an HsDesc, and just pass it on when
constructing our circuits.
I haven't added any accessor or constructor functions, because I
don't expect to need them.
2023-02-17 15:26:23 -05:00
757c6d333f
netdoc: Remove a TODO hs comment.
...
It said to check whether C enforces an absence of extraneous bytes
at the end of the link specifiers. It does, in
`hs_desc.c:decode_link_specifiers()`, where it says:
```
if (link_specifier_list_parse(&specs, decoded,
(size_t) decoded_len) < decoded_len) {
goto err;
}
```
The comparison with "decoded_len" checks whether all the bytes were
decoded.
2023-02-17 15:06:09 -05:00
2b515c8d25
tor-dirmgr: Work around an apparent regression in `time`.
...
Some code in our tests that worked fine with time 0.3.17 no
longer works with 0.3.19, despite the semver.
See https://github.com/time-rs/time/issues/552 for the upstream bug.
2023-02-17 14:36:19 -05:00
814563dfd6
tor-proto: Rename end_received.
...
It can now indicate _any_ cell that means we can forget about a stream.
2023-02-17 11:56:53 -05:00
d81a9c9aad
Use CmdChecker in our stream handling code.
...
This change makes sure that open streams and half-closed streams
have the same stream-type-dependent state machines with respect to
which cells are acceptable.
Fixes #774 .
Fixes #769 .
2023-02-17 11:52:09 -05:00
c5be3c089e
tor-proto: Introduce CmdChecker, and define it for our streams.
...
The role of CmdChecker is to verify that messages are arriving at
the appropriate sequence on a stream, with respect to the other
messages that have been received. Once the stream becomes
half-closed, the CmdChecker is also in charge of consuming incoming
messages on the stream and making sure that they are well-formed.
2023-02-17 11:50:27 -05:00
1297dfb5e3
Merge branch 'doc-iter' into 'main'
...
tor-netdoc: Abolish PauseAt in favour of using itertools
Closes #760
See merge request tpo/core/arti!1021
2023-02-17 12:34:32 +00:00
878531af90
Merge branch 'hs-cells-1' into 'main'
...
Start refactoring hs cell implementations
See merge request tpo/core/arti!1020
2023-02-16 15:36:31 +00:00
c4b451ffa6
Mark a quoted block as text, so doctest ignores it.
2023-02-16 09:00:07 -05:00
14417dc098
Add a TODO about combining several macros and patterns
2023-02-16 08:25:12 -05:00
b4e48b0695
Make a comment more accurate
2023-02-16 13:19:04 +00:00
66beab6c9e
tor-netdoc: Fix doc comment following removal of PauseAt
2023-02-16 11:53:32 +00:00
ca127b013f
Expire routerdescs as soon as any of their expiries expire
...
Fixes #772
2023-02-15 17:41:28 +00:00
676241c349
tor-netdoc: Run rustfmt (separately from the refactoring)
2023-02-15 17:17:01 +00:00
1993165f22
tor-netdoc: Abolish NetDocReader::iter and into_iter
...
Nothing uses these now; the NetDocReader is simply an iterator, itself.
2023-02-15 17:17:01 +00:00
4a91c732fe
tor-netdoc: Remove many needless calls to .iter() and .into_iter()
2023-02-15 17:17:01 +00:00
fa2e164939
tor-netdoc: Use impl Iterator for NetdocReader in .pause_at
...
This simplifies the return type!
2023-02-15 17:17:01 +00:00
db79e3f369
tor-netdoc: impl Iterator etc. for NetdocReader
...
This will simplify things at many call sites.
2023-02-15 17:17:01 +00:00
b58977b428
tor-netdoc: move PeekableIterator to util.rs
...
We're going to want this a bit more widely.
2023-02-15 17:17:01 +00:00
11660f84b0
tor-netdoc: Delete now-unused PauseAt type
2023-02-15 17:17:01 +00:00
15d93bd642
tor-netdoc: Use itertools to implement NetDocReader::pause_at
...
This is the last use of PauseAt.
2023-02-15 17:17:01 +00:00
089ebb7d11
tor-netdoc: Replace use of PauseAt::new_pred()
...
We can just make a new PauseAt, since it can take the inner peekable
iterator by mutable reference.
This seems to tidy the code up a bit too.
2023-02-15 17:16:57 +00:00
7491424ee9
tor-netdoc: Have sectionrules.parse take I, not &mut I
...
There's a blanket impl of Itertor for &mut impl Iterator, so this
isn't necessary, and it prevents us passing iterators by value
producing syntactic vinegar.
2023-02-15 17:16:57 +00:00
80d8d94c96
tor-netdoc: Import Peekable rather than referring to it by path
...
This makes things easier to read. We're about to add another
site (albeit, temporarily).
2023-02-15 17:16:57 +00:00
0f2218f4d3
Merge branch 'ticket_525_part3_take2' into 'main'
...
Finish #525 for relay messages: Only parse messages at the last instant.
Closes #773 and #525
See merge request tpo/core/arti!1017
2023-02-15 16:35:27 +00:00
19c0dd153a
tor-proto: Add a TODO about simplifying a common pattern.
2023-02-15 10:51:03 -05:00
21db73f182
tor-cell: add a TODO comment about simplifying Body away.
2023-02-15 10:48:19 -05:00
fce1c83f2e
tor-cell: Add another debug_assert to relay cell encoding
2023-02-15 10:48:19 -05:00
5d28e9e7d0
slicewriter: rename a local variable.
2023-02-15 10:48:19 -05:00
19c9593acf
Move slicewriter to tor-cell and make it private.
...
Also, add some comments about how it is likely to change.
2023-02-15 10:48:19 -05:00
1ee6bfa59c
tor-proto: note implications for future HS work
2023-02-15 10:48:19 -05:00
41b50b6c56
tor-proto: Push stream message parsing into the stream objects.
...
This closes #525 , and ensures, at last, that we don't parse any
message that we wouldn't accept.
2023-02-15 10:48:19 -05:00
58c3b8276c
tor-proto: Defer parsing of messages send to half-closed streams
...
This includes a partial solution for #769 , but also turned up
another bug (#774 ) while I was working on it. I'll close them both
once I have a real solution.
2023-02-15 10:48:19 -05:00
bd0f6f5adf
tor-proto: stop reactor (and kill circuit) if meta handler fails
...
If the meta handler reports an error, then the circuit has violated
its protocol, and needs to be shut down.
Fixes #773 .
2023-02-15 10:48:19 -05:00
2e483124cb
tor-proto: defer meta-cell parsing to the last moment.
2023-02-15 10:48:19 -05:00
0765243f5e
tor-proto: Use UnparsedRelayCell to start deferring cell processing.
...
In general, we want to avoid parsing these cells until we are
fairly sure that they are something we would accept.
2023-02-15 10:48:19 -05:00
e4bc7ef57b
tor-cell: Add RelayCell::into_msg.
2023-02-15 10:48:19 -05:00
3f1457ea04
tor-cell: Implement {Relay,Chan}Msg for every body type
...
This will make it ergonomic to decode a single body type without
having to declare a variant that accepts only a single message.
2023-02-15 10:48:19 -05:00
65cc7d0974
tor-cell: Note an opportunity for future optimization
2023-02-15 10:48:19 -05:00
a809a809ba
tor-cell: Add a new UnparsedRelayCell
...
We'll use this to router relay messages on a circuit to the
appropriate stream, and hand them to that stream, without parsing
the message until the stream has been determined.
2023-02-15 10:48:19 -05:00
ca3b33a1af
tor-cell: Refactor relay cells to copy much less
...
We now manipulate raw relay cell bodies as (an alias for)
`Box<[u8;509]>` rather than as (an alias for) `[u8;509]`. This
enables us to do much less copying. It will become more important
soon, as we defer parsing relay cell bodies even longer.
Related to #7 .
We also use SliceWriter to avoid allocating a Vec<> for every relay
message we want to encode, and instead encode directly into the
cell.
2023-02-15 10:48:19 -05:00
Ian Jackson
Nick Mathewson
Ian Jackson