Commit Graph

189 Commits

Author SHA1 Message Date
Carl Lerche
4c7ecf158d Add ReleaseCapacity handle. (#148)
This enables releasing stream capacity without having the `Body` handle.
2017-10-09 14:00:28 -07:00
Carl Lerche
56db50526d Don't unwrap logger init in tests 2017-10-09 13:10:52 -07:00
Sean McArthur
08a642ab11 fix ICE when compiling prioritization tests (#147) 2017-10-09 13:05:15 -07:00
Holt Chesley
2aee78c7d7 Issue 128: Convert frame::Reason to struct (#142)
Alter frame::Reason to a struct with a single u32 member.
Introduce Constants to the impl for existing Reasons. Change all usage
in the library and its tests to adopt this change,
using the new constants.
2017-10-08 13:13:07 -07:00
Sean McArthur
7b81be04aa reduce available window capacity when settings lowers window size 2017-10-06 14:20:32 -07:00
Sean McArthur
431442735d reset streams when receiving invalid psuedo headers 2017-10-06 13:48:30 -07:00
Sean McArthur
7d1732a70d change unimplemented to unreachable in handshake 2017-10-06 09:59:03 -07:00
Sean McArthur
720fb20bbf remove unused pieces from PingPong (#134)
Adds some extra tests as well, to be sure.
2017-10-05 19:16:14 -07:00
Eliza Weisman
2e3dcf602c StreamRef sends RST_STREAM on drop (#109)
This PR modifies the `Drop` implementation for `StreamRef` to reset the underlying stream if it is the last reference to that stream. Since both `Stream` and `Body` are internally just a `StreamRef`, this means they will both reset the stream on drop; thus, this closes #100.

The assertion that the store no longer contains the dropped stream ID at the end of the `Drop` method  had to be removed, as the stream has to be reset from inside of a `transition` block (which now manages releasing that ID for us), and the `transition` closure moves the value of `stream`, making the assertion no longer possible.

Modifications to some of the tests in `flow_control.rs` were also necessary, in order to prevent `StreamRef`s from being dropped too early.
2017-10-05 18:05:18 -05:00
Sean McArthur
ecd2764f4b when receiving a GOAWAY, allow earlier streams to still process (#133)
Once all active streams have finished, send a GOAWAY back and close the
connection.
2017-10-05 15:32:13 -07:00
Sean McArthur
c4ca8f7def Client::poll_ready and send_request may return Connection Errors (#132)
Closes #131
2017-10-04 15:22:10 -07:00
Sean McArthur
f8efb053b9 split Client into (Client, Connection) (#107)
The Connection type is a `Future` that drives all of the IO of the
client connection.

The Client type is separate, and is used to send requests into the
connection.
2017-09-28 16:55:12 -07:00
Oliver Gould
0b289fd55d Fix stream-id double-accounting bug (#112)
Both Recv::open and Rev::recv_headers check new stream ids against the
previously stream id. The second such check fails.

Now, only Recv::open performs stream id checks.

Fixes #110
2017-09-26 10:42:12 -07:00
Carl Lerche
8911ee2a4b Remove h2-test-support crate (#119)
The h2-test-support caused the unstable flag to always be enabled.
2017-09-25 22:28:39 -07:00
Oliver Gould
b99c513334 Apply rustfmt to test/support crate (#116) 2017-09-25 08:29:54 -07:00
Oliver Gould
dad113e17b Disallow nightly failures (#115)
Always install rustfmt since nightly may change underneath it, causing
linking to break.

Apply rustfmt
2017-09-24 19:25:50 -07:00
Eliza Weisman
a72a6bc8f0 Rewrite flow control tests to use newer mock API (#108)
I've rewritten the tests `flow_control::stream_close_by_data_frame_releases_capacity()` and `flow_control::stream_close_by_trailers_frame_releases_capacity()` to use the new mock API. This will make modifying these tests a bit easier in order to expect the correct behavior in my reset-on-drop branch.
2017-09-21 12:58:50 -05:00
Sean McArthur
6ec7f38cd7 add test for client sending over max concurrent limit (#105) 2017-09-19 15:31:35 -07:00
Sean McArthur
db8c109817 Client::poll_ready() returns an Error if next stream ID would overflow (#103)
Closes #102
2017-09-19 14:16:32 -07:00
Sean McArthur
0c8bd75224 check for StreamId overflow (#68) 2017-09-19 13:10:48 -07:00
Carl Lerche
e049dcb62b Rename Client::request -> send_request (#98)
All other fns have a send prefix.
2017-09-18 21:22:15 -07:00
Sean McArthur
a8a4cd2be1 add Client config to disable server push
- Adds `Client::builder().enable_push(false)` to disable push
- Client sends a GO_AWAY if receiving a push when it's disabled
2017-09-18 10:49:35 -07:00
Sean McArthur
c32015d48e add support for configuring max frame size
- Adds `max_frame_size` to client and server builders
- Pushes max_frame_size into Codec
- Detects when the Codec triggers an error from a frame too big
- Sends a GOAWAY when FRAME_SIZE_ERROR is encountered reading a frame
2017-09-14 17:03:43 -07:00
Carl Lerche
f84a1bdd1f Notify connection on connection window expansion (#86)
When capacity is released back to the connection and a connection level
window update needs to be sent out, the connection task needs to be
notified in order for the send to actually happen.
2017-09-14 13:50:52 -07:00
Sean McArthur
3ec0e85e56 add test when stream window overflows before conn window 2017-09-13 14:32:27 -07:00
Sean McArthur
e2cda1860b fix Body to return errors when there is recv error 2017-09-13 14:32:27 -07:00
Sean McArthur
ed472f109c add client::Builder to configure Clients 2017-09-13 14:32:27 -07:00
Sean McArthur
f7d14861e5 rustfmt: add trailing commas in match arms, set fn call to block stle (#85) 2017-09-12 19:29:06 -07:00
Oliver Gould
897bf84163 Use rustfmt to enforce consistent formatting
This change adds a .rustfmt.toml that includes ALL supported settings,
12 of which we have overridden to attempt to cater to our own
proclivities.

rustfmt is checked in the rust-nightly CI job.
2017-09-12 22:29:35 +00:00
Carl Lerche
93925e6d1f Limit send flow control bug to window_size (#78)
Senders could set the available capacity greater than the current
`window_size`.  This caused a panic when the sender attempted
to send more than the receiver could accept.
2017-09-12 10:48:11 -07:00
Eliza Weisman
11de86b34e Make Codec max frame length configurable (#57) 2017-09-11 12:56:40 -07:00
Sean McArthur
460afa41c8 Add connection window overflow test (#72) 2017-09-11 10:17:43 -07:00
Carl Lerche
5c0efcf8c4 Ref count stream state and release when final (#73)
Previously, stream state was never released so that long-lived connections
leaked memory.

Now, stream states are reference-counted and freed from the stream slab
when complete.  Locally reset streams are retained so that received frames
may be ignored.
2017-09-10 16:01:19 -07:00
Carl Lerche
daa54b9512 Rewrite a flow test 2017-09-08 13:42:14 -07:00
Oliver Gould
17bebe719a Increment stream window capacity when sending window update (#67) 2017-09-08 12:24:06 -07:00
Carl Lerche
c47717204c Flow control bug fix (#70)
The requested capacity was not decreased as data is written.
2017-09-08 12:15:46 -07:00
Sean McArthur
14f3f87da6 Merge pull request #66 from carllerche/flow-control-bugs
Fix some flow control bugs
2017-09-07 21:02:33 -07:00
Sean McArthur
b9e9fc64a7 Merge pull request #64 from seanmonstar/tests-ergo-frames
add ergonomics to testing with frames
2017-09-07 20:16:39 -07:00
Sean McArthur
018f00b8db update flow control tests to use frame builders 2017-09-07 17:06:09 -07:00
Sean McArthur
aef11623ac add recv_frame and send_frame extensions for Handle Futures 2017-09-07 17:05:43 -07:00
Sean McArthur
cc70ac1ef2 fix style to match crate 2017-09-07 17:05:07 -07:00
Carl Lerche
ae6bad6cef Fix some flow control bugs
When a stream is closed, connection level capacity should be released
back to the connection and then assigned to other streams waiting for
capacity.
2017-09-07 16:45:20 -07:00
Carl Lerche
9b42dafd24 Extract drive combinator to test support 2017-09-07 15:36:54 -07:00
Carl Lerche
38bbf30b2f Fix bug in prioritization (#63)
The stream buffered data counter was never decremented.
2017-09-07 14:12:21 -07:00
Sean McArthur
342d283cc5 add ergonomics to testing with frames 2017-09-07 13:09:57 -07:00
Sean McArthur
452e49dc3e aggregate WINDOW_UPDATE frames until change is over 50% available 2017-09-07 08:20:03 -07:00
Carl Lerche
cd76aca6d4 Add test infrastructure to work directly with frames (#56)
This adds a `Codec` based testing API. This is a bit less annoying than writing
at the raw H2 wire protocol level...
2017-09-06 14:18:37 -07:00
Carl Lerche
0cc611df35 Add Codec::set_max_send_frame_size 2017-09-05 14:01:32 -07:00
Carl Lerche
c2e6eb35d8 Track HTTP crate 2017-09-05 10:21:31 -07:00
Carl Lerche
b4fa5134f9 Data frame (#50)
* Rename DataFlag -> DataFlags
* Polish Data frame API
2017-09-05 10:00:05 -07:00