* Use File.list and File.walk within a try with resource
The API contract of File.list and File.walk requires them to be closed after use.
* Fix from review
Left out filter
* Fix from review
Factored out deleteFile with better debug
* Fix from review
Can delete files whilst walking
* Fix from review
Restored sweepFile
fixed minor code suggestions
* Issue #5605 unconsumed input on sendError
Add Connection:close if content can't be consumed during a sendError. Processed after the request has returned to the container.
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* Update from review
+ Add close on all uncommitted requests when content cannot be consumed.
* Update from review
+ fixed comment
+ space comma
* Only consume input in COMPLETE if response is >=200 (ie not an upgrade or similar)
* Updated to be less adventurous
I do not think it was valid to always consumeAll in COMPLETE as this could break upgrades with both 101s and 200s
Instead I have reverted to having this consumeAll logic only:
+ in sendError once control has passed back to the container and we are about to generate an error page.
+ in front of all the sendRedirection that we do without calling the application first.
Extra tests also added
* Updated to be less adventurous
reverted test
* Testcase for odd sendError(400) issue.
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Fix for odd sendError(400) issue.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
* Testcase for odd sendError(400) issue.
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Always try to consumeAll on all requests
* Refinements after testing in 10
* Refinements after testing in 10
Fixed test
* Fixed comment from review
* Updates from review
+ added redirect methods that consumeAll
+ ensureContentConsumedOrConnectionClose renamed to ensureConsumeAllOrNotPersistent
+ ensureConsumeAllOrNotPersistent now handles HTTP/1.0 and HTTP/1.1 differently
* better consumeAll implementation
* update from review
+ better javadoc
+ filter out keep-alive
+ added more tests
* update from review
+ better javadoc
* update from review
+ fixed form redirection test for http 1.0 and 1.1
* update from review
+ HttpGenerator removes keep-alive if close present
+ Use isRedirection
Co-authored-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
Co-authored-by: Simone Bordet <simone.bordet@gmail.com>
* For #5650 protect from bad SslSession
Better protection from a bad SslSession like seen in #5650
* review indicated that there is too much depth to defend. So just a little cleanup in this PR.
* Cleanup request/response recycle #4711
Reordered recycle in request and response to field order so that we can check that all fields are recycled.
* Fixed ordering of reader consumption
* update from review
* Fix#5562 Improve HTTP Field cache allocation
Fix#5562 by initially putting cacheable fields into a inexpensive arraylist.
Only create the Trie (with space and complexity costs) if a second request is received.
* Fixed NPE
* Feedback from review
Create `HttpHeader.isPseudo()`` method
improved clarity with `createFieldCacheIfNeeded()``
* Feedback from review
Only defer Trie creation to first cacheable field, not until next request.
* Updates from review
* Update from review
+ more javadoc
+ empty set return
* SessionCookieConfig name may be null
Protect against NPE by make a null name in SessionCookieConfig deactive session cookies.
* SessionCookieConfig name may be null
Protect against NPE by make a null name in SessionCookieConfig deactive session cookies.
* SessionCookieConfig name may be null
Protect against NPE by make a null name in SessionCookieConfig deactive session cookies.
* feedback from review
added static method to convert null name to default.
Moved jetty-dir.css from jetty-util to jetty-server,
so that it can be found by ResourceHandler when using JPMS.
Updated DefaultServlet to call a ResourceHandler method
to retrieve the stylesheet.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
* Issue #5022 Filter Cache cleanup
Issue #5022 Filter Cache cleanup:
+ Fixed many compiler warnings
+ removed old LazyList leftovers
+ Don't create holder string for source unless required
+ Only have a single type of chain, so it can be wrapped regardless of cache
+ Reverse mappings lists to make filter chain creation easier
+ build chain directly rather than build a list then a chain
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* added comment to explain ordering
Signed-off-by: gregw <gregw@webtide.com>
* More cleanups
* fixed toString format
turn off debug in OSGI test
If SessionHandler.newHttpSession(Request) fails to create a session
it returns null. Request.getSession(true) cannot throw a checked
exception, nor can it return null, so we should throw ISE.
Signed-off-by: Jan Bartel <janb@webtide.com>
* Issue #5357 - Updating to https://eclipse.org/
- Removing redundant <url> refs in pom.xml
- Correcting bad indenting from merge
- Correcting mailing list references
- Correcting bugs.eclipse.org references
- Correcting text file references
- Correcting html references
- Correcting further references
- Correcting download.eclipse.org reference
- Adding test for demo-base /proxy/current/
- Ensuring jetty-client is included in javadoc-proxy.war/WEB-INF/lib
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
+ Merge ProxyPass tests from CheckReverseProxyHeadersTest into
ForwardedRequestCustomizerTest
+ Deleted CheckReverseProxyHeadersTest.java
+ Add more tests for ForcedHost configuration
+ Updated ForwardedRequestCustomizer to conform to expectations
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
+ In the case of HEAD, the servlet-api response is a wrapper
of javax.servlet.http.HttpServlet$NoBodyResponse
We know the content_length, use it.
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
If the request is async dispatched, the check state.isSuspended() is not
correct to determine if the request was async or not. The check
state.isAsyncStarted() should be used instead.
Signed-off-by: Lachlan Roberts <lachlan@webtide.com>
On the client:
* Origin.Address.host is passed through HostPort.normalizeHost(),
so that if it is IPv6 is bracketed.
Now the ipv6 address passed to an `HttClient` request is bracketed.
* HttpRequest was de-bracketing the host, but now it does not anymore.
On the server:
* Request.getLocalAddr(), getLocalName(), getRemoteAddr(),
getRemoteHost(), getServerName(), when dealing with an IPv6 address,
return it bracketed.
The reason to return bracketed IPv6 also from *Addr() methods is that
if it is used with InetAddress/InetSocketAddress it still works, but
often it is interpreted as a URI host so brackets are necessary.
* DoSFilter was blindly bracketing - now it does not.
Added a number of test cases, and fixed those that expected
non-bracketed IPv6.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
* Issue #5088 Review ContextHandler locking
The locking was primarily as a memory guard for the availability status, which was already volatile.
Have instead using an AtomicReference with a simple state machine layered on top of start/stop lifecycle.
There was also protection for AttributesMap, which is no longer needed as AttributesMap is now concurrent.
* Issue #5088
updates from review
* Issue #5088
updates from review (better this time)
* Fixes#5057 Included root context path
Root context path in include should be empty string.
* Issue #5057
merged context path methods as result of review.
ServletContent.getContextPath now returns the encoded contextPath (if anybody is silly enough to have one).
Fixes#4971 - Simplify Connection.upgradeFrom()/upgradeTo().
Now the upgrade-from connection produces a "floating" buffer
(not belonging to a pool), so that it can release the original buffer.
The upgrade-to connection is free to copy or store this "floating" buffer.
Strengthened ByteBufferPool behavior when releasing non-pooled
ByteBuffers: the buffer is now discarded.
Updated javadocs and all implementations.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
Fixes#4967 - Possible buffer corruption in HTTP/2 session failures
Partially reverted the changes introduced in #4855, because they
were working only when sends were synchronous.
Introduced ByteBufferPool.remove(ByteBuffer) to fix the issue.
Now when a concurrent failure happens while frames are being
generated or sent, the buffer is discarded instead of being
recycled, therefore resolving the buffer corruption.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
* Fixes#4855 - Occasional h2spec failures on CI
In case of bad usage of the HTTP/2 API, we don't want to close()
the stream but just fail the callback, because the stream
may be performing actions triggered by a legit API usage.
In case of a call to `AsyncListener.onError()`, applications may decide to call
AsyncContext.complete() and that would be a correct usage of the Servlet API.
This case was not well handled and was wrongly producing a WARN log with an
`IllegalStateException`.
Completely rewritten `HttpTransportOverHTTP2.TransportCallback`.
The rewrite handles correctly asynchronous failures that now are executed
sequentially (and not concurrently) with writes.
If a write is in progress, the failure will just change the state and at the
end of the write a check on the state will determine what actions to take.
A session failure is now handled in HTTP2Session by first failing all the
streams - which notifies the Stream.Listeners - and then failing the session
- which notifies the Session.Listener.
The stream failures are executed concurrently by dispatching each one to a
different thread; this means that the stream failure callbacks are executed
concurrently (likely sending RST_STREAM frames).
The session failure callback is completed only when all the stream failure
callbacks have completed, to ensure that a GOAWAY frame is processed after
all the RST_STREAM frames.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
* Issue #4936 - Adding LargeHeaderTest to replicate issue
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Issue #4936 - Updating LargeHeaderTest to use ServerConnector
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Issue #4936 - Fail LargeHeaderTest if client detects issues.
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Issue #4936 large response header buffer corruption
If the response buffer is too large, the header buffer was released
but not nulled, then an exception thrown, which again released the
not nulled buffer. The buffer thus ends up in the buffer pool twice!
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* Issue #4936 large response header buffer corruption
removed old comment
Signed-off-by: Greg Wilkins <gregw@webtide.com>
Co-authored-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Optimisation for single context
It is a frequent deployment mode to have only a single context.
In that case, the ContextHandlerCollection can bypass a bit of
looping/matching/selecting and just call the single context,
which it works out itself anyway if the request applies to it.
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* Optimisation for single context
updates from review
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* Spun out from #4814 Improve Attributes Handling
Improve attribute handling to reduce garbage and improve lookup.
Introduced a Wrapper so that request can remove any layers on reset.
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* Issue #4814 - Exposing AttributeMap.getAttributeNameSet() on Attributes.
The underlying AttributesMap already has a .getAttributeNameSet()
method, expose it on the Attributes interface.
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
* Allow a set to override a secure attribute.
Signed-off-by: Greg Wilkins <gregw@webtide.com>
* Issue #4814 - Attributes.getAttributeNames() is now defaulted
The Attributes.getAttributeNames() will use the
.getAttributeNameSet() by default now.
Updated all Attributes.Wrapper impls to use this new behavior
Signed-off-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
Co-authored-by: Joakim Erdfelt <joakim.erdfelt@gmail.com>
If SNI is required, wrap the KeyManagers with SniX509ExtendedKeyManager.
Updated the main keystore file to only have one certificate (instead of two),
since there never was the need for two certificates in the tests.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
Introduced NetworkTrafficSocketChannelEndPoint to replace
NetworkTrafficSelectChannelEndPoint, now deprecated.
Code and javadocs cleanup.
Moved the tests to jetty-client so that also the client is tested.
Signed-off-by: Simone Bordet <simone.bordet@gmail.com>
Whilst investigating #4711 for jetty-10, it was noticed that trailers are not nulled on recycled Response instances, nor on reset.
Signed-off-by: Greg Wilkins <gregw@webtide.com>
Greg Wilkins
Joakim Erdfelt
Lachlan Roberts
Simone Bordet
Jan Bartel
Thomas Draebing
olivier lamy
Ludovic Orban