We now use our http::Response in HttpHelper::sendFileAndShutdown.
Change-Id: I77503fce7a66a11435bad896d3d5d6df611544be
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Since we don't want to propagate Poco types
as well as our pair containers, it's best
to have these are named functions rather
than as operator<<.
Change-Id: I9772b8f314262147b6c445464bf3ce8cb598b56d
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Since we already have an http response instance (in all but
one case, prior to this change), there is no need to pass
the mime-type explicitly to the sendFileAndShutdown
function, only to set it on the http response.
There are already too many arguments and the mime-type
is surely redundant.
Change-Id: Iab64074dc111573b87fb8fa9b907c26a4160910c
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
It seems that when TCP_NODELAY fails to be set,
the failure is permanent. As such, there is no
point in filling the logs with the same error.
This patch logs the error only once, per process,
and supresses further logs from Socket::setNoDelay().
Change-Id: I52c6b8cca35a8c281b4c4639d61a7e2521775d49
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
rearrange so coverity can see the size check
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: Ifad60b1a2aba2cf9c6cfbf4ea8e450b9846de576
in which case don't send it, which then implies we might have
an empty shareFDs which we can treat the same as a nonexisting
shareFDs
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: I80a78a01c69dbee5ee28a64442a5069a6c2b4dbe
The range header allows a client to specify that they would like only
part of a file, e.g. only the first 1000 bytes of a video. When playing
video on Apple devices (Macs, iPads, iPhones etc.) this is required to
avoid Safari rejecting the video as broken. This is the first part of a
fix to embedded videos on Safari.
This commit does not implement specifying multiple ranges at once (e.g.
bytes=0-9,-100 to get the first 10 and last 100 bytes).
This feature is necessary to play video on Apple devices, but it is not
sufficient. In particular, I believe some further client-side fixes will
be needed to make the video play properly.
Signed-off-by: Skyler Grey <skyler.grey@collabora.com>
Change-Id: Id89a06d374b7d0efbf2b3184d8618df61684dcb6
Instead of fighting the SolarMutex to get the messages
processed by the main loop. Simple and no additional
threading, mutexes etc.
messages from the external uno client are just
written to URPtoLoFD and the core reads from that
messages to the external uno client are written
to URPfromLoFD by core, that fd is in poll, and
activity there triggers a read by the DocBroker
to send it to the external uno client.
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: Ib1f0a0d5fb5ab22eee476d5d740b290c51de59dc
Failed with:
fuzzer/HttpEcho.cpp:111:17: error: 'removeSockets' is a private member of 'SocketPoll'
And:
kit/Delta.hpp:208:(.text._ZN14DeltaGenerator14DeltaBitmapRow7initRowEPKjj[_ZN14DeltaGenerator14DeltaBitmapRow7initRowEPKjj]+0x127): undefined reference to `simd_initPixRowSimd'
Signed-off-by: Miklos Vajna <vmiklos@collabora.com>
Change-Id: I6d980698b43ca3545d9eae5f40eabaf4e442ca17
Now that we depend on EnableExperimental in Log.cpp,
we must define it in the standalone binaries that
otherwise don't define it.
Change-Id: Ic56032eaf6df7d0d5d707a60eeddf2d75c9041ac
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
silence various of these in tests and utilities
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: I4fcc95b1b5d5c95e5ac40782550f95c08a3467bd
newRequest is always called with a const Request& already so might as
well make this copy inside the function rather than passing a copy
by value and moving it.
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: Ib572637d29590b8b470a42da3293065ed86d7dd0
iss ctor takes a const std::string& and constructs with a copy of str as content
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: I2805c869cb034a9b87e6e95cc58419b54a078b63
configure currently requires a min of c++17
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: I613dac6e66c3e77183d46623346969179bccc489
map is never null so make it a reference
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: I9f6ad863d0a1102dd7e26a5668b1c6ba71249580
std::make_shared and std::make_unique
are superior to explict smart pointer
construction. Where we have private
constructors, we can't use them.
Change-Id: I492122f58e958113c6e6d31445c6614ad98c89aa
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Using our own http library, we request
cool.html with realistic parameters and
validate many key values sent in the
request.
Change-Id: Ie3c0ac042f64bf5d6ed21354a33efac801b32846
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
We don't want to dump the whole paste buffer as hex, or indeed any
large mis-encoded string - instead crop to the problematic section
for easier debugging, and no performance hit.
Change-Id: I30518beea436895a42b3429c3cb6e16e2093cca9
Signed-off-by: Michael Meeks <michael.meeks@collabora.com>
Revert "Avoid crash when pasting image in debug mode"
This reverts commit 78558fe9af.
Instead of detecting paste command and not trigerring assertion:
fix frame type so it will be binary in case of paste.
Signed-off-by: Szymon Kłos <szymon.klos@collabora.com>
Change-Id: If5c726463459be7eaffae1522df522dfa1d74443
If COOL is running in debug mode ther is a
check that incoming messages are UTF-8.
We skip this tests if the message refers
to an image pasted.
Signed-off-by: luigiiucci <luigi.iucci@collabora.com>
Change-Id: I4f7e4a647deb22ec2a032d4d6347d910184c0701
Fortunately, we have no need for any locals
in this case.
Change-Id: I954953bb2a8c79662152ad4cc2fb5422bc2b0baa
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Previously, we didn't expect such exceptions,
but increasingly this can happen, as we have
more and more async processing. The main
issue with not catching exceptions is that
we do not properly discard read and
parsed data from the socket. This is a
major issue, of course.
Change-Id: I4490baed7069cd55d103301c1d3be31d9b2d9b2c
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
github_android_online_co-23.05/net/HttpRequest.hpp:1391:63: error: no matching function for call to 'min'
<< Util::dumpHex(std::string(data.data(), std::min(data.size(), 256UL))));
^~~~~~~~
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: I5b81096c696565b9e4eb4f3ca39c16bc1e5b9182
Remove un-necessary gating of wakeup() on stop un-related to
callbacks or new sockets.
This reverts commit 25f2581a30.
Change-Id: I9f9f30fed34c973b86206677168071abb81afa6f
Signed-off-by: Michael Meeks <michael.meeks@collabora.com>
UnitBase::isUnitTesting() and UnitWSD::isUnitTesting() to avoid
calling methods that crash on mobile.
Signed-off-by: Patrick Luby <plubius@neooffice.org>
Change-Id: Ie6bc2d238ac9e475b02ceef5809c55acc129dd52
Still needed to add a bunch of #if !MOBILAPP to disable all of the
Online server's unit testing classes so that the iOS app won't crash
when opening a document.
TODO: the iOS app will not accept text input at all. Is some code
unexpectedly consuming message data sent over the mobile fake socket?
Signed-off-by: Patrick Luby <plubius@neooffice.org>
Change-Id: Ie874ee8e452188a7b4cdf6d5f87708e5ed9b0235
probably went astray with:
commit 52f2600cf9
Date: Wed Feb 16 17:22:18 2022 -0500
wsd: test: embue all tests with a name and unify logs
Signed-off-by: Caolán McNamara <caolan.mcnamara@collabora.com>
Change-Id: Idf4f9bfce7ef188ac15a7cf62baadc3dd5b59dcd
Shorter waits to break the polling
loop when the test finishes and we
need to terminate.
Change-Id: I79d9d89675d0a37f25a03b265b3ae636a90e57d5
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
This merges multiple implementations of
assertCorrectThread and simplifies its
usage.
Change-Id: I7be5dea62c6046fb0412d7f885fcbcc4b66e3fd9
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
The normal path is now streamlined and error handling
is at the end. We also now always set the disposition
to closed and fire onDisconnect() when disconnected
or hit an error.
Change-Id: I984ad71601b92b8042dc7984e7339f0804c8083b
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
We use our own status codes, which
are type-safe and use streaming
operators to serialize and log.
Change-Id: I0eba7b16694866b5a79476a7ef4b1b78f7f9c176
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
This is needed to correct overload
resolution of these operators.
Change-Id: I02c0859674efe112102a8d3833bfb0a30b1a6574
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
For some reason we only checked for the closed
socket case as failure and returned 0 in that case,
for error cases, we returned 1. Likely the API
had been modified in the early days, but this
return code was left lopsided.
This meant that even when the handshake failed, we
still called readIncomingData or writeOutgoindData,
depending on whether we wanted to read or write,
causing a rare race-condition.
When a client (HTTP request) connects to a server,
it needs to send the request, right after the
SSL handshake. SSL_do_handshake could need data
from the socket to complete the handshake. In such
a case it returns WANT_READ. Unfortunately,
because we always called SSL_read, the missing data
could have arrived between the SSL_do_handshake call
and the SSL_read call (a rather short duration, to
be sure, but an open window all the same).
SSL_read would of course read said data from the
socket and, since it still needs to finish the
handshake, will buffer it. It then returns the very
same error that the SSL_do_handshake returned:
WANT_READ. Of course we will oblige by polling with
POLLIN, which will time out (there is no more data
to come, and the server is waiting for *our* request
and has nothing to send us).
The only way this deadlock could break if
SSL_do_handshake was called (which will consume
the buffered data, return 1 to indicate handshake
has completed). Since we wouldn't call it unless
and until we get POLLIN, per WANT_READ, which won't
happen in this case. And since SSL_read doesn't call
SSL_do_handshake either, the request times out and
that's the end of it.
The fix is to not call SSL_read when the handshake
isn't complete and needs more data, which we do now.
Fortunately, we have very few SSL clients, outside
of unit-tests. Most notably the WOPI client. But
even then it's not a heavily used connection and
might not even be SSL-enabled (for LAN servers).
Change-Id: I04fd3dae151904194f3d7579dbf8c671b2580ffb
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
A rare edge-case happens when we have a socket
that is closed (and we get POLLHUP or POLLERR)
but upon reading the socket we get EAGAIN.
This was observed when simulating EAGAIN,
and it is possible that this is quite impossible
in practice (since we read only when we get
POLLIN), but at least for the unit-tests
we need to handle this case, so we don't
get random failures.
Change-Id: I77af1726066507af5d5ada68fe11b479a4e579e5
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
We don't have to read the wakeup pipe
under a lock. So we no longer do.
Change-Id: I6bd724b9748add3022b4f9aa2268094b9818f3e5
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
We now always handle socket polls without skipping
when the number of sockets change. We verify the
socket FD of each socket before handlingPoll to
catch any violations of the preconditions.
This should avoid missing any events and also
handles timeouts better (which are checked
in handlePoll). It also protects against
unexpected modifications of the sockets, with
proper logging and assertion.
Change-Id: I5659eb57231a490e6c813e7a0222443b534713c6
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Reading larger blocks should avoid having
the pipe back up when we accidentally
attempt more wakes than actual ones.
Change-Id: I7766230f60dbf069668ee7919f766e9093df7017
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
There is no use-case for calling the
thread function of the SocketPoll
from outside.
Change-Id: Id8e87369494817aaab749d03d1cd4cd3724c2da1
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
This replaces the hard-coded status-code
numbers with named compile-time constants.
Change-Id: Ibe678fb2c533b29efd696e4430f5377523eeb298
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
I had added some extra debugging output to the !MOBILEAPP branch to
log values of timeoutMaxMicroS and timeoutMaxMS and was wondering why
I saw things like:
timeoutMaxMicroS=5000000 timeoutMaxMS=5009
and
timeoutMaxMicroS=0 timeoutMaxMS=9
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: Iac8599ce5b00ef90d62eabc29c5d92858e276bb6
Apparently it woke up the world unnecessarily.
Change-Id: Iad65215da898b017860e7d7b803771f657a3e1ab
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
When shutting down, SocketPoll threads asynchronously
stop and exit. This fundamental race means the warning
is useless and noisy.
Change-Id: I3ca9044c9a68689abb7e8f692fffd10509eadab6
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
sendCloseFrame does check for the socket to be
there, but it logs an error if it isn't because
it assumes we expect to have a socket.
Since we are shutting down, this code could
be triggered on destruction when we no
longer have the socket.
Change-Id: I622a24394632159aa71c1c21c33c0f8a2d5c6250
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
It seems we were always pinging immediately on a
newly created WebSocket, even before upgrading to
WS, which logged a warning. That was because we
create the WebSocketHandler instance with a
timed-out lastPingSentTime.
Here we correctly set lastPingSentTime such that
it times out *after* InitialPingDelayMicroS has
elapsed. This gives it time to upgrade the
socket to WS and avoid the warning.
Change-Id: I6004348b9b4bd29f614d9e010fb7649da2bca338
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
We no longer emit a fatal (FTL) log when
forced-exiting successfully.
Also, improved are some other logs to
better reflect the severity of the issue.
Change-Id: I22e79f685825f7ecd47cec76c9be9683deff2d55
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Messages are high-frequency events and logging
them at debug level is too noisy. Increase the
level to trace.
Debug level should produce legible entries that
outline the main activities, rather than log
each message. That is best done at trace level.
Change-Id: I722ab8b58e0adcab6ecb2f8c571966af0d952051
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Ashod Nakashian
Michael Meeks
Caolán McNamara
Skyler Grey
Miklos Vajna
Szymon Kłos
luigiiucci
Patrick Luby
Shehr Bano
FedericoHeichou
Tor Lillqvist