And improve the logging support in unit-tests to
help troubleshoot issues faster and more accurately.
Also makes the code more readable (hopefully).
Change-Id: I4f8aafb5245e2f774b03231591a74544f9ec84aa
Reviewed-on: https://gerrit.libreoffice.org/48645
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
And we read the "LastModifiedTime" from the JSON payload in PutFile
response, not from the headers. Change that so that wsd now reads the
correct time.
Change-Id: Ic930ef4cfb2744367d04ba460fa1cbeb045b7356
... instead of directly stop()ing it. The close request approach also
makes sure that outgoing document is saved to storage.
Change-Id: I44f61db00dbd326dec80f59f4a2cbb617048aa94
Reviewed-on: https://gerrit.libreoffice.org/49122
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Now all logging is done after checking if the
level in question is enabled or not (thanks to
the macros LOG_XXX), which saves unnecessary
conversions and stringification when said level
is disabled.
Change-Id: Icde31e067f60269563896f04f8b0d65643720766
Reviewed-on: https://gerrit.libreoffice.org/47885
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Michael Meeks <michael.meeks@collabora.com>
"loolnb" used to mean the concept of using non-blocking sockets back
when that was experimental proof-of-concept work in progress, but that
work has been merged in long ago and there is no neeed for the term
any longer.
Change-Id: I4c49b625fe8be8892b82115eaaa47bcba0f29500
Reviewed-on: https://gerrit.libreoffice.org/47699
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Michael Meeks <michael.meeks@collabora.com>
Trying to combine the Poco's http server together with our polling loop leads
only to problem; so instead let's introduce a hook where we can do the WOPI
serving directly in the unit test.
Change-Id: Id3fec6ff93c3ad652aa4e0fc6309c5b7639728cb
MS-WOPI:
"The following HTTP header MUST be included in all WOPI requests."
"Authorization "Bearer" and <token>"
Change-Id: Ie5327286e93b86a49a0248ff5886bf3ea99df768
Changes protocol to use 'wid' instead of 'hash' everywhere. Wire-ids
are monotonically increasing integers that can be mapped to hash
values for all of the hash values and tiles we cache internally.
Change-Id: Ibcb25817bab0f453e93d52a6f99d3ff65059e47d
make check fails for me like this:
Test name: HTTPCrashTest::testCrashKit
equality assertion failed
- Expected: 1001
- Actual : 65535
Failures !!!
Run: 1 Failure total: 1 Failures: 1 Errors: 0
But when I run loolwsd and ./test manually (./test is invoked by gdb) and I
step through the code, then the test passes. So I guess what happens is that we
read from the socket too fast, and the error we're looking for is just not
there yet. Add the same amount of sleep here (0.5s) than what's used in
connectLOKit(), with that the test passes fine.
(The sleep is in test-only code.)
Change-Id: Iff105c45f21c40c2fb0a649fc9fd9a9065e7c952
Reviewed-on: https://gerrit.libreoffice.org/38846
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
.uno:DeletePage deletes the "current" page, while the intent here is to
delete all slides except the first one. Be explicit about this.
Change-Id: Ia8a8a5bf907e3d79cc646d54803447525375ce72
Reviewed-on: https://gerrit.libreoffice.org/37833
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>
Dung out overlapping return enumerations. Move more work into 'move'
callbacks at a safer time, etc.
Change-Id: I62ba5a35f12073b7b9c8de4674be9dae519a8aca
The following scenario causes rendering failure
where blank tiles are returned in CP-5.3.
1. Load doc where the cursor is saved to a top cell.
2. Page down to grow the document rows.
3. Render tile at last row.
4. Load a new view to the same doc (do nothing else).
5. Render same last-row tile, now it's transparent.
Change-Id: I8f7caf61a7d221d1ccf56762b1e609de987976c5
Reviewed-on: https://gerrit.libreoffice.org/37132
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
The following scenario causes rendering failure
where blank tiles are returned.
1. Load doc where the cursor is saved to a top cell.
2. Page down (typically several 100th row).
3. Load a new view to the same doc (do nothing else).
4. In the first view up-arrow to move cursor and invalidate.
5. New tile is rendered incorrectly.
Change-Id: I06c7627d1b74d9e3be3e83d9d9a09cb5479ba660
Reviewed-on: https://gerrit.libreoffice.org/37129
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Passive clients that don't load the document can
be disruptive as they are not useful for saving
the document, so we need to ignore them and
use reliable sessions.
Change-Id: I162ec00823ba5af776fcb55f6d58149f2a56d7bb
Reviewed-on: https://gerrit.libreoffice.org/36712
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
So failing that first assert doesn't give bogus
test duration.
Change-Id: Iaad2e5654e1264bd126193205b5218fd0f6637ef
Reviewed-on: https://gerrit.libreoffice.org/36324
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Tests should have sensible limits so they don't
go overboard and fail needlessly causing noise.
Change-Id: Idd556c348cc0e97e38c710fdbf76fe20c76d8f9b
Reviewed-on: https://gerrit.libreoffice.org/36241
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Apparently some setups produce a slightly different
output than the expected. Here we tolerate for those
outputs as well.
Change-Id: Ia4beeb653ff6182e1403a59fbd05c6a46b9277ac
Reviewed-on: https://gerrit.libreoffice.org/36080
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
This was a workaround to Poco's limitation
of requiring socket receiveFrame be given
preallocated buffer, which couldn't be
exceeded by a larger payload. This meant
the receiver had to know the maximum
payload in advance.
Since only the Kit uses Poco sockets,
and the Kit never receives large payloads,
this preamble is now obsolete.
100% (94/94) of old-style tests PASS.
Change-Id: I76776f89497409e5755e335a3e25553e91cf0876
Reviewed-on: https://gerrit.libreoffice.org/36037
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Issue the fake disk space error
only after fully loading the doc.
This prevents handling the error
while loading, which terminates
the session.
Change-Id: I5acd9454b1aa9fb5d1f886fb23a76a2d808d4852
Reviewed-on: https://gerrit.libreoffice.org/36027
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Without this, online doesn't link, atleast for me. The problem seems to
be because in presence of target specific _LDFLAGS, AM_LDFLAGS is
ignored and it is only the latter where we are specifying ZLIB_LIBS.
Further, since the LDFLAGS values are same for all targets, we can
simplify here and rather use AM_LDFLAGS and remove all target specific
LDFLAGS.
Change-Id: I2f1c30c7f34b32c638a4ac60fab709b32f1e6c9a
Reviewed-on: https://gerrit.libreoffice.org/35959
Reviewed-by: pranavk <pranavk@collabora.co.uk>
Tested-by: pranavk <pranavk@collabora.co.uk>
Since this a fake session that doesn't
have a client socket, we push the
messages directly into the ClientSession.
But since the DocBroker poll thread will
probably not be ready by then, there
is no child process and the other
document bits needed to load (or indeed
process any client messages).
So we defer all the fake messages in
a poll callback to insure they are done
in the correct order.
Change-Id: Id81dc4288b305829149e6e9c81d0f7da719c59ad
Reviewed-on: https://gerrit.libreoffice.org/35712
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
While message emplacement happens in the DocumentBroker poll, we
can be sure that the next iteration of the poll will call
hasQueuedWrites before polling.
These are really GET requests that aren't
WebSocket upgrade. Should rename to something
less misleading.
Re-enabled testSlideShow which depended on this.
Change-Id: I52b7f67b650fcdcbae7c2bff020b756099263141
The payload of paste can't have length 0.
Now we silently skip such cases, although perhaps we
should be more strict and disconnect the offending
client.
Change-Id: Iaa2e7373277f9e7d85209aec56a2f8ee0ef7e801
Reviewed-on: https://gerrit.libreoffice.org/34112
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Now it is possible to use:
./loolwsd_fuzzer --config-file=loolwsd.xml --o:storage.filesystem[@allow]=true --o:logging.level=fatal --fuzz=/tmp/looltrace
Ie. no need to specify the LibreOffice install location. Ideally worth
disabling the logging output too, to gain higher performance.
Change-Id: I4fa5f275cd4f4a52fe2cd07e658cea726f6f31c2
We have enabled LOK_FEATURE_NO_TILED_ANNOTATIONS flag already
in Kit.cpp, which means these tests, which assume comments are
rendered as tiles, are no longer relevant, so lets disable them
for now. Surely sometime in future they would completely be
removed and replaced by new tests that assume comments are *not*
rendered in tiles and rather are added/modified/removed by using
comments API. See LOK_CALLBACK_COMMENT for details.
Change-Id: I48b7a0a0aa116d164a3654618ab0e95252d29cf3
This reduces the chance for random test failures.
Change-Id: I961a502d57a3cb4946b3b7febc6e6b03e7f36aa1
Reviewed-on: https://gerrit.libreoffice.org/33858
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
In unittests we need to count the
number of lool processes. This can
take a long time depending on how many
they are and if they are exiting or
new ones being created.
New logic is to reset the timeout
when the numbers are changing until
it settles on either the expected
(when we return), or some other value.
Change-Id: Ia8fcc7ef60586ebec6d8a280174b4dd2eae0ccc4
Reviewed-on: https://gerrit.libreoffice.org/33672
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Now we don't get a situation where there would be a tremendous amount of
invalidates & tile render requests piled in the queue, so we can do it
deterministic again.
The only thing that could potentially pile in the queue are the keypresses
events sent from the clients, but that is a different problem anyway.
This reverts commit c326228774.
Change-Id: I98e199eab0187bf5f47ce322ac1b1b2e3b976b85
Which is not odd, as it seems to be a forkit process that tries to load
unit-oob.so, and the symbol in question,
UnitWSD::testHandleRequest(UnitWSD::TestRequest,
UnitHTTPServerRequest&, UnitHTTPServerResponse&), is in LOOWSD.cpp.
Oh well. Whatever.
Change-Id: I0e0409b7247acdcd00e830d6fa63e73f27ea6c3a
The server tells the client the hash of each tile it sends (calculated
from the contents of the tile, not its PNG encoding). When the client
asks for a tile to be refreshed, it tells the server what the hash of
the existing tile is. If the server notices that the tile contents
hasn't actually changed, it doesn't PNG encode it and doesn't send it
to the client.
The intent is that this will reduce load on the server and also avoid
unnecessary tile traffic.
Change-Id: Ia06ca68655ea984ed4319f24f4470afda322eccf
If we are logging a message, we want to see the first line of it in
its entirety if possible. Especially now with more parameters being
added to tile messages, 120 was not enough to see the added
interesting ones.
Bin the silly test that used knowledge of what the limit is. We should
not test a coindidental arbitrary number that is not a documented part
of an API. If we want to test the default abbreviation functionality,
we need to at least make that default limit (now 500) public in
Protocol.hpp.
Change-Id: Iea59ba46e8331e2a839c792146f123fed9df2b82
When there are many views requesting tiles,
disconnecting any should not cancel tiles
for other views.
Change-Id: I6029432c4b9ab931811dcb3b03b0441b413a9971
Reviewed-on: https://gerrit.libreoffice.org/32872
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Since the Impress slides use large fonts, we end up
overflowing the part bounds with this test.
Luckily the top row of text has smaller font, and
more vertical space within the slide, so now
we edit the top-left textbox and this test
doesn't fail anymore (tested 50 times) when
it failed ~1 in 4 previously.
Change-Id: I13d8cc4462d48b128578a9d3fdde90a79c607c4b
Reviewed-on: https://gerrit.libreoffice.org/32862
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
This makes debugging much easier as one can
readily match WSD logs with a given test.
Change-Id: I8f2c83d67189038699af3f24dee205bc7efb5c28
Reviewed-on: https://gerrit.libreoffice.org/32860
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Previously tilecombine had its own version, which is
nonesensical, since it's not really a tile.
Now it passes the version to the tiles when
parsing and serializes version per-tile.
Change-Id: I5db8d94880431e3d2a40b6787c6fe51a05771305
Reviewed-on: https://gerrit.libreoffice.org/32633
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
As far as I see, we tend to use the same namespace name as the
basename of the corresponding include file, and this stuff is defined
in a file called Png.hpp.
Change-Id: Id859e13e94568abd9f1d5b4ef3bfbbb0c156db11
canceltiles is not guaranteed to prevent
tiles from getting sent back to the client.
There is an obvious race between receiving
tile requets and cancelling them.
This reduces failures by repeating the
unittest when there is spurious failure.
Change-Id: Icf299e2212e175dc4c0cbc1d2f91c37725f2b261
Reviewed-on: https://gerrit.libreoffice.org/32631
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Dedicated tests for each type of document.
Change-Id: I399e32a64ad0fe25f0cf4795cf428da922eea2f5
Reviewed-on: https://gerrit.libreoffice.org/32564
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Load multiple views on multiple documents
concurrently and verify that the number
of views for each case is consistent and
as expected.
Change-Id: Ie550dd42880da7bdd0aef6d642e9b000be5c3f84
Reviewed-on: https://gerrit.libreoffice.org/32556
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
It always failed. No idea why, or what this stuff is. See also
6bd899f3fa44f2d78092333a399e141889b43dca in cp-5.1 which might be
related, or not.
Change-Id: I95f964e3f99c681ec9eefd092e523babd919a5c0
the change in UnitPrefork.cpp looks like it might fix something
important
Change-Id: I95741c0256b22ac0bb4caadbd113b96584f784b5
Reviewed-on: https://gerrit.libreoffice.org/32343
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
and apply the nullptr plugin.
Lots of hacking in my LO tree required to make this work, will probably
end up needing to add an extra parameter to the LO side.
Change-Id: I02ae1dcdece9d9ddf05f7757f6696e3a5d7d1f14
Reviewed-on: https://gerrit.libreoffice.org/32339
Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>
This was added while debugging some problem with forward slashes
in docurl in websocket endpoint automatically being decoded on
some machines. But it turned out it was due to the apache server
setting. So, this test serves no purpose now and tests something
really obvious.
Change-Id: I4658354d53c481cf7554804383892f501bc408cd
To avoid degrading performance for everyone
because of a single slow/bad connection, we
send data to clients each in its own thread.
Change-Id: I6f980c25a404c4d05bcdb1979849ea3d2776c7b9
Reviewed-on: https://gerrit.libreoffice.org/31984
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
This adds SenderQueue and a wrapper of messages to
send back to clients.
Currently no threading takes place, but the messages
are pumped through the queue nonetheless.
Change-Id: Id9997539c0a2a351cbf406f649c268dd3643e88e
Reviewed-on: https://gerrit.libreoffice.org/31883
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Tiles no longer hog the queue ahead of all else.
We now give priority to callback events, so clients
get to know the document state sooner.
Since tiles take long to render, an equal time
is given to non-tiles (capped at 100ms).
Finally, Impress preview tiles are given
the lowest priority and rendered only when
the queue is drained.
Change-Id: I922c1e11200e5675f50d86b83baee1588cbbf66f
Reviewed-on: https://gerrit.libreoffice.org/31394
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
Messages larger than a certain size are preambled
with a 'nextmessage' message that hold the size
of the subsequent message.
This is a workaround to a limitation the Poco
WebSocket API where if the buffer size is
smaller than the received frame the socket
ends up in a bad state and must be closed.
Unfortunately the new API that avoids this
workaround is not yet released by Poco.
Here we minimize the need for 'nextmessage'
to truely large messages. The limit is now
raised from above 1KB to over 63KB.
We may raise this limit further, but that will
cost each socket that much dedicated buffer size.
Change-Id: I01e4c68cdbe67e413c04a9725152224a87ab8267
Reviewed-on: https://gerrit.libreoffice.org/31286
Reviewed-by: Ashod Nakashian <ashnakash@gmail.com>
Tested-by: Ashod Nakashian <ashnakash@gmail.com>
