In MobileApp builds, the 'else if' was evidentally
getting chained to the unrelated if/else clauses
before the '#if !MOBILEAPP' at line 434 of
DocumentBroker.cpp.
Now this 'else if' is a standalone conditional
in MobileApp builds and it's part of the
'remove idle documents' conditional chail otherwise.
While fixing that, the missing indentation
of a for-loop under a conditaional is also
fixed, to avoid confusing maintainers.
Change-Id: If3de18f385d84cb22d757cf73743fd8541c36df9
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Templates are special because the file that gets
loaded is not in the same format as the final one.
An implicit save-as is issued in Kit for template
loads right after loading to create the final
document format. It is this file (as opposed to the
template) that needs to be uploaded, which we do
right after we get the 'status:' message in
DocBroker to actually create the first version
in storage.
Change-Id: I23cbf527f1059b0b4059f15069737ab35ac860e7
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
To avoid writing (saving) while reading (uploading) or
any other undesirable racse on the document file,
we atomically rename the file after saving on disk
to pass ownership.
After saving in Core, we rename the document in the
jail to .upload (by appending it). DocumentBroker
looks for the file with that extension and atomically
renames it to .uploading (by appening the 'ing' suffix).
This way, the Kit only renames from the original to
.upload and DocBroker renames only .upload to .uploading.
This guarantees that we never rename the same file
concurrently.
Uploading decision is strictly based on the modified
timestamp of the .uploading file, compared to the
timestamp of the last file we uploaded successfully.
Change-Id: I03520cd8c87605f6dad417e7a978204f76fc0c38
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
This makes sure that if the document isn't modified, we
will detect no change from the storage as well and
avoid making an unnecessary upload.
Change-Id: I8ba2850921d75c96b2aa4f6deb86891b4fad0ea0
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
The client always issued a save before renameFile.
This worked fine until we make uploading asynchronous.
Now that uploading is done asychronously, the renameFile
command arrives while uploading is still in progress
and fails because renaming is done via the same
storage API used for uploading.
It is best to do both the saving (including uploading)
and renaming directly in WSD, instead of relying on
the client to save explicitly before renaming.
Change-Id: I99f6294f4c83130bc13ec8e3b013ba0192f0f84c
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Some user-initiated activities, such as renaming
the document filename, is either a multi-stage
activity or involves async operations.
Using a new Activity enum in DocumentState, we
now track each of the different activities that
needs special attention or handling.
Change-Id: I9cd2d04b10a97387ce02f5c3cc98b41a1732af6a
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
With async uploading we have to process the
stop-condition only after we get the response from
the async operation. For now, we don't stop
if we fail to upload, giving a chance to retrying.
Timeout should be used in the poll loop to
limit how long we retry.
Change-Id: I8e8f8618f92f34ef2f1853e499f4f8d6ce932b40
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
This is to also include the 'force' case that is
needed when unloading a document.
Now that the force flag is reliably set, we don't
need to override it in needToUploadToStorage.
Change-Id: Ic0672e6c7bcb53a17ec6a0cbd2d2f8b434fbc300
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Now we can handle the response of the storage
after the async upload is complete (or timed out).
Change-Id: I29d450646bddb07f02bb17d257e7e0fa372ce357
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Add the necessary support for handling async
document upload to storage in DocumentBroker.
Change-Id: Id7a58c245d729cef58669bafb60db450f7ebbba2
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
loleaflet does not look for a &enabletraceeventlogging=yes query
parameter. That was removed many weeks ago.
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I33b36191f1d21c2c0177061ff704aa9bcd5f9ef3
The path to the local file is relative to the jail.
Change-Id: Ie008aee4afc264339706b5b4715af35cd71ac55d
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Modelled on how it is done in core. ProfileZone is derived from
NamedEvent which is derived from TraceEvent. Here we don't keep any
separate ProfileZone.hpp, though.
This was needed to introduce generation of "instant" events here, too.
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I6583134e96001641c50339deb4197fca6ab7d5d5
(Also in variable names.)
Better to just use terms that say what we mean, i.e. most verbose and
least verbose.
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I80d74fda8b80bd34d194c3df97d246a41368189b
This was already worked on iOS since commit
75a3ab02ca (Add an iOS setting for the
user's name or nickname, 2021-04-07), start using the same API on
Android as well.
Signed-off-by: Miklos Vajna <vmiklos@collabora.com>
Change-Id: I37825d143b0ed92bd84f9a1512313e51ff5e761c
Add a way to set it from the client.
For now, in the Help > About dialog, the L key toggges the logging
level between the default (whatever is set in loolwsd.xml or on the
loolwsd command line) and the maximum a client is allowed to set
(which also can be set in loolwsd.xml or on the loolwsd command line).
Also while at it, uncouple the toggling of the Trace Event generation
from the triple-click tha toggles tile debugging. It is now on the T
key.
Also swap the meaning of "min" and "max" for the logging levels. Even
if the Poco::Message::Priority values are such that the less urgent
priorities have a numerbcally higher value, conceptually it is more
clear to say the more urgent priorities (levels) are "higher".
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I5383b16ead7d2d378901525eb0d16513d88f045e
We don't want a client to be able to turn logging off completely
either.
Note, just to avoid potential confusion: This is about the LOG_INF(),
LOG_WRN(), LOG_TRC() etc logging here in Online. Not related to my
recent work on Trace Event format logging.
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I3de527cb9f69d9a5269b1db2e1d4602e7117f22e
(Not used yet.)
The intent is that the JS code will be able send the server a message
that adjusts the logging level for the WSD thread(s) and Kit processes
handling that client.
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I88ac301f002b1f3a55a5d8c9e3650babdedc1e0d
I had accidentally used core naming conventions for ProfileZone.
Change-Id: If10eceb7a5a29d26662d5b21a9289681d814a12e
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Don't bother collecting them into a vector, like in core. Instead,
just call a static member function ProfileZone::addOneRecording() for
each Event Trace object (string) to be emitted. That function then
needs to be implemented separately in each executable. In WSD it logs
the object to the Event Trace file. In Kit it sends it to WSD for
logging. In Unit tests we use a dummy implementation.
(If Event Trace logging is not enabled at all in loolwsd.xml (the
default), nothing is done, of course.)
When receiving the "traceeventrecording start" or "traceeventrecording
stop" message from the client, turn ProfileZone recoring on or off in
the WSD and Kit process. (Probably in WSD the flag should be
per-client.)
Change-Id: Ie1127d65dd44ed77e7eeab4b0f0a90cce95dc4a2
Signed-off-by: Tor Lillqvist <tml@collabora.com>
That is our convention, and this also avoids a global variable in the
kit process.
Change-Id: I37d2d53aa7eb24f9848fa8ef98bc57d75db90d13
Signed-off-by: Tor Lillqvist <tml@collabora.com>
For now, an argument object should have all keys and values as strings
not containing any spaces because of the trivial way in which it is
parsed.
This is a bit sad, yes, but necessary until we move to some more
structured syntax for the messages.
Also fix an incorrect comment.
Change-Id: I8a1408a4a1787b66a3cf7b26b3d92c07df244c47
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Instead of requiring that information in the query string of the loleaflet.html URL.
Change-Id: I8c41e87c7f561561adeb03ec34ce0c19fe9d7fa5
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Make it contain a proper empty JSON array if no trace objects have
been written to it.
Change-Id: Ided1992a822659561a909ac545536261b91adbac
Signed-off-by: Tor Lillqvist <tml@collabora.com>
It enables turning Trace Event recording on (and off again). The
option is passed down to the client through loleaflet.html, and to the
KIT processes. If the option is not used, the new JS functions that
send trace events to the server turn into no-ops to avoid wasting
bandwidth.
It is always on in a "make run".
Change-Id: Iafe1919ccba7c376137d3e0568b857e20780bbc8
Signed-off-by: Tor Lillqvist <tml@collabora.com>
wsd/ClientSession.cpp:411:31: error: object backing the pointer will be destroyed at the end of the full-expression [-Werror,-Wdangling-gsl]
const char* str = tokens[2].data();
^~~~~~~~~
Signed-off-by: Miklos Vajna <vmiklos@collabora.com>
Change-Id: I67e4aa5f61e049ad13663dc1e11ae2bd67399bea
Add a function emitInstantTraceEvent() to leaflet that sends an
"Instant" Trace Event to the server for logging.
Add a function createAsyncTraceEvent() that creates an object that
records the timestamp of its creation and sends the 'b' event to the
server for logging, and when you call finish() on the object, sends
the corresponding 'e' event.
Finally, add a function createCompleteTraceEvent() that creates an
object for a "Complete" Trace Event that includes the end timestamp
*and* the duration. The event is sent to the server when you call
finish() on the object.
Loleaflet sends timestamps in the above messages from
performance.now(). To enable the server to turn those into absolute
timestamps, the loolclient message is amended to include the current
Date.now() and performance.now() values.
Note that the intent is that when generating Trace Event logs, the
server and the web browser run on the same machine, so there is no
wall-clock synchronisation issues between server and client.
Change-Id: Ie9e68b093b769cc942e1e1d17083febeb07ccf5e
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Also avoid generating abbreviations in various message handling loops
unless debugging is enabled.
Change-Id: I22f4929b0bfd4da36917db6882bb2f5f5be02780
Signed-off-by: Michael Meeks <michael.meeks@collabora.com>
Avoiding default values for parameters makes the code easier to read.
Especially true for bool parameters. But sure, just a question of
taste.
Change-Id: I473f70bdfafe3a9ccfb325def8760d78fee7e9a6
Signed-off-by: Tor Lillqvist <tml@collabora.com>
When toggling tile debugging in the Help>About, toggle Trace Event
recordning, too. Add a comment with an example of how to turn on
SAL_LOG overriding at the same place.
The following new messages from client to server are introduced:
traceeventrecording <start/stop>
Starts or stops comphelper::TraceEvent recording.
sallogoverride <string>
Overrides the SAL_LOG value, or stops overriding if no parameter
or parameter is "default".
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I8b56c28cd99d39115cd796c44e5051d934d21a1f
Signed-off-by: Tor Lillqvist <tml@collabora.com>
It is (for now) produced (by the ProfileZone things in core) whenever
logging level is "trace". The Event Trace file pathname can be given
in the loolesd.xml file or on the loolwsd command line.
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Change-Id: I6d5829121e71460a4600ee94d2ebf51043c8893f
This enables writing multiple WS frames
at a time, up to the capacity (rounded
to the next full-frame's size).
Change-Id: I30d25f9ba1e405156d1e44886b2f3ac8017f0f3c
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
When writing to the socket, it's always more efficient
to fill the buffer up to the hardware limit for each
write. This is doubly important for efficiency with
SSL, due to the overhead of encrypting multiple
small buffers instead of one large one.
Currently we don't write more than one message
at a time, primarily due to limitations in
the Poco sockets in the unit-tests, which
have a hard time consuming multiple WS frames
with a single poll (subsequent calls to poll
doesn't enter signalled state until new data
arrives, possibly because the data is read and
buffered internally, making the whole scheme
of using poll unreliable and meaningless).
Change-Id: Ic2e2cf1babfb5ab4116efd93f392977ba234d92b
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
... without copying the token.
And use it in TileDesc::parse(), which is known to be a hot path.
Signed-off-by: Miklos Vajna <vmiklos@collabora.com>
Change-Id: I0dcf2eb26c93254cdc6a1c11f9708daf213a825d
And use it in TileDesc::parse(), which is known to be a hot path.
Signed-off-by: Miklos Vajna <vmiklos@collabora.com>
Change-Id: I20375d7a1c31f61662446979e4d6799fd45b49d3
In some rare cases Poco throws "Exists" exception.
It's not clear what causes it, so this should
log more of the original exception, hoping that
it would shed some light.
Change-Id: I51723f35c9e062c56608e8f7b80c9a65d3f0488f
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Even if we use http:// locally via a tunnel the new outgoing socket
code needs SSL initialized to connect securely to remote servers.
Change-Id: If64a3838267182757591a8026097bf08d9ba732f
Signed-off-by: Michael Meeks <michael.meeks@collabora.com>
Used for comments added to documents by the user.
The setting, like templateListURL, can be set through a mobile device
management system.
There is no trivial way on iOS for an app to programmatically find out
the name of the user, and that is good from a privacy point of view.
Fixes https://github.com/CollaboraOnline/online/issues/1843
Change-Id: Ie68fcbacf886ec8f1c74021a71879b38d4180c15
Signed-off-by: Tor Lillqvist <tml@collabora.com>
Per the RFC, the server must set the 'Connection'
header to 'close' to signify that the current
request is the last.
With this patch, we now properly notify all the
clients when we do not intend on resuming communication.
Although, technically, there is no reason why we
shouldn't allow the client to reuse the connection,
we may consider that in the future.
See https://tools.ietf.org/html/rfc7230#section-6.3
Change-Id: Ic6250c0afe8124314a9122452040f6a75728a525
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Since the return value of syncRequest and syncDownload
was derived from Response anyway, there is no reason
to have multiple values for callers to look at.
This simplifies the API.
Change-Id: I0f136e515dd0ef6eda84f6a7cd662b260809d2f1
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
Rendering PDF documents to images can consume
large amount of memory. Here we limit the
maximum allowed resolution to 384. This is
about 400% zoom resolution, given that the
100% resolution is 96 dpi.
If this proves limiting, it can be increased
to 480 dpi, which is about 500% zoom scale.
Change-Id: I719118d517bbaf8b204411223f1e26a31b97ca99
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
The defined envar is used by Core to override its
default resolution (of 96 dpi), when specified.
Change-Id: Ie7164e78f98990ac88b02dbfe2becbae8c134e9c
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
User-Agent is designed for client-side use only,
in http requests. For servers, the Server header
is designed to announce the server name and version.
This tries to normalize the use and documents
the proper intent and usage.
Change-Id: I42d68d65611cab64c45adf03fe74f9466798b093
Signed-off-by: Ashod Nakashian <ashod.nakashian@collabora.co.uk>
