I have made a pseudo-release tarball (for internal testing purposed) with the
version 1.0.0.
Use the following policy for versioning: When doing a release tarball, bump
the version in configure.ac so that the third (micro) version number is
even. Don't commit that to git, but do run make dist. After you have the
tarball, bump the micro version number again to an odd number. Commit. This
way, anything from git will always have an odd micro version number.
Useful when building loolwsd on a system that doesn't have new enough
LibreOfficeKit headers available. (It makes sense to use them only if the
resulting loolwsd will then be run on a system with a new enough LibreOffice
installation anyway.)
Pass the --with-lokit-path=bundled/include configure option to use them.
It doesn't work to debug a program that has file capabilities set, it seems,
so to debug the loolwsd master process, one in practice needs to run sudo gdb
on it. But it is not necessarily a good idea to run all of the code as
root. When configured for debugging (--enable-debug), reset real and effective
uid to a non-root one, either one given with an --uid option (typically that
of the developer), or "nobody".
Catch IOException when trying to copy the document and send error message.
Otherwise, the exception will just propagate up to handleRequest() and the
connection will be silently closed. Closing the connection is fine as such, I
think, but we need to send an error message first.
Currently we require the current version, 1.6.0, and as soon as 1.6.1 is out,
I will start requiring that, because I want to use a feature I submitted to
it.
Surround most of configure.ac with AC_LANG_PUSH([C++]) while at it, as this is
all C++ code anyway.
Otherwise, if we use the same port number and same HTTPServer, if enough
clients try to contact us and, we won't be able to accept child processes
having been spawned.
Also add some temporary debugging output here and there to debug lifecycle
management issues.
Just a stopgap measure so far. Clearly will need some logic to get rid of too
many children if they are just hanging around with no clients incoming. Also,
we need some sanity check not to spawn an unlimited number of children.
We don't actually need this plugin because the server
dictates how the selection will work.
Will remove the whole plugin when I'm sure this won't have other
applications
Otherwise two or more linkOrCopy() functions running simultaneously will lead
to a mess.
Actually nftw() is not guaranteed to be thread-safe, I think, so we should
really use something that is, like Poco's SimpleRecursiveDirectoryIterator.
It probably is not a good idea to keep depleting the entropy source needlessly
by having a separate RNG in the preSpawn(). Use just one shared RNG and
protect access to it with a mutex.
Mihai Varga
Tor Lillqvist