not a stability thing but helps with tracking leaks

This doesn't show much at the moment but the log_close was purging the cached log entries even i...

In cases where a copy is made from the queue, ie when on resync of lagging clients or a move to ...

HTML5 apps like chrome issue range requests which can clash with time limited set connections. A...

If there are a high number of admin requests for mount streams then the lock required can cause
...

Originally the only meta block was icy, so using a refbuf to store the icy data was fine but now...

This is primarily for trapping large ID3 but could be important for other frames. The actual
ex...

Fix status-json leaking IPs

Fix CVE-2018-18820

fixup auth url buffer overrun possibility

Fix makefile

Add missing makefile

Move to Relative Paths to support being behind a reverse proxy.

Tag with correct versions

Fix issue where we require x-forward-for set for new all-x-forwarded-…

Use boolean instead of int in XML config file

Allow us to enable x-forwarded-for on all IPs

handle very large frames (typicall ID3 tags) without dropping to failed parsing.
drop the restri...

Needs to be renamed really, but this should only be used for content that has been
prepared and ...

seems like browsers have issues with ID3 embedded in content when not at the beginning. This
pos...

The increase generally meant a slightly inaccurate rate used for fallback files

A recent commit to lessen the frequency of purges tripped over the check for the end time
check....

A number of internal checks are based on the current assessment of the incoming bitrate, which m...

Wrong api call with log handle, looks like this was causing a problem for socket closures
on sou...

While glibc is ok with this, we cannot assume certain behaviour across the board, eg strtol can ...

The id is attached to the source_t not the mount in these cases

make source push a 0 value on each run through to allow for a regular drop of the bitrate if
not...

we don't want to send a response on a dead socket.

if a listener is kicked back to intro with a partial block then reset timer so the drop case
is ...

This should on be on pre IPv6 setups and only with relays involved.

usually the trusted CA pem is located by the openssl library in a system wide place but this
all...

mostly to prevent source clients bouncing around too much. allow listeners to move around a
bit ...

when falling back to file, the average bitrate taken from the source is taken to be the average
...

Allow for reporting the intro file being played and how much of it. Uses the <preroll-log> block...

Keep alive is really for multiple short lived connections, source client onnections stay
connect...

This seems to affect windows more than linux due to lock implementation but needed fixing. Would...

this now uses the sync routine as a separate function and delays the opening/checking of the
fil...

A lot of the changes here are actually just down to structures being shuffled around.

Callback ...

the return type may need expandin on to include samplerate/chan/bitrate but for now
just deal wi...

original patch did not handle file processing case. This still needs a little work for
the heade...

make log messages switchable on a flag and tighten control on recognised but
ignored formats

rejects listeners via the 403/redirect mechanism

to save a little re-calculation and checks that are usually pointless, but occur on
each listene...

The adds an extra item in the client. I'm treating this as a generic temp value holder
which cou...

The mount list is now in two blobs, a template list (wildcards) and a specific
tree. This is to ...

with keepalive, the ssl part of the listener was not cleaned up so was wasting resources.
When a...

a leftover from when the source was mutex locked, now it is only read locked from a
listener con...

low bitrate default (eg mp3/aac) streams can adjust there delays quite large, but we
don't want ...

when falling back to file withe private buffers (eg auth intro data, or private queue
buffer cop...

When unable to open log files and fallback to stderr occurs, the triggers were causing
a recursi...

The purge routine needs to be sensitive to the async write point, not just the number
of entries...