git-svn-id: https://svn.musicpd.org/mpd/trunk@4585 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4584 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4583 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4582 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4575 09075e82-0dd4-0310-85a5-a0d7c8717e4f

renamce cstrtok to buffer2array. please don't rename functions; especially to names that look extremely std-lib-ish.  also, don't use isspace, apparently it's local dependent and potentially consideres ' ' or '\t' not to be a space, or considers other characters to be a space.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4574 09075e82-0dd4-0310-85a5-a0d7c8717e4f

-use tree for tagTracker
-eliminate the master process

git-svn-id: https://svn.musicpd.org/mpd/trunk@4571 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4570 09075e82-0dd4-0310-85a5-a0d7c8717e4f

Also added a unit test to check for errors/bugs to make sure we
don't have regressions.

Bug found by Qball.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4569 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4542 09075e82-0dd4-0310-85a5-a0d7c8717e4f

Our id3 tag buffer should be an array of bytes, not an array of pointers to bytes.  Now I know where those warnings came from...

git-svn-id: https://svn.musicpd.org/mpd/trunk@4540 09075e82-0dd4-0310-85a5-a0d7c8717e4f

Any escaped instances of \ must already be inside an already
quoted string, though.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4539 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4538 09075e82-0dd4-0310-85a5-a0d7c8717e4f

(and fdprintf was never meant to be reentrant, either)

A huge thanks to welshbyte for reporting the bug and being very
helpful in helping me fix it.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4537 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4536 09075e82-0dd4-0310-85a5-a0d7c8717e4f

merged back into trunk

git-svn-id: https://svn.musicpd.org/mpd/trunk@4535 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4533 09075e82-0dd4-0310-85a5-a0d7c8717e4f

tree optimization: reduce the number of compares required for insertion and deletion by storing the position in the parent node of each child

git-svn-id: https://svn.musicpd.org/mpd/trunk@4532 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4518 09075e82-0dd4-0310-85a5-a0d7c8717e4f

This was originally introduced in r3718, but reverted r3859 since the
original r3718 commit was incorrect (and I was too excited about
the speedup and also lacking in UTF-8 files to notice :x)

git-svn-id: https://svn.musicpd.org/mpd/trunk@4517 09075e82-0dd4-0310-85a5-a0d7c8717e4f

Oops, I broke pause/resuming from a statefile r4514
Everything should be fixed out.

Also we now avoid opening the audio device until we have a
playable audio_format set.  This is a long-standing bug that got
exposed more blatantly with the single array.

Thanks to MattD in #mpd for reporting my breakage.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4516 09075e82-0dd4-0310-85a5-a0d7c8717e4f

Just malloc all of the audioOutput array in one shot
to avoid fragmentation and to improve cache locality
when iterating through the array.

We also know name and type members of the AudioOutput
struct won't change in the config, so there's no
need to strdup them.

newAudioOutput => initAudioOutput

git-svn-id: https://svn.musicpd.org/mpd/trunk@4515 09075e82-0dd4-0310-85a5-a0d7c8717e4f

It just made things more confusing.  We'll just store
the states in playerData_pd->audioDevicesStates and be
done with it (it's a unsigned byte now).

git-svn-id: https://svn.musicpd.org/mpd/trunk@4514 09075e82-0dd4-0310-85a5-a0d7c8717e4f

Some people have more than 8 devices (the old limit).  It's
pretty easy to support as many as our hardware and OS allows
so we might as well.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4513 09075e82-0dd4-0310-85a5-a0d7c8717e4f

* Moved all logging-related stuff into log.c
(and not myfprintf.c)

* ISO C90-compliant strftime usage:
%e and %R replaced with %d and %H:%M respectively

* Got rid of variadic macros since some old-school compilers
don't like them

* compiling with -DNDEBUG disables the DEBUG() macro

git-svn-id: https://svn.musicpd.org/mpd/trunk@4512 09075e82-0dd4-0310-85a5-a0d7c8717e4f

No point in doing all that extra work for one variable...

git-svn-id: https://svn.musicpd.org/mpd/trunk@4511 09075e82-0dd4-0310-85a5-a0d7c8717e4f

playerData.c:
proper error checking

directory.c:
properly check myFgets() for errors
(it returns NULL on error)

inputPlugins/mp3_plugin.c
get rid of commas at the end of enums

interface.c:
we weren't using long long, so strtoll isn't needed
get rid of void-pointer arithmetic

sllist.c:
get rid of void-pointer arithmetic

compress.c:
get rid of C++ comments, some compilers don't accept them

Note that I personally like void pointer arithmetic, but some
ancient compilers don't support them :(

git-svn-id: https://svn.musicpd.org/mpd/trunk@4510 09075e82-0dd4-0310-85a5-a0d7c8717e4f

bcopy is old-school, yo

git-svn-id: https://svn.musicpd.org/mpd/trunk@4509 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4508 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4507 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4506 09075e82-0dd4-0310-85a5-a0d7c8717e4f

next is to clean up the code and make obvious optimizations

git-svn-id: https://svn.musicpd.org/mpd/trunk@4502 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4498 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4497 09075e82-0dd4-0310-85a5-a0d7c8717e4f

*) add copyright notice
*) implemented node deletion, but not yet tested

git-svn-id: https://svn.musicpd.org/mpd/trunk@4496 09075e82-0dd4-0310-85a5-a0d7c8717e4f

git-svn-id: https://svn.musicpd.org/mpd/trunk@4495 09075e82-0dd4-0310-85a5-a0d7c8717e4f

This way it's easier to manage and extend.

git-svn-id: https://svn.musicpd.org/mpd/trunk@4494 09075e82-0dd4-0310-85a5-a0d7c8717e4f

This shaves another 5-6k because we've removed the paranoid
fflush() calls after every fprintf.  Now we only fflush()
when we need to

git-svn-id: https://svn.musicpd.org/mpd/trunk@4493 09075e82-0dd4-0310-85a5-a0d7c8717e4f

*) when CHILDREN_PER_NODE is large, use binary search
*) add a iterator implementation
*) some code cleanup

git-svn-id: https://svn.musicpd.org/mpd/trunk@4492 09075e82-0dd4-0310-85a5-a0d7c8717e4f

strncpy isn't really safe because it doesn't guarantee null termination,
and we have had to work around it in several places.
strlcpy (from OpenBSD) isn't great, either because it often leaves
errors going unchecked (by truncating strings).

So we'll add the pathcpy_trunc() function with is basically strlcpy
with a hardcoded MAXPATHLEN as the limit, and we'll acknowledge
truncation since we only work on paths and MAXPATHLEN should be
set correctly by the system headers[1].

file-specific notes:

inputStream_http:
eyeballing the changes here, it seems to look alright but I
haven't actually tested it myself.

ls:
don't even bother printing a file if the filename is too long
(and when is it ever?) since we won't be able to read it anyways.

metadataChunk:
it's only metadata, and it's only for showin the user, so truncating
it here souldn't be a big issue.
memset to zero in init is unecessary, so lets not waste cycles

[1] - If the system headers are screwed up, then we're majorly
screwed regardless of what we do :x

git-svn-id: https://svn.musicpd.org/mpd/trunk@4491 09075e82-0dd4-0310-85a5-a0d7c8717e4f