about summary refs log tree commit homepage
DateCommit message (Collapse)
2015-01-19update copyright years and links to mailing list archives
The documentation part is managed by the new Documentation/update-copyright script. For the future, the rest may be managed by the update-copyright tool in gnulib
2015-01-19Revert "fadefx: remove module"
This reverts commit 3471463325be6c990b3abd18b4d34f723440d19a. While not strictly necessary, it makes the user syntax in splitfx files much terser. Changes from the original version should make it easier-to-read and the diagram should help a lot with understanding.
2015-01-19consolidate spawn fix for Ruby [Bug #8770]
Ensure we can apply the workaround to dtas-sourceedit and our test cases while also simplifying the existing call sites a little. This will also make for less code churn in 3-5 years down the line when we drop <= 2.1 support.
2015-01-19avoid aliasing global "spawn" method
This makes debugging, grepping, and following code confusing at times and also unexpected breaks usage of the global "spawn" method.
2015-01-19splitfx: export TBEG and TLEN variables
These are the raw sample counts for the "trim" effect and may be useful for arithmetic in the shell.
2015-01-19doc: dtas-env(7) manpage for environment vars
This rounds out the documentation a bit and hopefully introduces/encourages some commonality between the playback and processing/production components of dtas. Clarify some splitfx-related environment variables while we're at it.
2015-01-19sinkedit: flesh out all parameters
dtas-sinkedit now shows default parameters in addition to user-changed parameters, allowing easier editing. Also, we need to be able to revert back to using the default pipe_size on Linux by setting pipe_size to nil (as an empty field in YAML).
2015-01-19dtas-sourceedit: update player as user saves in the editor
This allows changes in the source YAML file to be reflected immediately in player after the user saves the file in their favorite $EDITOR. Previously, a user would need to: 1) start dtas-sourceedit, spawning $EDITOR 2) edit the file 3) save changes 4) exit $EDITOR 5) repeat starting from 1) until happy with the results Now, the workflow allows avoiding the context switch between their $EDITOR and terminal to restart dtas-sourcedit: 1) start dtas-sourceedit, spawning $EDITOR 2) edit the file 3) save changes 4) repeat starting from 1) until happy with the results 5) exit $EDITOR In my experience, this greatly speeds up tuning of the playback change, giving all the repeatability and flexibility of editing text files while having the immediacy of an interactive UI. Keep in mind this can cause problems for those with auto-save enabled in their $EDITOR buffer at inopportune times, so a -N/--no-watch option is added.
2015-01-19dtas-console: avoid crashing on failed cue seeks
Attempting to perform operations on a file which which cannot handle it should not cause dtas-console to die with a backtrace
2015-01-19splitfx: remove unnecessary assignment
We already assigned fmt before entering the case statement, so make the case statement smaller.
2015-01-19partstats: use Etc.nprocessors on Ruby 2.2+
This will likely become more available and faster than GNU nproc(1) over time.
2015-01-06player: enqueued commands cannot use bypass
We cannot specify an audio format for bypass mode, so inherit the last one we used and hope it works.
2015-01-03splitfx: remove redundant Hash#merge call
This was unnecessary since we already iterate through the overrides hash and merge "manually" based on command-line arguments (which allows us to represent nested hashes).
2015-01-03dtas-splitfx: use the top-level command used by -player
If somebody is using dtas-player to help edit a file for dtas-splitfx, it is likely they will want to use it when generating the final files (regardless of "target" format).
2015-01-03splitfx: add INDIR and INBASE environment variables
These are intended to act like `$(@D)' and `$(@F)' in GNU make(1) and to ease managing temporary files for some effects (e.g. noiseprof + noisered in sox) for splitfx users.
2014-12-29source/splitfx: fail gracefully on missing files
We cannot afford to break the entire player because somebody enqueued a non-existent file (or enqueued and later renamed it).
2014-12-29trimfx: require audio format at initialization
This makes things easier for scheduling/expansion since we won't have to deal with floating point numbers when we work directly with with sample counts (like the rest of dtas).
2014-12-29trimfx: comments and cleanups
It's been a while, and I've lost my train of thought regarding this system a bit :<
2014-12-28splitfx: use $FX env consistently in targets, too
While we're at it, document the splitfx manpage and make the example suitable for tests.
2014-12-28examples/splitfx.sample.yml: use more common effect
The "-l" and "-r" switches used by the sox ladspa effect (for latency compensation and replicate, respectively) are not available until sox v14.4.2, so avoid using it since it can fail tests on common systems today.
2014-12-28sigevent/efd: minor simplifications
This reduces memory overhead by a few bytes by eliminating constant lookups and unnecessary inclusions.
2014-12-28source/splitfx: default tracklist for single-track files
While the splitfx source is intended for applying effects to untracked audio files (e.g. transfers of vinyl records or live concert recordings), it should be useful for applying effects to an already-tracked recording.
2014-12-28source/splitfx: standardize on $FX for sox effects
This should allow users to setup effects in a more standardized fashion and avoid needing to specify a "command:" field in their splitfx YAML files in the general case.
2014-12-28source/splitfx: restart source on YAML modifications
Since splitfx YAML files are intended to be frequently edited and modified by the user, we'll support automatically restarting the source when the user saves changes via their favorite $EDITOR This change is only for Linux users. However, sleepy_penguin supports kqueue nowadays so a patch to support such functionality would be appreciated.
2014-12-27player: support "source restart" command
This becomes useful for systems without inotify when we're editing YAML (or whatever) files frequently and want changes to be reflected right away during playback. This is a weaker version of the plain "restart" command, which restarts the entire playback chain.
2014-12-27player: support playing splitfx YAML files
This allows splitfx users to test CUE breakpoints and run file-specific effects without interrupting their normal flow.
2014-12-27dtas-console: keys for "cue next/prev"
The '!' and '@' keys are used in mplayer to skip chapters, so perhaps this is a good analogy.
2014-12-27test: favor Process.spawn when possible
Process.spawn allows vfork() + execve() usage in Ruby 2.2.0 to improve performance over normal fork() + execve().
2014-12-24restart "source ed" after env changes
Merely converting the `before' source information to a hash is not enough, as the `env' sub-hash is mosified in-place and shared with the `after' hash. So use `inspect' to serialize and snapshot the env and rely on a string comparison.
2014-12-22dtas 0.9.0 - minor features and speedups v0.9.0
The major new feature is '>' and '<' keys are now supported in the dtas-console interface for dtas-player. "dtas-tl cat" also received a minor speedup for big tracklists via syscall reductions shortlog of changes since 0.8.0: process: update comment for bug workaround process: update comment for bug workaround #2 compat_onenine: simplify pipe wrapper tracklist: favor &:sym_to_proc style dtas-console: '>' and '<' keys for tracklist next/prev dtas-tl: halve write() syscalls when emitting tracklists reduce syscalls in recvmsg invocations doc: flesh out "tl" subcommand docs The best is yet to come!
2014-12-22doc: flesh out "tl" subcommand docs
The following subcommands are now documented: * current * current-id * next * prev
2014-12-22reduce syscalls in recvmsg invocations
Favor IO.select over IO#wait since the latter makes another ioctl syscall (which we'll make anyways for IO#nread). Having BasicSocket#recvmsg and recvmsg_nonblock detect the buffer size requires extra syscalls, so pass explict maxmesglen, flags, and maxcontrollen args to elide auto-detection since we already have the buffered amount from IO#nread. This cuts more syscalls from the "dtas-tl cat" invocation with larger tracklists.
2014-12-22dtas-tl: halve write() syscalls when emitting tracklists
Notably, this speeds up "dtas-tl cat" output a little bit. Hrm, Ruby really should be using writev here, we'll need to investigate at a later date.
2014-12-21dtas-console: '>' and '<' keys for tracklist next/prev
These key bindings are used in mplayer, too.
2014-12-08tracklist: favor &:sym_to_proc style
This is terser, reads better and generates smaller bytecode.
2014-12-08compat_onenine: simplify pipe wrapper
Array#each already returns the array, so there's no need to use the more confusing Array#map! block
2014-12-08process: update comment for bug workaround #2
Followup to commit 698e1f04580839ea29647f285b39b88fcbb46071 "process: update comment for bug workaround"
2014-12-08process: update comment for bug workaround
We're certain this is a workaround for a bug. We may remove the workaround at some point in the future; but not until the rest of the world has had a chance to upgrade, too.
2014-10-21dtas 0.8.0 - player bugfixes v0.8.0
Several bugfixes for the dtas-player and playback-related components. Audio-editing support is still being worked on. See git repository for full details: git clone git://80x24.org/dtas Terminal-browser-friendly HTML mail archives are also up at: http://80x24.org/dtas-all/ The mailing list for anything related to dtas remains at: dtas-all@nongnu.org The list is open to all without subscription (no HTML email). dtas is for and by users who never want to deal with GUIs. shortlog of changes since 0.7.0: dtas-console: add note to install "curses" gem if missing test/helper: compat w/ coverage in Feature #9508 player: style cleanup to favor &:methods update copyrights and email address for 2014 trimfx: initial cut of scheduling + gap filling test/helper: ancient minitest compatibility test_splitfx: fix tests without opus{enc,dec} remove builtin-$FADEFX support fadefx: remove module tests: hoist out pluck generation xs: favor &: block style for simple cases rely on filesystem encoding client_handler: minor cleanup (favor &:proc) doc: document "tl get"/"dtas-tl cat" escaping gotcha dtas-console: force encoding for current locale unix_server: fix for infinite loop sink: favor &:proc form instead of blocks favor &:proc form instead of blocks in more places unix_accepted: wait for readability on EAGAIN
2014-10-21unix_accepted: wait for readability on EAGAIN
We only drop dead connections, not live ones. This is noticeable on my home machine when using the "powersave" CPU governor.
2014-10-21favor &:proc form instead of blocks in more places
This generates smaller bytecode and avoids unnecessary captures. Perhaps the Ruby optimizer can be taught to handle this automatically.
2014-10-18sink: favor &:proc form instead of blocks
This generates smaller bytecode and avoids unnecessary captures.
2014-10-18unix_server: fix for infinite loop
Occasionally, killing a sink from an external process could result in an infinite loop due to the lack of close notification from __dst_error (in DTAS::Buffer) up to the top-level event loop. Since it is not easy to notify the top-level event loop, we'll detect closed IOs after-the-fact and retry as needed in a cold rescue path. This fixes an occasional error (usually when using a non-parallel make) in the test suite: TestPlayerIntegration#test_sink_killed_during_play: Timeout::Error: execution expired $HOME/dtas/test/player_integration.rb:57:in `sleep' $HOME/dtas/test/player_integration.rb:57:in `block in wait_pid_dead' $HOME/$RUBY/lib/ruby/2.2.0/timeout.rb:91:in `block in timeout' $HOME/$RUBY/lib/ruby/2.2.0/timeout.rb:35:in `block in catch' $HOME/$RUBY/lib/ruby/2.2.0/timeout.rb:35:in `catch' $HOME/$RUBY/lib/ruby/2.2.0/timeout.rb:35:in `catch' $HOME/$RUBY/lib/ruby/2.2.0/timeout.rb:106:in `timeout' $HOME/dtas/test/player_integration.rb:54:in `wait_pid_dead' test/test_player_integration.rb:42:in `test_sink_killed_during_play'
2014-09-22dtas-console: force encoding for current locale
This is the one place we display the filename for users, so it should hopefully make it displayable. Users who care about proper display should use a locale which matches their music collection (or rename their music collection to match).
2014-08-26doc: document "tl get"/"dtas-tl cat" escaping gotcha
We shell-escape filenames, so they may not show up properly when shuttled across the wire.
2014-08-26client_handler: minor cleanup (favor &:proc)
This reduces captures and is more consistent with our other code.
2014-08-16rely on filesystem encoding
This avoids compatibility problems as File.expand_path respects the "filesystem" encoding of Ruby. I'm not 100% sure this is correct, as Ruby encoding support still confuses me, but this fixes a problem where I attempted to add UTF-8 filename with non-US-ASCII characters in it on ruby-trunk.
2014-08-16xs: favor &: block style for simple cases
This avoids captures
2014-07-06tests: hoist out pluck generation
This allows the same sound to be reused in tests.
2014-07-05fadefx: remove module
This will not be needed after all.