[Ffmpeg-devel-irc] ffmpeg-devel.log.20150719

Mon Jul 20 02:05:02 CEST 2015

[00:03:51 CEST] <durandal_1707> should post patch for amerge that uses .needs_fifo and watch Nicolas reaction
[00:09:22 CEST] <wm4> durandal_1707: relicensed the vapoursynth filter
[00:09:28 CEST] <wm4> but be aware that it's a huge hack
[00:09:42 CEST] <wm4> it's no fun to connect 2 APIs that are essentially the opposite of each others
[00:09:56 CEST] <wm4> it's also inefficient and might not work fully correctly with all scripts
[00:10:27 CEST] <wm4> also you should probably ignore the hacked Lua backend (the proper solution is extending vapoursynth script by a proper Lua backend or so)
[00:10:38 CEST] <wm4> all in all the data flow is pretty complicated
[00:10:51 CEST] <wm4> maybe you even find a better way to do this
[00:11:55 CEST] <durandal_1707> I noticed that some plugins can't pick right pixel format
[00:12:21 CEST] <J_Darnley> Did someone say Lua?
[00:14:21 CEST] <wm4> durandal_1707: what do you mean yb that?
[00:14:52 CEST] <wm4> J_Darnley: vapoursynth uses python as main scripting language, but scripting and core are separate and it was pretty trivial to add a primitive Lua scripting backend
[00:17:06 CEST] <J_Darnley> Nice.  I might be able to look past all the C++ then.
[00:17:57 CEST] <wm4> the API is in C
[00:39:33 CEST] <durandal_1707> wm4: what I mean: need to add lavfi=format... Before and after vapoursynth
[01:00:10 CEST] <cone-487> ffmpeg 03James Almer 07master:78347549a444: avutil/intmath: check for ICC before GCC
[01:48:46 CEST] <cone-487> ffmpeg 03James Almer 07master:bd487645327b: avutil/x86/bswap: force inline asm versions with ICC
[03:24:51 CEST] <cone-487> ffmpeg 03Michael Niedermayer 07master:54882156dd09: avcodec/hevc_parser: Treat extradata parsing differently, as it should not contain an AU
[11:29:53 CEST] <cone-506> ffmpeg 03Alexandra Hájková 07master:406627287e01: asfdec: do not read replicated data when their length is 0
[11:29:53 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:f509c9503a8f: Merge commit '406627287e015ce381795e85e2557b12bf60ca35'
[11:38:35 CEST] <cone-506> ffmpeg 03Alexandra Hájková 07master:0989d3ad1fbd: asfdec: convert condition for the replicated data reading to be safer
[11:38:36 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:80a37fc3e10a: Merge commit '0989d3ad1fbd7509815208b0a5792918492d2a68'
[11:45:28 CEST] <cone-506> ffmpeg 03Alexandra Hájková 07master:2883ef34b59c: asfdec: read the replicated data in a separate function
[11:45:29 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:74aba00700d8: Merge commit '2883ef34b59c9b427c4cfad4620c3235e5778406'
[11:52:57 CEST] <cone-506> ffmpeg 03Alexandra Hájková 07master:e61f39849c2e: asfdec: make nb_sub to be unsigned int
[11:52:58 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:3d4297f851ec: Merge commit 'e61f39849c2e2b7f492c17b42058242ed2fa4d57'
[12:07:01 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:56c2f3772701: avconv: drop update_sample_fmt()
[12:07:02 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:8534c8c77a50: Merge commit '56c2f37727015212a404cae0f444d8bc8704d691'
[12:51:19 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:59245e0c5e10: avconv: set the encoding/decoding_needed flags earlier
[12:51:20 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:843be56ee189: ffmpeg_opt: Fix stream copy flag for attachments
[12:51:21 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:db3e12a82855: Merge commit '59245e0c5e10a849e67c632cccf4f677b2442e82'
[13:23:36 CEST] <ubitux> kierank: if you're interested in the parseutils, you might want to raise attention on 71956371a7bb911dee4b206cd991bbdca2e0fb62 and 49ba6e56bd321da0e066de6533592131894b8902
[13:23:36 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:6d5d9246042a: avconv: move handling the 2pass logfile into avconv_opt
[13:23:37 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:247f4d1f181c: Merge commit '6d5d9246042acb804a652e6fedfb7afe0ca85614'
[13:45:57 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:bd2ab27c488a: avconv: use read_file() for reading the 2pass stats
[13:45:58 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:edd110ee4526: Merge commit 'bd2ab27c488ae92c7820efe11d4f53d84e94d58e'
[13:53:40 CEST] <durandal_170> anybody tried sidechain compressor?
[14:01:02 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:ab7b038906f3: avconv: factor out the output stream initialization
[14:01:03 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:15fbf3e72a64: Merge commit 'ab7b038906f3e40ed474676d8e3029902a2078f5'
[14:01:28 CEST] <D404|Ghetto> i see michaelni re-did my patch for me for PAT/PMT
[14:01:35 CEST] <D404|Ghetto> i guess laziness really does pay aoff
[14:01:46 CEST] Action: D404|Ghetto waits as a laser sight moves towards his head
[14:11:03 CEST] <durandal_170> when gonna push vf_reverse?
[14:11:30 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:1959351aecf0: avconv: move the no streams failure to open_output_file()
[14:11:31 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:3b0534efdcde: Merge commit '1959351aecf09fc3e90208ff775f4849801dc13f'
[14:18:59 CEST] <D404|Ghetto> durandal_170: i was waiting for ubitux to reply
[14:19:02 CEST] <D404|Ghetto> he never did.
[14:19:26 CEST] <ubitux> i just did, sorry :)
[14:20:11 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:be101bc1e357: avconv: create the complex filtergraphs earlier
[14:20:12 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:c7dd2365f432: Merge commit 'be101bc1e357c50fcb740bc4870b3bacc93a5727'
[14:32:01 CEST] <durandal_170> now should be possible to do ducking with lavfi
[14:33:59 CEST] <ubitux> do we have a "all colors" image in fate that could be used for testing filters that do a 1:1 pixel color processing 
[14:34:01 CEST] <ubitux> ?
[15:18:20 CEST] <D404|Ghetto> slight problem: i forget the ffmpeg push ssh url
[15:18:28 CEST] <D404|Ghetto> since i am on no-net atm
[15:18:33 CEST] <D404|Ghetto> (my usual dev box is down)
[15:24:30 CEST] Action: D404|Ghetto pokes durandal_170 or ubitux 
[15:25:16 CEST] <ubitux> git at source.ffmpeg.org:ffmpeg
[15:25:31 CEST] <D404|Ghetto> ah
[15:25:35 CEST] <D404|Ghetto> i had .git at the end of the url.
[15:26:01 CEST] <D404|Ghetto> gmm doesnt work.
[15:26:13 CEST] <ubitux> you sure you have ':' and not '/'?
[15:26:48 CEST] <durandal_170> ubitux: nope, create one
[15:27:01 CEST] <ubitux> yeah, will probably do
[15:27:16 CEST] <D404|Ghetto> i seem to have edited .git/config wrong
[15:27:39 CEST] <Compn> 	url = git://source.ffmpeg.org/ffmpeg.git
[15:27:40 CEST] <ubitux> i finally figured out accurately the formula for the absolute settings in selectivecolor, but i can't find the proper logic for the relative one :(
[15:27:58 CEST] <Compn> D404|Ghetto : thats whats in my .git/config
[15:28:13 CEST] <D404|Ghetto> Compn: it's pushurl
[15:28:16 CEST] <D404|Ghetto> i had it as 'push'
[15:28:35 CEST] <cone-506> ffmpeg 03Derek Buitenhuis 07master:8271a84094aa: avfilter: Add reverse filter
[15:28:43 CEST] <Compn> ah
[15:29:15 CEST] <D404|Ghetto> the magic cli is git remote set-url --push origin <url>
[15:29:21 CEST] <D404|Ghetto> dem subcommands.
[15:29:35 CEST] <Compn> is that in our git howto ?
[15:29:40 CEST] <Compn> if not should put it there
[15:29:42 CEST] <Compn> :)
[15:32:03 CEST] <D404|Ghetto> anyway i have a filter in lavfi
[15:32:06 CEST] <D404|Ghetto> hell has frozen over
[15:32:13 CEST] <D404|Ghetto> ubitux, the great demon lord, shall rise
[15:32:24 CEST] <durandal_170> D404|Ghetto: now write areverse
[15:32:35 CEST] <ubitux> and seeking support
[15:32:40 CEST] <D404|Ghetto> ive never even heard of reversing audio
[15:32:46 CEST] <D404|Ghetto> oh well yesi have
[15:32:50 CEST] <D404|Ghetto> but only for dun
[15:32:52 CEST] <D404|Ghetto> fun*
[15:32:58 CEST] <D404|Ghetto> are there even real applications
[15:33:36 CEST] <durandal_170> sox have it, uses temp file
[15:34:02 CEST] <D404|Ghetto> vf_sox?
[15:34:24 CEST] <durandal_170> bah to complicated
[15:34:39 CEST] <D404|Ghetto> surely more useful than fri0r
[15:34:41 CEST] <durandal_170> sox program
[15:34:41 CEST] <D404|Ghetto> or w/e it is
[15:35:22 CEST] <durandal_170> You never heard of SoX?
[15:36:11 CEST] <D404|Ghetto> of course i have
[15:36:14 CEST] <D404|Ghetto> there is also libsox
[15:36:18 CEST] <D404|Ghetto> which is what i was refering to
[15:36:31 CEST] <D404|Ghetto> e.g. we wrap other things like frie0r or w/e similariyl in lavfi
[15:38:09 CEST] <ubitux> D404|Ghetto: "This iflter qequires memory"
[15:38:26 CEST] Action: D404|Ghetto runs
[15:38:40 CEST] <ubitux> well don't forget to fix it after a lap or two
[15:40:13 CEST] <wm4> some alsa player thing can play audio backwards by setting negative speed
[15:40:27 CEST] <wm4> but I don't know if that can even play compressed audio
[15:44:54 CEST] <wm4> hm it can do at least mp3
[15:44:56 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:6d592fbd0d8e: avconv: split creating and (re-)configuring complex filtergraphs
[15:44:57 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:bc3f19641ce3: Merge commit '6d592fbd0d8e89ecade3fc93b36ea200213dc01c'
[15:44:58 CEST] <wm4> I wonder how it works
[15:44:58 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:2dbaec690d9f: ffmpeg_filter: Check that filters have connected outputs before accessing the output stream
[15:59:42 CEST] <wm4> hm, lots of hackery and assumes files are freely and exactly seekable
[16:01:58 CEST] <D404|Ghetto> why does osmething like also accept anything but pcm?
[16:02:02 CEST] <D404|Ghetto> alsa*
[16:05:36 CEST] <wm4> D404|Ghetto: it's just a separate player
[16:05:45 CEST] <D404|Ghetto> o
[16:05:49 CEST] <wm4> I don't even know why it's named like this, because it supports other audio APIs too
[16:05:54 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:6b15874fc2c3: af_resample: do not touch the timestamps if we are not resampling
[16:05:55 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:7093e215d02e: Merge commit '6b15874fc2c3f565732201f7907ae1112727d6ae'
[16:06:20 CEST] <wm4> and I only remember it because I found its ability to set speed to negative values kind of fancy
[16:08:51 CEST] <ubitux> h
[16:08:54 CEST] <ubitux> oups
[16:31:50 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:f5c4d38c7834: qsvdec: properly handle asynchronous decoding
[16:31:51 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:f929081f2e64: Merge commit 'f5c4d38c78347b09478e21a661befff4b2d44643'
[16:44:44 CEST] <cone-506> ffmpeg 03Anton Khirnov 07master:69b92f1b99f3: qsvenc: properly handle asynchronous encoding
[16:44:45 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:9c943e077bfc: Merge commit '69b92f1b99f3f210be19ee6ec06f6c0de1733031'
[17:37:39 CEST] <Daan> Dear, Anyone who can develop a live audio spectrum like in this video: https://www.youtube.com/watch?v=c-1bTqZB8cE ? Regards,
[17:43:25 CEST] <BtbN> http://video.stackexchange.com/questions/9644/how-do-i-turn-audio-into-video-that-is-show-the-waveforms-in-a-video
[18:25:20 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:ea8785cedae2: avcodec/utils: Check that channel count is valid in avcodec_open2();
[18:28:07 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:f66d2bf94909: mpegaudiodec: copy AVFloatDSPContext from first context to all contexts
[18:28:08 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:2e1226a695dd: vc1dec: use get_bits_long and limit the read bits to 32
[18:28:09 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:729337295982: s302m: fix arithmetic exception
[18:28:11 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:eddf146ada12: nutdec: check maxpos in read_sm_data before returning success
[18:28:11 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:95bd0f3a4b65: wavpack: use get_bits_long to read up to 32 bits
[18:28:12 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:43f8a422b343: huffyuvdec: validate image size
[18:28:13 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:acfad331adde: pthread_frame: forward error codes when flushing
[18:28:14 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:1d1adf5ff420: ffmpeg: exit_on_error if decoding a packet failed
[18:28:15 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:70b97a89d2d7: ffmpeg: only count got_output/errors in decode_error_stat
[18:28:16 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:f0af6e705f3b: wavpack: limit extra_bits to 32 and use get_bits_long
[18:28:17 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:762a5878a6b0: webp: fix infinite loop in webp_decode_frame
[18:28:18 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:26cb351452e7: snow: remove an obsolete av_assert2
[18:28:19 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:f06d9dced4c7: hevc: check slice address length
[18:28:20 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:27816fb9ef26: imc: use correct position for flcoeffs2 calculation
[18:28:21 CEST] <cone-506> ffmpeg 03Andreas Cadhalpun 07release/2.7:15466db69e60: Changelog update
[19:56:04 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:4df66c7cd6f7: avformat/mp3dec: split position sync code out
[20:07:20 CEST] <durandal_170> anybody against random filter?
[20:11:52 CEST] <Compn> its too random
[20:11:55 CEST] Action: Compn jokes
[20:30:00 CEST] <cone-506> ffmpeg 03Vesselin Bontchev 07master:0a551cbe97e0: Add support for Audible AAX (and AAX+) files
[20:45:10 CEST] <cone-506> ffmpeg 03Chris Spencer 07master:eb468b0be03e: avdevice/decklink: Add missing libraries when building with DeckLink support on Windows.
[21:12:52 CEST] <durandal_170> huh I added bunch of lines to lavfi
[21:24:35 CEST] <cone-506> ffmpeg 03Paul B Mahol 07master:a8fbb7b62597: avfilter: add random video filter
[21:34:42 CEST] <j-b> Can someone break the DRM?
[21:40:50 CEST] <ubitux> durandal_170: you can remove !ctx->is_disabled in random filter
[21:41:52 CEST] <durandal_170> j-b: DRM?
[21:43:06 CEST] <j-b> durandal_170: the AAX commit
[21:45:06 CEST] <durandal_170> afaik it is bruteforce, key is 4 bytes
[21:46:31 CEST] <j-b> -activation_bytes
[21:46:36 CEST] <nevcairiel> j-b: you can't "break" such a DRM, i mean, the implementaiton is known, you jsut need the key
[21:46:46 CEST] <nevcairiel> and the only way to "break" such a key is to bruteforce it =p
[21:47:01 CEST] <j-b> sure, but maybe the master-key was known, or similar
[21:47:08 CEST] <j-b> like for the MD drm
[21:47:12 CEST] <jamrial> anyone with clang want to report this recent failure http://fate.ffmpeg.org/report.cgi?time=20150719174819&slot=x86_64-darwin-clang-3.7-O3 upstream before 3.7 is released?
[21:47:22 CEST] <ubitux> .str="77214d4b196a87cd520045fd20a51d67"
[21:47:31 CEST] <ubitux> so there are 2 keys?
[21:47:52 CEST] <ubitux> audible_fixed_key ok
[21:49:27 CEST] <nevcairiel> i think you only need the activation key
[21:49:42 CEST] <ubitux> note: asan seems not happy about channelmap
[21:49:57 CEST] <nevcairiel> the other one is apparently  constant
[21:49:59 CEST] <nevcairiel> but shrug
[21:51:38 CEST] <j-b> sure, but testing 2^32 keys might not be the best :)
[21:51:51 CEST] <rcombs> well if you have DRM-encrypted content you usually have the key as well
[21:51:57 CEST] <durandal_170> ubitux: anything obvious?
[21:52:17 CEST] <j-b> can we infer something out of the fact it's aac in it?
[21:52:25 CEST] <ubitux> durandal_170: http://fate.ffmpeg.org/report.cgi?time=20150719183914&slot=x86_64-archlinux-gcc-asan
[21:52:34 CEST] <nevcairiel> jamrial: is there a way to sync the msys2 package database from some mirror after I already broke it by syncing from sf.net? :p
[21:52:53 CEST] <jamrial> use -yy instead of -y to force it :p
[21:53:14 CEST] <nevcairiel> ah
[21:54:03 CEST] <jamrial> sf is back online for that matter
[21:54:06 CEST] <nevcairiel> finally my fate box has tar again :D
[21:54:13 CEST] <rcombs> j-b: AFAIK AES isn't vulnerable to any known-plaintext attacks?
[21:54:16 CEST] <jamrial> or was ealier today
[21:54:16 CEST] <nevcairiel> i tried sf just now and still got the error message
[21:55:26 CEST] <jamrial> weird
[21:56:58 CEST] <rcombs> and I don't know of any relevant practically-exploitable flaws in SHA-1 either
[21:57:16 CEST] <rcombs> just figure out how to yank the key out of wherever you're getting the files from
[22:05:23 CEST] <nevcairiel> oh gcc 5.2 was also released
[22:05:27 CEST] <nevcairiel> can finally upgrade to that
[22:05:38 CEST] <JEEB> ye
[22:06:22 CEST] <nevcairiel> although the bug stats in their status report look kinda frightening, there is more issues now than before =p
[22:06:33 CEST] <nevcairiel> but that may as well be things just not reported before
[22:08:06 CEST] <jamrial> 5.1 was good, at least for ffmpeg
[22:09:32 CEST] <jamrial> it had some issues with c++11 and c++14, though
[22:12:16 CEST] <nevcairiel> tomorrow is also supposedly vs2015 release, should give my fatebox a proper cleanuo
[22:12:29 CEST] <nevcairiel> maybe setup a fresh vm to purge some of the cruft accumulated there =p
[22:14:08 CEST] <jamrial> will you drop the 2012 client once you get vs2015 running?
[22:14:15 CEST] <nevcairiel> probably
[22:14:36 CEST] <nevcairiel> i think that box even still has 2010 installed, although nothing is running them anymore
[22:15:17 CEST] <nevcairiel> personally i havent used 2012 with ffmpeg for ever now
[22:15:52 CEST] <nevcairiel> maybe i'll keep one, but dont run both 32 and 64 anymore
[22:16:15 CEST] <nevcairiel> just to check the converter still kinda works
[22:16:33 CEST] <jamrial> i only see a 32bits one in fate.ffmpeg.org
[22:17:19 CEST] <nevcairiel> oh right, i have a 64-bit for libav
[22:21:54 CEST] <nevcairiel> i should check how long the fate builds usually take, I could also add more and reduce the frequency a bit, if people think a 2012 build is still useful
[22:23:05 CEST] <jamrial> imo, no
[22:24:59 CEST] <jamrial> it suffers from a known bug that Microsoft doesn't plan to fix, and needs c99-to-c89
[22:25:27 CEST] <nevcairiel> they did fix it
[22:25:30 CEST] <nevcairiel> in 2013
[22:25:31 CEST] <nevcairiel> :P
[22:25:34 CEST] <jamrial> heh
[22:25:44 CEST] <nevcairiel> thats the answer you'll get from them
[22:25:49 CEST] <jamrial> better use your rigs to run gcc 4.9, 5.2 and vs2013/5
[22:26:08 CEST] <jamrial> don't drop the 4.9 clients. we barely have any of those now that ubitux migrated to 5 :p
[22:26:14 CEST] <nevcairiel> i'm not setup to run two gccs :(
[22:26:22 CEST] <nevcairiel> but i probably could set that up somehow
[22:39:00 CEST] <nevcairiel> in factits probably easier to get this going now that i use msys2 instead of this unified msys blob i used before that behaved a bit peculiar when mingw/gcc wasn't in an exact predefined spot =p
[22:49:06 CEST] <cone-506> ffmpeg 03James Almer 07master:2db8c42ddc58: avformat/mp3dec: remove unused variable
[22:53:36 CEST] <cone-506> ffmpeg 03James Almer 07master:4ebe31abf1b4: RELEASE: update to 2.7.git
[22:53:39 CEST] <jamrial> ^that was long overdue
[22:56:19 CEST] <jamrial> michaelni: the n2.8-dev tag is missing
[22:59:41 CEST] <michaelni> jamrial, feel free to add it, i forgot
[22:59:53 CEST] Action: michaelni always forgets
[23:00:02 CEST] <jamrial> not idea how to do it. never added tags with git :p
[23:04:00 CEST] <michaelni> IIRC git tag -a n2.8-dev HASH
[23:04:05 CEST] <michaelni> dunno which hash
[23:04:18 CEST] <michaelni> and then push origin  n2.8-dev (-n)
[23:04:34 CEST] <michaelni> but double check its correct before pushing
[23:04:41 CEST] <michaelni> it cannot be changed later 
[23:17:49 CEST] <cone-506> ffmpeg 03James Almer 07n2.8-dev:HEAD: RELEASE: update to 2.7.git
[23:20:09 CEST] <jamrial> guess it worked
[23:23:13 CEST] <michaelni> thx
[23:39:06 CEST] <cone-506> ffmpeg 03Michael Niedermayer 07master:36a87c2dc7f3: avcodec/dds: fix paletted files
[00:00:00 CEST] --- Mon Jul 20 2015