collect scripts in oxtools

.bzrignore

@@ -0,0 +1 @@
+build

README

@@ -0,0 +1,27 @@
+oxgt-tools
+
+Tools
+  oxframe
+    get frame from movie, image can be resized by adding --width or --height
+  oxtimeline
+    create timeline from video
+  oxinfo
+    output information about video, output can in xml, json or cfg format
+  oxposter
+    render 0xdb poster
+  oxicon
+    extract icon from frame
+
+Python API
+  import oxgst
+  
+  video = oxgst.Video(videoFile)
+  video.frame(pos_in_nanoseconds)
+  >>> <PIL Image>
+  
+  timeline = oxgst.Timeline(videoFile)
+  timeline.extract(timeline_prefix, width, height)
+
+  info = oxgst.Info(videoFile)
+  info.metadata
+  {videoCodec:..}

bin/oxframe

@@ -0,0 +1,56 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2008
+
+import os
+import sys
+from optparse import OptionParser
+
+import pygst
+pygst.require("0.10")
+import gst
+import Image
+
+root = os.path.join(os.path.abspath(os.path.dirname(__file__)), '..')
+if os.path.exists(os.path.join(root, 'oxgst')):
+    sys.path.insert(0, root)
+
+from oxgst import Video
+
+
+if __name__ == '__main__':
+    width = None
+    height = None
+    size = None
+    parser = OptionParser()
+    parser.add_option('-x', '--width', dest='width', help='scale image to given width')
+    parser.add_option('-y', '--height', dest='height', help='scale image to given height')
+    parser.add_option('-p', '--pos', dest='pos', help='frame position in milliseconds')
+    parser.add_option('-i', '--input', dest='input', help='video input')
+    parser.add_option('-o', '--output', dest='output', help='path to save frame to, jpg, png supported')
+    (opts, args) = parser.parse_args()
+    if None in (opts.input, opts.output, opts.pos):
+        parser.print_help()
+        sys.exit()
+    if opts.width:
+        width = int(opts.width)
+    if opts.height:
+        height = int(opts.height)
+    video = Video(opts.input)
+    timestamp = int(float(opts.pos) * gst.SECOND)
+
+    frame = video.frame(timestamp)
+    if frame:
+        if width:
+            height = int(float(frame.size[1])/frame.size[0] * width)
+            height = height + height%2
+            size = (width, height)
+        elif height:
+            width = int(float(frame.size[0])/frame.size[1] * height)
+            width = width + width%2
+            size = (width, height)
+        if size:
+            frame = frame.resize(size, Image.ANTIALIAS)
+        frame.save(opts.output)
+

bin/oxicon

@@ -0,0 +1,15 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+import os
+import sys
+
+root = os.path.join(os.path.abspath(os.path.dirname(__file__)), '..')
+if os.path.exists(os.path.join(root, 'oxgst')):
+    sys.path.insert(0, root)
+
+from oxposter import icon
+
+if __name__ == "__main__":
+    icon.main()
+

bin/oxinfo

@@ -0,0 +1,48 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2009
+
+import os
+import sys
+import Image
+from optparse import OptionParser
+import xml.etree.ElementTree as ET
+
+root = os.path.join(os.path.abspath(os.path.dirname(__file__)), '..')
+if os.path.exists(os.path.join(root, 'oxgst')):
+    sys.path.insert(0, root)
+
+from oxgst import Info
+
+if __name__ == "__main__":
+
+    parser = OptionParser()
+    parser.add_option('-f', '--format', dest='format', help='output format: cfg, json, xml default: cfg')
+    (opts, args) = parser.parse_args()
+
+    if not args:
+        parser.print_help()
+        sys.exit()
+
+    inputFile = args[0]
+    i = Info(inputFile)
+    info = i.metadata
+    if opts.format == 'xml':
+        xml = ET.Element("gstinfo")
+        el = ET.SubElement(xml, "path")
+        el.text = inputFile
+        for key in sorted(info):
+            el = ET.SubElement(xml, key)
+            el.text = unicode(info[key])
+        print u'<?xml version="1.0" encoding="utf-8"?>\n' + ET.tostring(xml).replace('><', '>\n  <').replace('  </', '</')
+
+    elif opts.format == 'json':
+        import simplejson
+        info['path'] = inputFile
+        print simplejson.dumps(info).replace(', ', ',\n ')
+    else:
+        print "[%s]" % inputFile
+        for key in sorted(info):
+            print "%s = %s" % (key, info[key])
+

bin/oxposter

@@ -0,0 +1,15 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+import os
+import sys
+
+root = os.path.join(os.path.abspath(os.path.dirname(__file__)), '..')
+if os.path.exists(os.path.join(root, 'oxgst')):
+    sys.path.insert(0, root)
+
+from oxposter import poster
+
+if __name__ == "__main__":
+    poster.main()
+

bin/oxtimeline

@@ -0,0 +1,38 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2008
+
+import os
+import sys
+from optparse import OptionParser
+
+import Image
+
+root = os.path.join(os.path.abspath(os.path.dirname(__file__)), '..')
+if os.path.exists(os.path.join(root, 'oxgst')):
+    sys.path.insert(0, root)
+
+
+import oxgst
+
+if __name__ == '__main__':
+    parser = OptionParser()
+    parser.add_option('-x', '--width', dest='width', help='pixels per tile, defaults to 1500px', default=1500)
+    parser.add_option('-y', '--height', dest='height', help='timeline height, defaults to 64px', default=64)
+    parser.add_option('-o', '--prefix', dest='prefix', help='prefix for timeline tiles')
+    parser.add_option('-i', '--input', dest='input', help='video input')
+    (opts, args) = parser.parse_args()
+
+    if None in (opts.prefix, opts.input):
+        parser.print_help()
+        sys.exit()
+
+    timeline = oxgst.Timeline(opts.input)
+    timeline.extract(opts.prefix, opts.width, opts.height)
+    oxgst.timeline.createTimelineMultiline(opts.prefix)
+    oxgst.timeline.makeTiles(opts.prefix, 16)
+    oxgst.timeline.makeTimelineOverview(opts.prefix, 300)
+    oxgst.timeline.makeTimelineOverview(opts.prefix, 600)
+
+

oxgst/__init__.py

@@ -0,0 +1,7 @@
+import timeline
+import video
+import info
+from video import Video
+from timeline import Timeline
+from info import Info
+

oxgst/imagesink.py

@@ -0,0 +1,67 @@
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2008
+
+import gobject
+
+import pygst
+pygst.require("0.10")
+import gst
+import Image
+
+from singledecodebin import SingleDecodeBin
+
+
+class ImageSink(gst.BaseSink):
+
+    __gsignals__ = {
+        "frame" : (gobject.SIGNAL_RUN_LAST,
+                       gobject.TYPE_NONE,
+                       ( gobject.TYPE_PYOBJECT, gobject.TYPE_UINT64 ))
+        }
+
+    __gsttemplates__ = (
+        gst.PadTemplate ("sink",
+                         gst.PAD_SINK,
+                         gst.PAD_ALWAYS,
+                         gst.Caps("video/x-raw-rgb,"
+                                  "bpp = (int) 24, depth = (int) 24,"
+                                  "endianness = (int) BIG_ENDIAN,"
+                                  "red_mask = (int) 0x00FF0000, "
+                                  "green_mask = (int) 0x0000FF00, "
+                                  "blue_mask = (int) 0x000000FF, "
+                                  "width = (int) [ 1, max ], "
+                                  "height = (int) [ 1, max ], "
+                                  "framerate = (fraction) [ 0, max ]"))
+        )
+
+    def __init__(self):
+        gst.BaseSink.__init__(self)
+        self.width = 1
+        self.height = 1
+        self.set_sync(False)
+
+    def do_set_caps(self, caps):
+        self.log("caps %s" % caps.to_string())
+        self.log("padcaps %s" % self.get_pad("sink").get_caps().to_string())
+        self.width = caps[0]["width"]
+        self.height = caps[0]["height"]
+        self.framerate = caps[0]["framerate"]
+
+        if not caps[0].get_name() == "video/x-raw-rgb":
+            return False
+        return True
+
+    def do_render(self, buf):
+        self.log("buffer %s %d" % (gst.TIME_ARGS(buf.timestamp),
+                                   len(buf.data)))
+
+        frame = Image.fromstring('RGB', (self.width, self.height), buf.data)
+        self.emit('frame', frame, buf.timestamp)
+        return gst.FLOW_OK
+
+    def do_preroll(self, buf):
+        return self.do_render(buf)
+
+gobject.type_register(ImageSink)
+

oxgst/info.py

@@ -0,0 +1,128 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2008
+import gobject
+gobject.threads_init()
+
+import os
+import pygst
+pygst.require("0.10")
+import gst
+
+codec_list = {
+    'MPEG-1 layer 3': 'MPEG-1 layer 3',
+    'MPEG-1 layer 3 audio': 'MPEG-1 layer 3',
+    'VP6 Flash video': 'VP6',
+    'AC-3 audio': 'AC-3',
+    'Uncompressed 16-bit PCM audio': 'Uncompressed 16-bit PCM',
+    'Generic DV': 'DV Video',
+}
+
+class Info:
+    video_done = True
+    audio_done = True
+    video = None
+    audio = None
+    metadata = {}
+    tags = {}
+
+    def __init__(self, videofile):
+        self.mainloop = gobject.MainLoop()
+        self.pipeline = gst.parse_launch('filesrc name=input ! decodebin name=dbin')
+        self.input  = self.pipeline.get_by_name('input')
+        self.input.props.location = videofile
+        self.dbin  = self.pipeline.get_by_name('dbin')
+
+        self.metadata['size'] = os.stat(videofile).st_size
+
+        self.bus = self.pipeline.get_bus()
+        self.dbin.connect('new-decoded-pad', self.demux_pad_added)
+
+        self.bus.add_signal_watch()
+        self.watch_id = self.bus.connect("message", self.onBusMessage)
+
+        self.pipeline.set_state(gst.STATE_PAUSED)
+        self.pipeline.get_state()
+
+        #duration
+        pads = None
+        if self.video:
+            pads = self.video.sink_pads()
+        elif self.audio:
+            pads = self.audio.sink_pads()
+        if pads:
+            q = gst.query_new_duration(gst.FORMAT_TIME)
+            for pad in pads:
+                if pad.get_peer() and pad.get_peer().query(q):
+                    format, self.duration = q.parse_duration()
+                    self.metadata["duration"] = self.duration/gst.MSECOND
+
+        self.mainloop.run()
+        if 'video-codec' in self.tags:
+            self.metadata['video-codec'] = codec_list.get(self.tags['video-codec'], self.tags['video-codec'])
+        if 'audio-codec' in self.tags:
+            self.metadata['audio-codec'] = codec_list.get(self.tags['audio-codec'], self.tags['audio-codec'])
+
+    def get_audio_info_cb(self, sink, buffer, pad):
+        caps = sink.sink_pads().next().get_negotiated_caps()
+        for s in caps:
+            self.metadata["channels"] = s['channels']
+            self.metadata["samplerate"] = s['rate']
+        self.audio.disconnect(self.audio_cb)
+        self.audio_done = True
+        if self.audio_done and self.video_done:
+            self.bus.post(gst.message_new_eos(self.pipeline))
+
+    def get_frame_info_cb(self, sink, buffer, pad):
+        caps = sink.sink_pads().next().get_negotiated_caps()
+        for s in caps:
+            self.metadata["width"] = s['width']
+            self.metadata["height"] = s['height']
+            self.metadata["framerate"] = float(s['framerate'])
+            if 'pixel-aspect-ratio' in s.keys():
+                self.metadata["pixel-aspect-ratio"] = "%d:%d" % (s['pixel-aspect-ratio'].num, s['pixel-aspect-ratio'].denom)
+        self.video.disconnect(self.video_cb)
+        self.video_done = True
+        if self.audio_done and self.video_done:
+            self.bus.post(gst.message_new_eos(self.pipeline))
+
+    def demux_pad_added(self, element, pad, bool):
+        caps = pad.get_caps()
+        structure = caps[0]
+        stream_type = structure.get_name()
+        if stream_type.startswith('video'):
+            colorspace = gst.element_factory_make("ffmpegcolorspace");
+            self.pipeline.add (colorspace);
+            colorspace.set_state (gst.STATE_PLAYING);
+            pad.link (colorspace.get_pad("sink"));
+
+            self.video_done = False
+            self.video = gst.element_factory_make("fakesink")
+            self.video.props.signal_handoffs = True
+            self.pipeline.add(self.video)
+            self.video.set_state (gst.STATE_PLAYING);
+            colorspace.link (self.video);
+            self.video_cb = self.video.connect("handoff", self.get_frame_info_cb)
+        elif stream_type.startswith('audio'):
+            self.audio_done = False
+            self.audio = gst.element_factory_make("fakesink")
+            self.audio.props.signal_handoffs = True
+            self.pipeline.add(self.audio)
+            self.audio.set_state (gst.STATE_PLAYING);
+            pad.link(self.audio.get_pad('sink'))
+            self.audio_cb = self.audio.connect("handoff", self.get_audio_info_cb)
+
+    def quit(self):
+        self.pipeline.set_state(gst.STATE_NULL)
+        self.pipeline.get_state()
+        self.mainloop.quit()
+
+    def onBusMessage(self, bus, message):
+        if message.type == gst.MESSAGE_TAG:
+            for key in message.parse_tag().keys():
+                self.tags[key] = message.structure[key]
+        if message.src == self.pipeline and message.type == gst.MESSAGE_EOS:
+            self.quit()
+
+

oxgst/singledecodebin.py

@@ -0,0 +1,305 @@
+# -*- coding: utf-8 -*-
+# -*- Mode: Python; -*-
+# vi:si:et:sw=4:sts=4:ts=4
+#
+#       pitivi/elements/singledecodebin.py
+#
+# Copyright (c) 2005, Edward Hervey <bilboed@bilboed.com>
+#
+# This program is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 of the License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this program; if not, write to the
+# Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+# Boston, MA 02111-1307, USA.
+
+"""
+Single-stream queue-less decodebin
+"""
+
+import gobject
+import gst
+
+def is_raw(caps):
+    """ returns True if the caps are RAW """
+    rep = caps.to_string()
+    valid = ["video/x-raw", "audio/x-raw", "text/plain", "text/x-pango-markup"]
+    for val in valid:
+        if rep.startswith(val):
+            return True
+    return False
+
+class SingleDecodeBin(gst.Bin):
+
+    __gsttemplates__ = (
+        gst.PadTemplate ("sinkpadtemplate",
+                         gst.PAD_SINK,
+                         gst.PAD_ALWAYS,
+                         gst.caps_new_any()),
+        gst.PadTemplate ("srcpadtemplate",
+                         gst.PAD_SRC,
+                         gst.PAD_SOMETIMES,
+                         gst.caps_new_any())
+        )
+    def __init__(self, caps=None, uri=None, *args, **kwargs):
+        gst.Bin.__init__(self, *args, **kwargs)
+        if not caps:
+            caps = gst.caps_new_any()
+        self.caps = caps
+        self.typefind = gst.element_factory_make("typefind", "internal-typefind")
+        self.add(self.typefind)
+
+        self.uri = uri
+        if self.uri and gst.uri_is_valid(self.uri):
+            self.urisrc = gst.element_make_from_uri(gst.URI_SRC, uri, "urisrc")
+            self.log("created urisrc %s / %r" % (self.urisrc.get_name(),
+                                                 self.urisrc))
+            self.add(self.urisrc)
+            self.urisrc.link(self.typefind)
+        else:
+            self._sinkpad = gst.GhostPad("sink", self.typefind.get_pad("sink"))
+            self._sinkpad.set_active(True)
+            self.add_pad(self._sinkpad)
+
+        self.typefind.connect("have_type", self._typefindHaveTypeCb)
+
+        self._srcpad = None
+
+        self._dynamics = []
+
+        self._validelements = [] #added elements
+
+        self._factories = self._getSortedFactoryList()
+
+
+    ## internal methods
+
+    def _controlDynamicElement(self, element):
+        self.log("element:%s" % element.get_name())
+        self._dynamics.append(element)
+        element.connect("pad-added", self._dynamicPadAddedCb)
+        element.connect("no-more-pads", self._dynamicNoMorePadsCb)
+
+    def _getSortedFactoryList(self):
+        """
+        Returns the list of demuxers, decoders and parsers available, sorted
+        by rank
+        """
+        def myfilter(fact):
+            if fact.get_rank() < 64 :
+                return False
+            klass = fact.get_klass()
+            if not ("Demuxer" in klass or "Decoder" in klass or "Parse" in klass):
+                return False
+            return True
+        reg = gst.registry_get_default()
+        res = [x for x in reg.get_feature_list(gst.ElementFactory) if myfilter(x)]
+        res.sort(lambda a, b: int(b.get_rank() - a.get_rank()))
+        return res
+
+    def _findCompatibleFactory(self, caps):
+        """
+        Returns a list of factories (sorted by rank) which can take caps as
+        input. Returns empty list if none are compatible
+        """
+        self.debug("caps:%s" % caps.to_string())
+        res = []
+        for factory in self._factories:
+            for template in factory.get_static_pad_templates():
+                if template.direction == gst.PAD_SINK:
+                    intersect = caps.intersect(template.static_caps.get())
+                    if not intersect.is_empty():
+                        res.append(factory)
+                        break
+        self.debug("returning %r" % res)
+        return res
+
+    def _closeLink(self, element):
+        """
+        Inspects element and tries to connect something on the srcpads.
+        If there are dynamic pads, it sets up a signal handler to
+        continue autoplugging when they become available.
+        """
+        to_connect = []
+        dynamic = False
+        templates = element.get_pad_template_list()
+        for template in templates:
+            if not template.direction == gst.PAD_SRC:
+                continue
+            if template.presence == gst.PAD_ALWAYS:
+                pad = element.get_pad(template.name_template)
+                to_connect.append(pad)
+            elif template.presence == gst.PAD_SOMETIMES:
+                pad = element.get_pad(template.name_template)
+                if pad:
+                    to_connect.append(pad)
+                else:
+                    dynamic = True
+            else:
+                self.log("Template %s is a request pad, ignoring" % pad.name_template)
+
+        if dynamic:
+            self.debug("%s is a dynamic element" % element.get_name())
+            self._controlDynamicElement(element)
+
+        for pad in to_connect:
+            self._closePadLink(element, pad, pad.get_caps())
+
+    def _tryToLink1(self, source, pad, factories):
+        """
+        Tries to link one of the factories' element to the given pad.
+
+        Returns the element that was successfully linked to the pad.
+        """
+        self.debug("source:%s, pad:%s , factories:%r" % (source.get_name(),
+                                                         pad.get_name(),
+                                                         factories))
+        result = None
+        for factory in factories:
+            element = factory.create()
+            if not element:
+                self.warning("weren't able to create element from %r" % factory)
+                continue
+
+            sinkpad = element.get_pad("sink")
+            if not sinkpad:
+                continue
+
+            self.add(element)
+            try:
+                pad.link(sinkpad)
+            except:
+                element.set_state(gst.STATE_NULL)
+                self.remove(element)
+                continue
+
+            self._closeLink(element)
+            element.set_state(gst.STATE_PAUSED)
+
+            result = element
+            break
+
+        return result
+
+    def _closePadLink(self, element, pad, caps):
+        """
+        Finds the list of elements that could connect to the pad.
+        If the pad has the desired caps, it will create a ghostpad.
+        If no compatible elements could be found, the search will stop.
+        """
+        self.debug("element:%s, pad:%s, caps:%s" % (element.get_name(),
+                                                    pad.get_name(),
+                                                    caps.to_string()))
+        if caps.is_empty():
+            self.log("unknown type")
+            return
+        if caps.is_any():
+            self.log("type is not know yet, waiting")
+            return
+        if caps.intersect(self.caps):
+            # This is the desired caps
+            if not self._srcpad:
+                self._wrapUp(element, pad)
+        elif is_raw(caps):
+            self.log("We hit a raw caps which isn't the wanted one")
+            # FIXME : recursively remove everything until demux/typefind
+
+        else:
+            # Find something
+            if len(caps) > 1:
+                self.log("many possible types, delaying")
+                return
+            facts = self._findCompatibleFactory(caps)
+            if not facts:
+                self.log("unknown type")
+                return
+            self._tryToLink1(element, pad, facts)
+
+    def _wrapUp(self, element, pad):
+        """
+        Ghost the given pad of element.
+        Remove non-used elements.
+        """
+
+        if self._srcpad:
+            return
+        self._markValidElements(element)
+        self._removeUnusedElements(self.typefind)
+        self.log("ghosting pad %s" % pad.get_name)
+        self._srcpad = gst.GhostPad("src", pad)
+        self._srcpad.set_active(True)
+        self.add_pad(self._srcpad)
+        self.post_message(gst.message_new_state_dirty(self))
+
+    def _markValidElements(self, element):
+        """
+        Mark this element and upstreams as valid
+        """
+        self.log("element:%s" % element.get_name())
+        if element == self.typefind:
+            return
+        self._validelements.append(element)
+        # find upstream element
+        pad = list(element.sink_pads())[0]
+        parent = pad.get_peer().get_parent()
+        self._markValidElements(parent)
+
+    def _removeUnusedElements(self, element):
+        """
+        Remove unused elements connected to srcpad(s) of element
+        """
+        self.log("element:%s" % element)
+        for pad in element.src_pads():
+            if pad.is_linked():
+                peer = pad.get_peer().get_parent()
+                self._removeUnusedElements(peer)
+                if not peer in self._validelements:
+                    self.log("removing %s" % peer.get_name())
+                    pad.unlink(pad.get_peer())
+                    peer.set_state(gst.STATE_NULL)
+                    self.remove(peer)
+
+    def _cleanUp(self):
+        self.log("")
+        if self._srcpad:
+            self.remove_pad(self._srcpad)
+        self._srcpad = None
+        for element in self._validelements:
+            element.set_state(gst.STATE_NULL)
+            self.remove(element)
+        self._validelements = []
+
+    ## Overrides
+
+    def do_change_state(self, transition):
+        self.debug("transition:%r" % transition)
+        res = gst.Bin.do_change_state(self, transition)
+        if transition in [gst.STATE_CHANGE_PAUSED_TO_READY, gst.STATE_CHANGE_READY_TO_NULL]:
+            self._cleanUp()
+        return res
+
+    ## Signal callbacks
+
+    def _typefindHaveTypeCb(self, typefind, probability, caps):
+        self.debug("probability:%d, caps:%s" % (probability, caps.to_string()))
+        self._closePadLink(typefind, typefind.get_pad("src"), caps)
+
+    ## Dynamic element Callbacks
+
+    def _dynamicPadAddedCb(self, element, pad):
+        self.log("element:%s, pad:%s" % (element.get_name(), pad.get_name()))
+        if not self._srcpad:
+            self._closePadLink(element, pad, pad.get_caps())
+
+    def _dynamicNoMorePadsCb(self, element):
+        self.log("element:%s" % element.get_name())
+
+gobject.type_register(SingleDecodeBin)

oxgst/timeline.py

@@ -0,0 +1,216 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2008
+import gobject
+gobject.threads_init()
+
+from glob import glob
+import math
+import os
+import time
+
+import pygst
+pygst.require("0.10")
+import gst
+import Image
+
+from singledecodebin import SingleDecodeBin
+from imagesink import ImageSink
+from video import Video
+
+
+class Timeline(Video):
+    def __init__(self, uri):
+        Video.__init__(self, uri)
+
+        bus = self.get_bus()
+        bus.add_signal_watch()
+        self.watch_id = bus.connect("message", self.onBusMessage)
+
+        self.mainloop = gobject.MainLoop()
+
+    def extract(self, prefix, width, height):
+        self.tile_width = width
+        self.tile_height = height
+        self.prefix = prefix
+        self.timeline_fps = 25
+        self.input_tile_width = int(math.ceil((float(self.framerate)/self.timeline_fps) * width))
+        ntiles = int(math.ceil(float(self.frames)/self.input_tile_width))
+        self.tiles = []
+        for i in range(ntiles):
+            tile = Image.new("RGB", (self.input_tile_width, height))
+            self.tiles.append(tile)
+        
+        self.set_state(gst.STATE_PLAYING)
+        self.mainloop.run()
+
+        for  i in range(ntiles):
+            tile = self.tiles[i]
+            if tile.size[0] != self.tile_width:
+                tile = tile.resize((self.tile_width, self.tile_height), Image.ANTIALIAS)
+            if i < (ntiles-1):
+                frames = self.input_tile_width
+            else:
+                frames = self.frames-((ntiles-1)*self.input_tile_width)
+            tile_width = int(math.ceil(self.timeline_fps*frames)/float(self.framerate))
+            if -2 < self.tile_width - tile_width < 2:
+                tile_width = self.tile_width
+            tile = tile.crop((0, 0, tile_width, self.tile_height))
+            filename = "%s.%s.%04d.png" % (self.prefix, self.tile_height, i)
+            tile.save(filename)
+
+    def done(self):
+        self.mainloop.quit()
+
+    def _sbinPadAddedCb(self, unused_sbin, pad):
+        self.log("pad : %s" % pad)
+        pad.link(self.csp.get_pad("sink"))
+
+    def _frameCb(self, unused_thsink, frame, timestamp):
+        self.log("image:%s, timestamp:%s" % (frame, gst.TIME_ARGS(timestamp)))
+
+        if not self._ready:
+            # we know we're prerolled when we get the initial thumbnail
+            self._ready = True
+        else:
+            framePos = int(math.ceil((float(timestamp) / (gst.SECOND) * float(self.framerate))))
+            tile = int(math.floor(float(framePos) / self.input_tile_width))
+            tilePos = framePos - (tile * self.input_tile_width)
+            frame = frame.resize((1, self.tile_height), Image.ANTIALIAS)
+            for i in range(self.tile_height):
+                self.tiles[tile].putpixel((tilePos, i), frame.getpixel((0, i)))
+
+            if self.mainloop and timestamp >= self.duration:
+                self.done()
+
+    def onBusMessage(self, bus, message):
+        if message.src == self and message.type == gst.MESSAGE_EOS:
+            self.done()
+
+def loadTimeline(timeline_prefix, height=64):
+    files = sorted(glob('%s.%s.*.png' % (timeline_prefix, height)))
+    f = Image.open(files[0])
+    width = f.size[0]
+    f = Image.open(files[-1])
+    duration = f.size[0] + (len(files)-1)*width
+    timeline = Image.new("RGB", (duration, height))
+    pos = 0
+    for f in files:
+        part = Image.open(f)
+        timeline.paste(part, (pos, 0, pos + part.size[0], height))
+        pos += part.size[0]
+    return timeline
+    
+def createTimelineMultiline(timeline_prefix, width=600, height=16):
+    lineWidth = width
+    timlelineHeight = height
+
+    timeline = loadTimeline(timeline_prefix)
+    duration = timeline.size[0]
+
+    width = duration/25 #one pixel per second
+    timeline = timeline.resize((width, timlelineHeight), Image.ANTIALIAS).convert('RGBA')
+
+    lineHeight = timlelineHeight + 2 * 4 
+
+    lines = int(math.ceil(width / lineWidth) + 1)
+    size = (lineWidth, lineHeight * lines)
+
+    timelineColor = (64, 64, 64)
+    i = Image.new("RGBA", size)
+
+    #padd end with nothing to fit to grid
+    t = Image.new("RGBA", (lineWidth * lines, timlelineHeight))
+    t.paste(timeline, (0, 0))
+
+    for currentLine in range(0, lines):
+        offset = currentLine * lineHeight + 4
+        toffset = currentLine * lineWidth
+        try:
+            tbox = t.crop((toffset, 0, toffset + lineWidth, timlelineHeight))
+            box = ((0, offset , tbox.size[0], offset + tbox.size[1]))
+            i.paste(tbox, box)
+        except:
+            broken = True
+            width = lineWidth
+            if currentLine == lines -1:
+                width = duration - (lines - 1) * lineWidth
+            box = ((0, offset , width, offset + timlelineHeight))
+            i.paste(timelineColor, box)
+    timeline_file = '%s.timeline.overview.png' % (timeline_prefix)
+    i.save(timeline_file, 'PNG')
+
+def makeTimelineByFramesPerPixel(timeline_prefix, frames_per_pixel, inpoint=0, outpoint=0, height=16):
+    pos = 0
+    input_scale = 25
+    
+    timeline_file = '%s.timeline.%s.png' % (timeline_prefix, width)
+    if outpoint > 0:
+        timeline_file = '%s.timeline.%s.%d-%d.png' % (timeline_prefix, width, inpoint, outpoint)
+
+    timeline = loadTimeline(timeline_prefix)
+    duration = timeline.size[0]
+
+
+    width = duration / frames_per_pixel
+
+    if inpoint<=0:
+        inpoint = 0
+    else:
+        inpoint = inpoint * input_scale
+    if outpoint<=0:
+        outpoint = pos
+    else:
+        outpoint = outpoint * input_scale
+
+    timeline = timeline.crop((inpoint, 0, outpoint, timeline.size[1])).resize((width, height), Image.ANTIALIAS)
+    timeline.save(timeline_file)
+    
+def makeTimelineOverview(timeline_prefix, width, inpoint=0, outpoint=0, duration=-1, height=16):
+    input_scale = 25
+    
+    timeline_file = '%s.timeline.%s.png' % (timeline_prefix, width)
+    if outpoint > 0:
+        timeline_file = '%s.timeline.%s.%d-%d.png' % (timeline_prefix, width, inpoint, outpoint)
+
+    timeline = loadTimeline(timeline_prefix)
+    duration = timeline.size[0]
+
+    if inpoint<=0:
+        inpoint = 0
+    else:
+        inpoint = inpoint * input_scale
+    if outpoint<=0:
+        outpoint = duration
+    else:
+        outpoint = outpoint * input_scale
+
+    timeline = timeline.crop((inpoint, 0, outpoint, timeline.size[1])).resize((width, height), Image.ANTIALIAS)
+    timeline.save(timeline_file)
+
+def makeTiles(timeline_prefix, height=16):
+    files = glob('%s.64.*.png' % timeline_prefix)
+    part_step = 60
+    output_width = 300
+    width = len(files) * part_step
+    timeline = Image.new("RGB", (width, height))
+
+    pos = 0
+    for f in sorted(files):
+        part = Image.open(f)
+        part_width = int(part.size[0] / 25) 
+        part = part.resize((part_width, height), Image.ANTIALIAS)
+        timeline.paste(part, (pos, 0, pos+part_width, height))
+        pos += part_width
+
+    timeline = timeline.crop((0, 0, pos, height))
+
+    pos = 0
+    i = 0
+    while pos < timeline.size[0]:
+      end = min(pos+output_width, timeline.size[0])
+      timeline.crop((pos, 0, end, timeline.size[1])).save('%s.%s.%04d.png' % (timeline_prefix, timeline.size[1], i))
+      pos += output_width
+      i += 5
+

oxgst/video.py

@@ -0,0 +1,131 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+# GPL 2008
+import gobject
+gobject.threads_init()
+
+import pygst
+pygst.require("0.10")
+import gst
+
+import Image
+import time
+
+from singledecodebin import SingleDecodeBin
+from imagesink import ImageSink
+
+class Video(gst.Pipeline):
+
+    def __init__(self, uri):
+        gst.Pipeline.__init__(self)
+        self.duration = -1
+        # queue of timestamps
+        self.queue = []
+        # queue callbacks
+        self.callback = {}
+        # extracted frames
+        self.frames = {}
+
+        # true only if we are prerolled
+        self._ready = False
+        uri = 'file://' + uri
+        self.log("uri : %s" % uri)
+
+        self.uri = uri
+
+        self.sbin = SingleDecodeBin(caps=gst.Caps("video/x-raw-rgb;video/x-raw-yuv"),
+                                    uri=self.uri)
+        self.csp = gst.element_factory_make("ffmpegcolorspace")
+        self.sink = ImageSink()
+        self.sink.connect('frame', self._frameCb)
+
+        self.add(self.sbin, self.csp, self.sink)
+        self.csp.link(self.sink)
+
+        self.sbin.connect('pad-added', self._sbinPadAddedCb)
+        self.set_state(gst.STATE_PAUSED)
+        self.get_state()
+        self.width = self.sink.width
+        self.height = self.sink.height
+        self.framerate = self.sink.framerate
+        self.getDuration()
+        self.frames = int((float(self.duration) / gst.SECOND) * float(self.framerate))
+
+    def _sbinPadAddedCb(self, unused_sbin, pad):
+        self.log("pad : %s" % pad)
+        pad.link(self.csp.get_pad("sink"))
+
+    def _frameCb(self, unused_thsink, frame, timestamp):
+        self.log("image:%s, timestamp:%s" % (frame, gst.TIME_ARGS(timestamp)))
+
+        if not self._ready:
+            # we know we're prerolled when we get the initial thumbnail
+            self._ready = True
+        elif timestamp in self.callback and self.callback[timestamp]:
+            self.callback[timestamp](frame, timestamp)
+            del self.callback[timestamp]
+
+        if timestamp in self.queue:
+            self.queue.remove(timestamp)
+
+        if self.queue:
+            # still some more thumbnails to process
+            gobject.idle_add(self._getFrame, self.queue.pop(0))
+
+    def getDuration(self):
+        if self.duration < 0:
+            pads = self.sink.sink_pads()
+            q = gst.query_new_duration(gst.FORMAT_TIME)
+            for pad in pads:
+                if pad.get_peer() and pad.get_peer().query(q):
+                  format, self.duration = q.parse_duration()
+        return self.duration
+
+    def getFrame(self, timestamp, callback):
+        """
+            Queue a frame request for the given timestamp,
+            callback is called once frame is extracted.
+            returns False if timestamp > duration
+        """
+        
+        self.log("timestamp %s" % gst.TIME_ARGS(timestamp))
+        if self.duration < 0:
+            self.getDuration()
+        
+        if timestamp > self.duration:
+            self.log("timestamp %s > duration %s" % (timestamp, self.duration))
+            return False
+
+        self.callback[timestamp] = callback
+
+        if self.queue or not self._ready:
+            self.log('ready')
+            self.queue.append(timestamp)
+        else:
+
+            self.queue.append(timestamp)
+            self._getFrame(timestamp)
+        return True
+
+    def _getFrame(self, timestamp):
+        if not self._ready:
+            return
+        self.log("timestamp : %s" % gst.TIME_ARGS(timestamp))
+        self.seek(1.0, gst.FORMAT_TIME, gst.SEEK_FLAG_FLUSH | gst.SEEK_FLAG_ACCURATE,
+                  gst.SEEK_TYPE_SET, timestamp,
+                  gst.SEEK_TYPE_NONE, -1)
+        return False
+
+    def frame(self, timestamp):
+        mainloop = gobject.MainLoop()
+        self.frames[timestamp] = None
+        def callback(frame, timestamp):
+            self.frames[timestamp] = frame
+            mainloop.quit()
+        if self.getFrame(timestamp, callback):
+            mainloop.run()
+        frame = self.frames[timestamp]
+        del self.frames[timestamp]
+        return frame
+

oxposter/icon.py

@@ -0,0 +1,54 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+
+from __future__ import division
+from optparse import OptionParser
+import os
+import sys
+
+import Image
+import ImageDraw
+
+data_root = os.path.join(os.path.dirname(__file__), 'data')
+
+def main():
+
+    parser = OptionParser()
+    parser.add_option('-f', '--frame', dest='frame', help='Poster frame (image file to be read)')
+    parser.add_option('-i', '--icon', dest='icon', help='Icon (image file to be written)')
+    (opts, args) = parser.parse_args()
+    if not opts.icon:
+        parser.print_help()
+        sys.exit()
+
+    iconSize = 1024
+    iconImage = Image.new('RGBA', (iconSize, iconSize))
+
+    if opts.frame:
+        frameImage = Image.open(opts.frame)
+        if frameImage.size[0] >= frameImage.size[1]:
+            frameWidth = int(frameImage.size[0] * iconSize / frameImage.size[1])
+            frameImage = frameImage.resize((frameWidth, iconSize), Image.ANTIALIAS)
+            crop = int((frameWidth - iconSize) / 2)
+            frameImage = frameImage.crop((crop, 0, crop + iconSize, iconSize))
+        else:
+            frameHeight = int(frameImage.size[1] * iconSize / frameImage.size[0])
+            frameImage = frameImage.resize((iconSize, frameHeight), Image.ANTIALIAS)
+            crop = int((frameHeight - iconSize) / 2)
+            frameImage = frameImage.crop((0, crop, iconSize, crop + iconSize))
+        iconImage.paste(frameImage, (0, 0, iconSize, iconSize))
+    else:
+        draw = ImageDraw.Draw(iconImage)
+        draw.polygon([(0, 0), (iconSize, 0), (iconSize, iconSize), (0, iconSize)], fill=(0, 0, 0))
+        for y in range(iconSize):
+            for x in range(iconSize):
+                if int((x + y + 192) / 128) % 2:
+                    iconImage.putpixel((x, y), (32, 32, 32))
+        
+    maskImage = Image.open(os.path.join(data_root, 'icon.mask.png'))
+    iconImage.putalpha(maskImage)
+
+    iconImage.save(opts.icon)
+
+

oxposter/poster.py

@@ -0,0 +1,155 @@
+# -*- coding: utf-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+
+from __future__ import division
+from optparse import OptionParser
+import os
+import sys
+
+import Image
+import ImageDraw
+from oxlib import truncateString, wrapString
+
+data_root = os.path.join(os.path.dirname(__file__), 'data')
+
+def main():
+
+    parser = OptionParser()
+    parser.add_option('-o', '--oxdb', dest='oxdb', help='0xdb Id')
+    parser.add_option('-i', '--imdb', dest='imdb', help='IMDb Id')
+    parser.add_option('-t', '--title', dest='title', help='Movie title')
+    parser.add_option('-d', '--director', dest='director', help='Director(s)')
+    parser.add_option('-f', '--frame', dest='frame', help='Poster frame (image file to be read)')
+    parser.add_option('-p', '--poster', dest='poster', help='Poster (image file to be written)')
+    parser.add_option('-r', '--restricted', action='store_true', dest='restricted', help='If set, poster frame will have overlay')
+    (opts, args) = parser.parse_args()
+    if None in (opts.oxdb, opts.title, opts.director, opts.poster):
+        parser.print_help()
+        sys.exit()
+    opts.oxdb = '%s%s' % (opts.oxdb[:2], opts.oxdb[2:8].upper())
+    opts.title = opts.title.decode('utf-8')
+    opts.director = opts.director.decode('utf-8')
+    if opts.imdb == None:
+        id = opts.oxdb
+    else:
+        id = opts.imdb
+
+    posterWidth = 640
+    posterHeight = 1024
+    posterImage = Image.new('RGB', (posterWidth, posterHeight))
+    fontImage = Image.open(os.path.join(data_root, 'font.monaco.bold.png'))
+    oxdbImage = Image.open(os.path.join(data_root, 'logo.0xdb.large.png'))
+
+    # frame section
+    frameHeight = int(posterWidth * 10 / 16)
+    if opts.frame:
+        frameImage = Image.open(opts.frame)
+        if frameImage.size[0] / frameImage.size[1] > posterWidth / frameHeight:
+            # poster frame is too wide
+            frameImage = frameImage.resize((int(frameImage.size[0] * frameHeight / frameImage.size[1]), frameHeight), Image.ANTIALIAS)
+            crop = int((frameImage.size[0] - posterWidth) / 2)
+            frameImage = frameImage.crop((crop, 0, crop + posterWidth, frameHeight))
+        else:
+            # poster frame is not wide enough
+            frameImage = frameImage.resize((posterWidth, int(frameImage.size[1] * posterWidth / frameImage.size[0])), Image.ANTIALIAS)
+            crop = int((frameImage.size[1] - frameHeight) / 2)
+            frameImage = frameImage.crop((0, crop, posterWidth, crop + frameHeight))
+        posterImage.paste(frameImage, (0, 0, posterWidth, frameHeight))
+    else:
+        draw = ImageDraw.Draw(posterImage)
+        draw.polygon([(0, 0), (posterWidth, 0), (posterWidth, frameHeight), (0, frameHeight)], fill=(0, 0, 0))
+        for y in range(frameHeight):
+            for x in range(posterWidth):
+                if int((x + y + 54) / 128) % 2:
+                    posterImage.putpixel((x, y), (32, 32, 32))
+
+    # restricted
+    if opts.restricted:
+        for y in range(frameHeight):
+            for x in range(posterWidth):
+                if int((x + y + 54) / 128) % 2:
+                    rgb = posterImage.getpixel((x, y))
+                    rgb = (int(rgb[0] / 2) + 128, int(rgb[1] / 2), int(rgb[2] / 2))
+                    posterImage.putpixel((x, y), rgb)
+
+    # director section
+    colorHeight = int(posterHeight / 2);
+    draw = ImageDraw.Draw(posterImage)
+    draw.polygon([(0, frameHeight), (posterWidth, frameHeight), (posterWidth, colorHeight), (0, colorHeight)], fill=(0, 0, 0))
+    director = wrapString(opts.director, 36, '\n', True)
+    while len(director.split('\n')) > 3:
+        director = opts.director.split(', ')
+        director.pop()
+        opts.director = ', '.join(director)
+        director = wrapString(opts.director, 36, '\n', True)
+    posterMargin = 16
+    imagewrite(posterImage, director, posterMargin, colorHeight - 8 - len(director.split('\n')) * 32, fontImage, 32)
+
+    # title section
+    backgroundColor = getRGB(opts.oxdb)
+    draw.polygon([(0, colorHeight), (posterWidth, colorHeight), (posterWidth, posterHeight), (0, posterHeight)], fill=backgroundColor)
+    title = wrapString(opts.title, 24, '\n', True)
+    lines = title.split('\n')
+    if lines > 8:
+        # following line commented out since the only known case
+        # (http://0xdb.org/0071458) looks better without '...'
+        # lines[7] = truncateString(lines[7] + '...', 24)
+        title = '\n'.join(lines[:8])
+    offset = -6
+    posterimage = imagewrite(posterImage, title, posterMargin, colorHeight + posterMargin + offset, fontImage, 48)
+    offset = 12
+    posterimage = imagewrite(posterImage, id, posterMargin, posterHeight - posterMargin - 96 + offset, fontImage, 96)
+
+    # 0xdb logo
+    x = posterWidth - oxdbImage.size[0] - posterMargin
+    y = posterHeight - oxdbImage.size[1] - posterMargin
+    for dy in range(oxdbImage.size[1]):
+        for dx in range(oxdbImage.size[0]):
+            rgb = posterImage.getpixel((x + dx, y + dy))
+            bw = oxdbImage.getpixel((dx, dy))[0]
+            rgb = tuple(map(lambda x : x + bw, rgb))
+            posterImage.putpixel((x + dx, y + dy), rgb)
+
+    posterImage.save(opts.poster)
+
+def getRGB(oxid):
+    i = int(int(oxid[2:8], 16) * 762 / 16777216)
+    if i < 127:
+        return (127, i, 0)
+    elif i < 254:
+        return (254 - i, 127, 0)
+    elif i < 381:
+        return (0, 127, i - 254)
+    elif i < 508:
+        return (0, 508 - i, 127)
+    elif i < 635:
+        return (i - 508, 0, 127)
+    else:
+        return (127, 0, 762 - i)
+
+def imagewrite(posterImage, string, left, top, fontImage, charHeight):
+    x = left
+    y = top
+    fontWidth = int(fontImage.size[0] / 16)
+    fontHeight = int(fontImage.size[1] / 16)
+    charWidth = int(fontWidth * charHeight / fontHeight)
+    for i in range(len(string)):
+        char = string[i:i+1]
+        if char == '\n':
+            x = left
+            y += charHeight
+        else:
+            ascii = ord(char)
+            fontLeft = (ascii % 16) * fontWidth
+            fontTop = int(ascii / 16) * fontHeight
+            letterImage = fontImage.crop((fontLeft, fontTop, fontLeft + fontWidth, fontTop + fontHeight))
+            letterImage = letterImage.resize((charWidth, charHeight), Image.ANTIALIAS)
+            for dy in range(charHeight):
+                for dx in range(charWidth):
+                    rgb = posterImage.getpixel((x + dx, y + dy))
+                    bw = int(letterImage.getpixel((dx, dy))[0] / 2)
+                    rgb = tuple(map(lambda x : x + bw, rgb))
+                    posterImage.putpixel((x + dx, y + dy), rgb)
+            x += charWidth
+    return posterImage
+

setup.py

@@ -0,0 +1,34 @@
+# setup.py
+# -*- coding: UTF-8 -*-
+# vi:si:et:sw=4:sts=4:ts=4
+import sys
+from glob import glob
+from distutils.core import setup
+
+setup(name="oxtools",
+    scripts = [
+        'bin/oxicon',
+        'bin/oxinfo',
+        'bin/oxframe',
+        'bin/oxposter',
+        'bin/oxtimeline',
+    ],
+    packages = [
+        'oxgst',
+        'oxposter',
+    ],
+    package_data = {
+        'oxposter': ['data/*.png'],
+    },
+    version="0.1",
+    author="j",
+    author_email="code@0xdb.org",
+    description="commandline tools and python api to extract information from movies",
+    classifiers = [
+        'Development Status :: 4 - Beta',
+        'Operating System :: OS Independent',
+        'Programming Language :: Python',
+        'Topic :: Software Development :: Libraries :: Python Modules',
+    ],
+)
+