Recherche avancée

Sur d’autres sites (7506)

• Paperclip Upload Video Without Styles

  16 mars 2017, par Moamen Naanou

  Background :

  In my app, there is a model has image attachment using paperclip gem, now as per new requirements, I have to let the same attachment accept MP4 video files as well and the only validation I have to implement is about file size.

  Product Model (with image attachment only) :

  class Product < ActiveRecord::Base
  
    has_attached_file :file,
  
                      :styles => { medium: '300x300>', thumb: '100x100>' },
  
                      :processors => [:thumbnail],
  
                      :url => '/assets/:id/:style/:hash.:extension',
  
                      :hash_digest=>'SHA1',
  
                      :use_timestamp => false
  
  
  
  
  
    validates_attachment_content_type :file, :content_type => /\A(image)\/.*\Z/
  
    validates_attachment_size :file, :less_than => 5.megabytes
  
  end

  So after reading many related questions, to accept a video file as well, I’ve used gem paperclip-av-transcoder and installed ffmpeg (Mac) using :

  brew install ffmpeg --with-fdk-aac --with-ffplay --with-freetype --with-frei0r --with-libas

  Then finally I’ve revised my model to be :

  class Product < ActiveRecord::Base
  
    has_attached_file :file,
  
                      :styles => lambda {|file| if file.instance.is_image?
  
                                                  { medium: '300x300>', thumb: '100x100>' }
  
                                                elsif file.instance.is_video?
  
                                                  {:thumb => { :geometry => "100x100#", :format => 'jpg', :time => 10}}
  
                                                end },
  
                      :processors => lambda {|file| if file.is_image?
  
                                                      [:thumbnail]
  
                                                    elsif file.is_video?
  
                                                      [:transcoder]
  
                                                    end},
  
                      :url => '/assets/:id/:style/:hash.:extension',
  
                      :hash_digest=>'SHA1',
  
                      :use_timestamp => false
  
  
  
  
  
    validates_attachment_content_type :file, :content_type => /\A(image|video)\/.*\Z/
  
    validates_attachment_size :file, :less_than => 5.megabytes
  
  
  
    def is_video?
  
      file_content_type =~ %r(video)
  
    end
  
  
  
    def is_image?
  
      file_content_type =~ %r(image)
  
    end
  
  end

  Now the file field accepts both videos & images but the video shouldn’t be compressed, because if it is compressed, I will get IO Error [Closed Stream]

  Question :

  I’m expecting compressed MP4 file to be received from Mobile client(through REST API) and I don’t want to generate any thumb image or any other style or do any processing on this video and without installing any processor (ffmpeg in my case) and if I did so currently, I’m getting the following error because ImageMagic doesn’t work with video files :

  Paperclip::Errors::NotIdentifiedByImageMagickError

  Any ideas ?

• QTableWidget and QProcess - update table based on multiple process results

  9 mars 2017, par Spencer

  I have a python program that runs through a QTableWidget and for each item it runs a QProcess (an FFMPEG process to be exact). What I’m trying to do is update the "parent" cell when the process completes. Right now there is a for loop that goes through each row and launches a process for each, and connects the finished signal of that process to a "finished" function, which updates the QTableWidget cell. I’m just having trouble properly telling the function WHICH sell to update - right now I am passing it the index of the current row (seeing as it is being spawned by the for loop) but what happens is by the time the processes start to finish it will only get the last row in the table... I’m quite new to Python and PyQt so it is possible there is some fundamental thing I have wrong here !

  I tried passing the actual QTabelWidgetItem instead of the index but I got this error : "RuntimeError : wrapped C/C++ object of type QTableWidgetItem has been deleted"

  My code, the function "finished" and line #132 are the relevant ones :

  import sys, os, re
  
  from PyQt4 import QtGui, QtCore
  
  
  
  class BatchTable(QtGui.QTableWidget):
  
      def __init__(self, parent):
  
          super(BatchTable, self).__init__(parent)
  
          self.setAcceptDrops(True)
  
          self.setColumnCount(4)
  
          self.setColumnWidth(1,50)
  
          self.hideColumn(3)
  
          self.horizontalHeader().setStretchLastSection(True)
  
          self.setHorizontalHeaderLabels(QtCore.QString("Status;Alpha;File;Full Path").split(";"))
  
  
  
          self.doubleClicked.connect(self.removeProject)
  
  
  
      def removeProject(self, myItem):
  
          row = myItem.row()
  
          self.removeRow(row)
  
  
  
      def dragEnterEvent(self, e):
  
          if e.mimeData().hasFormat('text/uri-list'):
  
              e.accept()
  
          else:
  
              print "nope"
  
              e.ignore()
  
  
  
      def dragMoveEvent(self, e):
  
          e.accept()
  
  
  
      def dropEvent(self, e):
  
          if e.mimeData().hasUrls:
  
              for url in e.mimeData().urls():
  
                  chkBoxItem = QtGui.QTableWidgetItem()
  
                  chkBoxItem.setFlags(QtCore.Qt.ItemIsUserCheckable | QtCore.Qt.ItemIsEnabled)
  
                  chkBoxItem.setCheckState(QtCore.Qt.Unchecked)
  
  
  
                  rowPosition = self.rowCount()
  
                  self.insertRow(rowPosition)
  
                  self.setItem(rowPosition, 0, QtGui.QTableWidgetItem("Ready"))
  
                  self.setItem(rowPosition, 1, chkBoxItem)
  
                  self.setItem(rowPosition, 2, QtGui.QTableWidgetItem(os.path.split(str(url.toLocalFile()))[1]))
  
                  self.setItem(rowPosition, 3, QtGui.QTableWidgetItem(url.toLocalFile()))
  
                  self.item(rowPosition, 0).setBackgroundColor(QtGui.QColor(80, 180, 30))
  
  
  
  class ffmpegBatch(QtGui.QWidget):
  
      def __init__(self):
  
          super(ffmpegBatch, self).__init__()
  
          self.initUI()
  
  
  
      def initUI(self):
  
  
  
          self.edit = QtGui.QTextEdit()
  
  
  
          cmdGroup = QtGui.QGroupBox("Commandline arguments")
  
          fpsLbl = QtGui.QLabel("FPS:")
  
          self.fpsCombo = QtGui.QComboBox()
  
          self.fpsCombo.addItem("29.97")
  
          self.fpsCombo.addItem("23.976")
  
          hbox1 = QtGui.QHBoxLayout()
  
          hbox1.addWidget(fpsLbl)
  
          hbox1.addWidget(self.fpsCombo)
  
          cmdGroup.setLayout(hbox1)
  
  
  
          saveGroup = QtGui.QGroupBox("Output")
  
          self.outputLocation = QtGui.QLineEdit()
  
          self.browseBtn = QtGui.QPushButton("Browse")
  
          saveLocationBox = QtGui.QHBoxLayout()
  
          # Todo: add "auto-step up two folders" button
  
          saveLocationBox.addWidget(self.outputLocation)
  
          saveLocationBox.addWidget(self.browseBtn)
  
          saveGroup.setLayout(saveLocationBox)
  
  
  
          runBtn = QtGui.QPushButton("Run Batch Transcode")
  
  
  
          mainBox = QtGui.QVBoxLayout()
  
          self.table = BatchTable(self)
  
          # TODO: add "copy from clipboard" feature
  
          mainBox.addWidget(self.table)
  
          mainBox.addWidget(cmdGroup)
  
          mainBox.addWidget(saveGroup)
  
          mainBox.addWidget(runBtn)
  
          mainBox.addWidget(self.edit)
  
  
  
          self.setLayout(mainBox)
  
          self.setGeometry(300, 300, 600, 500)
  
          self.setWindowTitle('FFMPEG Batch Converter')
  
  
  
          # triggers/events
  
          runBtn.clicked.connect(self.run)
  
  
  
      def RepresentsInt(self, s):
  
          try:
  
              int(s)
  
              return True
  
          except ValueError:
  
              return False
  
  
  
      def run(self):
  
          if (self.outputLocation.text() == ''):
  
              return
  
          for projIndex in range(self.table.rowCount()):
  
              # collect some data
  
              ffmpeg_app = "C:\\Program Files\\ffmpeg-20150702-git-03b2b40-win64-static\\bin\\ffmpeg"
  
              frameRate = self.fpsCombo.currentText()
  
              inputFile = self.table.model().index(projIndex,3).data().toString()
  
              outputPath = self.outputLocation.text()
  
              outputPath = outputPath.replace("/", "\\")
  
  
  
              # format the input for ffmpeg
  
              # find how the exact number range, stored as 'd'
  
              imageName = os.path.split(str(inputFile))[1]
  
              imageName, imageExt = os.path.splitext(imageName)
  
              length = len(imageName)
  
              d = 0
  
              while (self.RepresentsInt(imageName[length-2:length-1]) == True):
  
                  length = length-1
  
                  d = d+1
  
              inputPath = os.path.split(str(inputFile))[0]
  
              inputFile = imageName[0:length-1]
  
              inputFile = inputPath + "/" + inputFile + "%" + str(d+1) + "d" + imageExt
  
              inputFile = inputFile.replace("/", "\\")
  
  
  
              # format the output
  
              outputFile = outputPath + "\\" + imageName[0:length-2] + ".mov"
  
  
  
  
  
              # build the commandline
  
              cmd = '"' + ffmpeg_app + '"' + ' -y -r ' + frameRate + ' -i ' + '"' + inputFile + '"' + ' -vcodec dnxhd -b:v 145M -vf colormatrix=bt601:bt709 -flags +ildct ' + '"' + outputFile + '"'
  
  
  
              # launch the process
  
              proc = QtCore.QProcess(self)
  
              proc.finished.connect(lambda: self.finished(projIndex))
  
              proc.setProcessChannelMode(proc.MergedChannels)
  
              proc.start(cmd)
  
              proc.readyReadStandardOutput.connect(lambda: self.readStdOutput(proc, projIndex, 100))
  
              self.table.setItem(projIndex, 0, QtGui.QTableWidgetItem("Running..."))
  
              self.table.item(projIndex, 0).setBackgroundColor(QtGui.QColor(110, 145, 30))
  
  
  
      def readStdOutput(self, proc, projIndex, total):
  
          currentLine = QtCore.QString(proc.readAllStandardOutput())
  
          currentLine = str(currentLine)
  
          frameEnd = currentLine.find("fps", 0, 15)
  
          if frameEnd != -1:
  
              m = re.search("\d", currentLine)
  
              if m:
  
                  frame = currentLine[m.start():frameEnd]
  
                  percent = (float(frame)/total)*100
  
                  print "Percent: " + str(percent)
  
                  self.edit.append(str(percent))
  
                  self.table.setItem(projIndex, 0, QtGui.QTableWidgetItem("Encoded: " + str(percent) + "%"))
  
  
  
      def finished(self, projIndex):
  
          # TODO: This isn't totally working properly for multiple processes (seems to get confused)
  
          print "A process completed"
  
          print self.sender().readAllStandardOutput()
  
          if self.sender().exitStatus() == 0:
  
              self.table.setItem(projIndex, 0, QtGui.QTableWidgetItem("Encoded"))
  
              self.table.item(projIndex, 0).setBackgroundColor(QtGui.QColor(45, 145, 240))
  
  
  
  
  
  def main():
  
      app = QtGui.QApplication(sys.argv)
  
      ex = ffmpegBatch()
  
      ex.show()
  
      sys.exit(app.exec_())
  
  
  
  if __name__ == '__main__':
  
      main()

  (And yes I do know that my percentage update is totally wrong right now, still working on that...)

• What is AVHWAccel, and how can I use it ?

  15 mars 2017, par shintaroid

  I want to make use of hardware acceleration for decoding an h264 encoded MP4 file.

  My computing environment :

  Hardware: MacPro (2015 model)
  
  Software: FFmpeg (installed by brew)

  Here is the output of FFmpeg command :

  $ffmpeg -hwaccels
  
  Hardware acceleration methods:
  
  vda 
  
  videotoolbox

  According to this document, there are two options for my environment, that is, VDA and VideoToolBox. I tried VDA in C++ :

  Codec = avcodec_find_decoder_by_name("h264_vda");

  It kind of worked, but the output of the pixel format is UYVY422 which I have trouble to deal with (any suggestion on how to render UYVY422 in C++ ? The ideal format is yuv420p)

  So I want to try VideotoolBox, but there is no such simple thing like (it may work in the case of encoding though)

  Codec = avcodec_find_decoder_by_name("h264_videotoolbox");

  It seems I should use AVHWAccel, but what is AVHWAccel and how to use it ?

  Part of My C++ code :

  for( unsigned int i = 0; i < pFormatCtx->nb_streams; i++ ){
  
          if(pFormatCtx->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO){
  
              pCodecCtx = pFormatCtx->streams[i]->codec;
  
              video_stream = pFormatCtx->streams[i];
  
              if( pCodecCtx->codec_id == AV_CODEC_ID_H264 ){
  
                  //pCodec = avcodec_find_decoder(pCodecCtx->codec_id);
  
                  pCodec = avcodec_find_decoder_by_name("h264_vda");
  
                  break;
  
              }
  
          }
  
      }
  
      // open codec
  
      if( pCodec ){
  
          if((ret=avcodec_open2(pCodecCtx, pCodec, NULL)) < 0) {
  
          ....