How do I record these podcasts

CrashedAgain · 03-31-2010, 08:47 PM

How do I record the podcasts from this source http://www.cbc.ca/vinyltap/audio.html ?
I have tried mplayer I get a "stream not seekable" error.
I also tried arecord which I thought was supposed to record whatever sound is coming through the computer.
I tried to use the "sound recorder" that comes with Ubuntu but that records only from the microphone (source is "capture" with no other options.
Using Ubuntu 9.04 on an Acer Aspire one netbook.....
Having even less success with Ubuntu 9.04 on my Acer desktop computer.

itsbrad212 · 03-31-2010, 09:19 PM

What microphone do you use?

Kenhelm · 03-31-2010, 09:58 PM

You don't need to record the podcasts as their URLs are in the webpage.
The URLs can be used to download the podcasts as mp3 files.

This strips the 110 podcast URLs from the webpage:

Code:

wget -qO- http://www.cbc.ca/vinyltap/audio.html |
grep -o 'http://podcast[^"]*\.mp3'

# Output:
http://podcast.cbc.ca/mp3/vinyltapstreaming_20100403_29816.mp3
http://podcast.cbc.ca/mp3/vinyltapstreaming_20100403_29823.mp3
http://podcast.cbc.ca/mp3/vinyltapstreaming_20100327_29811.mp3
....
....
....
http://podcast.cbc.ca/mp3/vinyltapstreaming_20081004_8059.mp3
http://podcast.cbc.ca/mp3/vinyltapstreaming_20080928_8003.mp3
http://podcast.cbc.ca/mp3/vinyltapstreaming_20080928_8015.mp3

CrashedAgain · 04-01-2010, 09:39 AM

Thank you very much.

Next questions....

What is the best way to copy these mp3's? Mplayer works but is there a better way?

Is there a way to copy all of the files without having to d/l each one individually? (I've done some very elementary scripting but I'm not very familiar with it)

Kenhelm · 04-01-2010, 11:13 PM

Try downloading the podcast file into a directory then use 'mp3splt' to split it into smaller files.
For example, if you paste this command into a terminal it will download the podcast file into the terminal's present working directory.
The name of the file will be 'vinyltapstreaming_20100403_29816.mp3'

Code:

wget http://podcast.cbc.ca/mp3/vinyltapstreaming_20100403_29816.mp3

The file can also be taken from the Firefox 'Cache' directory.
The mp3 file is left there after the podcast has been played in Firefox.
But, if the Firefox cache size has been set too small to hold the mp3 file, it could be in the /tmp directory instead and in that case will probably be deleted after it has finished playing in Firefox.
Look for a file of around 24MB.

There are two versions of mp3splt:
mp3splt is a command line tool.
mp3splt-gtk has a graphical user interface.
Both are able to split up a mp3 file into several files without any loss of sound quality.
Screenshot of mp3splt-gtk: http://en.wikipedia.org/wiki/File:Mp3splt-gtk-0.3.png

CrashedAgain · 04-04-2010, 04:54 PM

Kenhelm....you misunderstand what I want. The problem is not that the files are too big, the question is how to download all 110 of them without having to do each download separately.

I made up this script which seems to be working

Code:

#!/bin/bash

#echo "Echo"
#sample http://podcast.cbc.ca/mp3/vinyltapstreaming_20081004_8062.mp3
Inlist=VinylTapList
COUNT=0

#cd to a dir which has lots of room
cd /media/sdb2/VinylTap
#get the files list
wget -qO- http://www.cbc.ca/vinyltap/audio.html | grep -o 'http://podcast[^"]*\.mp3' > VinylTapList

#cat $Inlist

for i in `cat $Inlist`; do
let COUNT=COUNT+1
echo "Downloading $COUNT of 110"
	
	DumpFileName=`expr substr $i 27 40` 
#	echo $DumpFileName
#use if to avoid duplicating downloaded files
	if [ -a $DumpFileName ]
  		then
    		echo "$DumpFileName exists."
		else 
		echo "$DumpFileName does not exist"
		
		wget $i 
	fi


#...was going to use mplayer 
	#mplayer -dumpstream $i -novideo -dumpfile "$DumpFileName"

done

It appears to take about 45 seconds per file so the entire job should take just under 2 hours.

It would have been a lot easier if I could just do wget http://podcast.cbc.ca/mp3/vinyltapstreaming_*.* but wget doesn't allow wildcards in http downloads. Also there may be a "skip existing files" option in wget but I couldn't find it, thus had to use the "if" construct.

You also mentioned trying to recover the file from Firefox's cache as a downloading method.

I have had some difficulty with using this method recently, I don't think Flash 10 deposits files in the cache and then leaves them there like previous versions of Flash did. Anyone else encountering this??