Re: Is it a repeat?

2013-06-28 Thread Vangelis forthnet

On Fri Jun 28 08:26:21 BST 2013, Chris Marriott wrote:

... although you can't use the PID to tell with any 
degree of accuracy WHEN a programme was first broadcast, ...


Hi Chris!
Have you read my previous post here:

http://lists.infradead.org/pipermail/get_iplayer/2013-June/004383.html

The PID alone (the 8-digit alphanumeric string) can't tell much by itself 
(as you say, it is only when you compare it to another PID that you can 
- maybe - deduce which one belongs to a more recent broadcast), but 
the URL:


http://www.bbc.co.uk/programmes/PID/broadcasts

does reveal precisely when the first broadcast took place...

@dinkypumpkin:

Any news whether the above URL could be somehow utilised by 
get_iplayer?


Many thanks in advance :-)
Vangelis.   


___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: Massive Failure of Radio Download

2013-06-28 Thread Roger Bell_West
On Fri, Jun 28, 2013 at 12:20:19PM -0700, Chris J Brady wrote:
>Any ideas to get it working again? CJB

Use get_iplayer instead, or ask on a list that isn't dedicated to
get_iplayer?

___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Massive Failure of Radio Download

2013-06-28 Thread Chris J Brady
Any ideas to get it working again? CJB

___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: Cache deleted

2013-06-28 Thread dinkypumpkin

On 28/06/2013 15:27, Arthur Murray wrote:

I don't know how it would affect others. If one channel like BBC One
doesn't refresh then I still wouldn't want my cache purged of all BBC
One pids only to have them all show up as new again at the next
successful refresh (which is done several times per day on my system).
  It is very frustrating to see old programs as new in the PVR cgi and
then have to try to pick the new ones out, so I'll use the patch
locally.


If you change your mind, submit something.  To make it work the way you 
want - which seems like the right way to do it - would require a little 
extra work, I think.  Instead of halting a refresh on the failure of any 
single channel to load, you could collect a list of the failed channels 
and copy all the entries for those channels from old cache to new cache, 
thus retaining as much current data possible and preserving timestamps 
until the problem channels can be refreshed.



___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: Add Series from cgi fails.

2013-06-28 Thread Arthur Murray
On Fri, Jun 28, 2013 at 4:58 AM, dinkypumpkin  wrote:
> This patch escapes every non-alphanumeric character in the programme name
> and then removes the escaping from whitespace characters for readability.
> The question mark is probably the only problematic character likely to
> appear in programme names, but we might as well make the escaping generic.
> If this change works for you I'll merge it into get_iplayer.

Yes the patch works for me, thanks.

___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: Cache deleted

2013-06-28 Thread Arthur Murray
On Tue, Jun 25, 2013 at 1:05 PM, dinkypumpkin  wrote:

> I'm not 100% convinced about your solution.  You're suggesting that the
> entire cache rebuild should be halted if even one channel doesn't refresh.
> That's something that happens from time to time, so a single failed channel
> could prevent your cache refreshing for a day or more. But then, what should
> the threshold be?  I guess it's arbitrary, so I'll leave it up to you.  Put
> whatever you think best into a patch or pull request we'll go with that.

I don't know how it would affect others. If one channel like BBC One
doesn't refresh then I still wouldn't want my cache purged of all BBC
One pids only to have them all show up as new again at the next
successful refresh (which is done several times per day on my system).
 It is very frustrating to see old programs as new in the PVR cgi and
then have to try to pick the new ones out, so I'll use the patch
locally.

___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: Add Series from cgi fails.

2013-06-28 Thread dinkypumpkin

On 28/06/2013 00:18, Arthur Murray wrote:

Using "Add Series" from the get_iplayer.cgi on a prog name with a
question mark in it ("Who Were the Greeks?") fills in the "search0"
field of the pvr file with the question mark and therefore does not
match. (search0 ^Who Were the Greeks?$).  If the question mark is
escaped it will match (search0 ^Who Were the Greeks\?$).  What is the
correct place to correct this, in the cgi, or in get_iplayer and
specifically where?

(Encoding the question mark to ? doesn't work, only escaping it does.)

# from cgi
# Add Series
$links .= label( {
 -id=>'nowrap',
 -class=>'search pointer_noul',
 -title=>"Add Series '$prog{$pid}->{name}' to PVR",
 -onClick=>"BackupFormVars(form);
form.NEXTPAGE.value='pvr_add';
form.SEARCH.value='".encode_entities("^$prog{$pid}->{name}\$")."'; ...


Sample of ugly fixing in cgi:

  my $encoded = encode_entities("^$prog{$pid}->{name}\$");
  $encoded =~ s/\?/\\\?/;

... form.SEARCH.value='$encoded';  ...


You're in the right place.  Try this patch:

https://github.com/dinkypumpkin/get_iplayer/commit/e43249cb3e857fffed07601d67c190d641076be8

You can apply it directly from:

https://github.com/dinkypumpkin/get_iplayer/commit/e43249cb3e857fffed07601d67c190d641076be8.patch

This patch escapes every non-alphanumeric character in the programme 
name and then removes the escaping from whitespace characters for 
readability.  The question mark is probably the only problematic 
character likely to appear in programme names, but we might as well make 
the escaping generic.  If this change works for you I'll merge it into 
get_iplayer.




___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: rtmpdump version error

2013-06-28 Thread dinkypumpkin

On 28/06/2013 00:45, dinkypumpkin wrote:

The bottom line is that if PATH doesn't contain %SystemRoot\system32, or
is malformed so that directory is effectively missing, then rtmpdump
won't run, leading to the get_iplayer errors.


Er, make that %SystemRoot%\system32



___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer


Re: Is it a repeat?

2013-06-28 Thread Chris Marriott



-Original Message- 
From: dinkypumpkin

Sent: Thursday, June 27, 2013 7:19 PM
To: get_iplayer
Subject: Re: Is it a repeat?


On 27/06/2013 12:52, Ian macdonald wrote:

Does this mean the PID stays the same, regardless of how many times
the program is repeated?


I think so. Even after they roll off iPlayer, programmes are still 
catalogued in the episode lists their series' sites, so I presume that 
identifiers are meant to be long-lived.


Yes, the PID for a particular programme sticks with it for repeats. I have 
something like 3 years' download history on my PC, and often get told that 
I've previously downloaded a programme when it's a repeat within that 
timeframe.


That's why I say that, although you can't use the PID to tell with any 
degree of accuracy WHEN a programme was first broadcast, you can often 
identify from the PID that a program IS a repeat. All current programmes 
have a PID starting in "B03x", so if you find a program with a "B01" or 
"B00" PID you know for certain that it's a repeat, and a "B02" PID, although 
not 100% definitely a repeat, very likely is. (The occasional "Pxxx" PIDs 
don't fit into this pattern, of course).


I don't know if this is of any practical use to anyone; it's simply what 
I've observed, given that I always download via PIDs.


Chris



___
get_iplayer mailing list
get_iplayer@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/get_iplayer