For several years I've been enqueueing whole "brands" for radio programmes to be recorded with a single command like:
get_iplayer --pvr-queue --type radio --pid=b0072q60 --pid-recursive and then downloading new episodes automatically in the middle of the night with: get_iplayer --pvr --metadata=generic I noticed that the files in the .get_iplayer/pvr/ directory were all named ONCE_<pid>, yet that didn't seem to matter. I never knowingly needed to enqueue the "brand" again for more downloads to occur in the future once new programmes had been broadcast. This all worked well until I upgraded from get_iplayer v2.99 to v3.00. It seems that the PVR ONCE_<pid> files are now being deleted once any programme has been downloaded that matches the brand. v3.01 seems to do the same. I've read through the release notes and I can't see this change mentioned. Looking at the code in $pvr->run() I can't see why these ONCE_ files weren't being deleted before. Perhaps something about the download was always failing previously, so $failcount was always non-zero for me? Perhaps the new scraping code means that far few programmes are being considered so it is possible to have downloaded them all, whereas in the past there were always programmes that were too old that could not be downloaded? It seems that $pvr->run() won't delete files that don't match /^ONCE_/ , but $pvr->queue doesn't provide a way to create files with any other name. Is there a better way to enqueue series and brand PIDs so that they continue to work indefinitely? Thanks. Mike. _______________________________________________ get_iplayer mailing list [email protected] http://lists.infradead.org/mailman/listinfo/get_iplayer
