Jinjurly Update: Help Wanted

[squidgestatus]

First of all, I'm happy to say that so far we have found (as of this moment) 2,337 of the 5,441 files that are missing (or 43%)!  This is amazing!  Hopefully by the weekend we can get to 50% or more!

The ssllllooooowww part of this is the processing of files.  Each one can take as few as 2 minutes, or as many as 15.  So this is definitely slow going - and the reason for my post.

If you are interested in helping out/becoming an archivist, great!  But we need a bit more than just an archivist at the moment; We need someone who is knowledgeable with using an FTP client to move files about.  So if you're interested, please let me know.  You can reach out publically in response to this post, but it'd be better if you emailed mific and I at: jinjurlymods@squidge.org

Thank you in advance if you're interested.  And thank you to everyone who has helped get us this far!

edit: Here's the process that someone has to go through, and thus, here's what I'm looking for:

• Locate a tan colored line in the spreadsheet where column B is unchecked.  Grab the URL from column A and open that entry in the Jinjurly dtabase.
• Back in the spreadsheet, go to column K and identify what has been found (an MP3, an M4B, a zip file, or combination). Download the file(s) to your local computer.
• Edit the entry in the Jinjurly database to identify the file name(s) that are there.
• Rename the downloaded files, and if appropriate, zip the MP3.
• Change the MP3 entry in the Jinjurly database to say "downloadable mp3", then copy the line and change "downloadable" to "streamable" and in the file name, change ".zip" to ".mp3" for the streamable entry.
• SFTP the downloaded files to the appropriate directory on the Jinjurly host.
• Edit any other necessary details of the Jinjurly entry, then save the edited Jinjurly entry.
• Test each MP3/ZIP/M4B entry in the edited Jinjurly entry to make sure the streaming one streams, and the zip file(s) download.
• Tick the checkbox in column B for the item you just restored and tested.
• Repeat

Comments

[morgandawn]

I boosted this
https://meeedeee.tumblr.com/post/674586720241680384/help-rescue-podfic-history-ftp-client-skills-the

[squidgestatus]

Thank you! I just reblogged your post!

[xancredible]

I already emailed about helping out, just wanted to say that according to the Stats tab on the spreadsheet, we cracked 50% :)

Thank you!

[squidgestatus]

And I don’t know if you saw, but I was counting hidden lines - so we’re actually close to 60%! We’re at 5,075 missing lines instead of 5,441.

Conversion from m4b to mp3

[brickgrass]

This seemed like the best place for me to dump this info, so here we go:

During the restoration of files, some of the rows on the spreadsheet only have m4b files available, so obviously the archivist restoring that file will need to convert from m4b -> mp3, using a program such as ffmpeg. (Infact almost certainly ffmpeg, most other converters I can think of are just using it, under the hood.) Just one problem! With default settings, a lot of quality can get lost in this conversion, which we obviously don't want. The solution for this is the extra command arguments that can be found in this stackoverflow solution: https://superuser.com/a/704535. (You may notice that the question is specifically asking about m4as, but as the similar name suggests, m4a and m4b files are effectively the same thing.)

So you don't have to click the link, the tldr is you want this ffmpeg -i input.m4a -c:v copy -c:a libmp3lame -q:a 4 output.mp3 because it'll do it's best to mimic the bitrate of the original file.

iirc the default windows command line ffmpeg is a little more restricted than the one I have available on mac. Apologies, but I don't have a windows PC to test on to be sure it works. So if you're on windows have a go with that command, and if it doesn't work then have a go at using it inside the WSL.