METS export

[eroux]

following https://github.com/BuddhistDigitalResourceCenter/library-issues/issues/151 : @TBRC-JimK would it be possible to produce a big zip with all the METS data you're producing? I don't know the works you're working on, but W22084 would be very useful to start with...

[xristy]

indeed W22084 has Hollis Id 014258613 so it will be one of the works being generated.

[jimk-bdrc]

@TBRC-Travis , is there some way you could make Windhorse /Volumes/DRS_Staging shareable over its Ethernet? If so, I think @eroux could find a rich source of sample METS data in any batch's `batchWxxxxx/batchWxxxx-Iyyyy/descriptor.xml

[jimk-bdrc]

And please do shoot me a list of works you need this for (they will all share one HOLLIS id) and I can generate lots of batches.

[eroux]

I can access RS3, but I don't know if "Windhorse" and "RS3" are the same... are they? If I can't get the whole zip easily, I think we can start with W22084, it will be enough for me to produce the code, then we can run the code on the whole thing later, no need for it right now!

[eroux]

Well, thinking about it, it would probably be smarter to get the METS for more Kangyurs and Tengyurs, as that's a big part of what we'll show in the first demo. These would be W1PD96685, W22084, W22703, W1PD96684, W26071, W1PD96682, W1PD95844, W1GS66030, W1KG13126, W23703 and W22704

[xristy]

just checking for the :workHollis on each Work will indicate the Hollis ID if any. Then Jim should be able to locate the ones that have already been generated; generate if a Hollis ID is available; or faux generate and then bundle up the METS info. The batches include the images which of course are not wanted, but it might do well to let everyone know that just the METS records are needed.

[eroux]

ok, I have to say the DRS is a bit mysterious to me, but am I correct if I understand you're suggesting I make a list of hollisID instead of RID? I'll do that tomorrow if that's the case, no problem

[xristy]

The batch builder is driven with a list of RID, Hollis ID pairs. I was simply indicating how to tell whether the Works of interest have a Hollis ID or not.

We can discuss the DRS in a chat.

[jimk-bdrc]

I should either have them built or may already have built them. I need about an hour to get to where I can check.

Jim Katz

On Feb 5, 2018, at 3:54 PM, Elie Roux notifications@github.com wrote:

Well, thinking about it, it would probably be smarter to get the METS for more Kangyurs and Tengyurs, as that's a big part of what we'll show in the first demo. These would be W1PD96685, W22084, W22703, W1PD96684, W26071, W1PD96682, W1PD95844, W1GS66030, W1KG13126, W23703 and W22704

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[jimk-bdrc]

Elie, the works you have listed are all I need - no need to search for the Hollis.

[TBRC-Travis]

Sent Elie a first round of METS data

[eroux]

well received thanks!

[jimk-bdrc]

Hi, I'm building these batches now, should be done by noon (Eastern US) latest. W1PD96685,14257985 W1PD96684,14257984 W26071,14259506 W1PD96682,14257983 W1PD95844,14257975 W1GS66030,14255521 W1KG13126,14256023 W23703,14259049

[jimk-bdrc]

An automated version of this is deployed since July 17. As make-drs-batch.sh builds batches, if the batch succeeds, the script calls tojsondimensions.py