search

diamm / diamm

The Digital Image Archive of Medieval Music
https://www.diamm.ac.uk
8 stars 1 forks source link

Potential BIG issue with image viewer #493

Open juliacmcf opened 4 years ago

juliacmcf commented 4 years ago

We had some serious issues with E-Bbc MS M. 971c which I think may be occurring in other MSS.

  1. Although this MS is a single leaf (r and v) we were seeing random UV images in place of the verso and not the main image. This seemed to be because the verso had about 15 UV secondary images attached to it. When I turned off the 'public' view of the secondary images this seemed to work; this could be an issue with every MS where I have put in the secondary images on the page record.
  2. In this MS we keep getting the image for 1v showing up in place of 1r (so there are two images in the viewer, but both of 1v). There are no secondary images for 1r in the page record. We re-fetched the images and it showed correctly, but then went wrong again and will only show correctly immediately after a re-fetch.
  3. because the UV images are essential to reading the MS, I created FOUR further 'pages' (because the UV images were shot as top half and bottom half for each side of the leaf). At one point I managed to get these to show correctly (after several re-fetches), but having gone back to the MS just now, only 5 of the 6 images are showing, and they are not the correct images (so the image for the verso appears twice (there is no image for the recto) and the top half of the recto UV appears twice).

I'm a bit reluctant to delete all the work someone did adding the secondary images to fo. 1v to see if that will fix the problem, but as other pages that don't have secondary images are also having problems I'm not sure that's the issue. Ms M. 971 has a similar problem, where (at the moment) a picture of the cover is being displayed instead of f. 3v.

juliacmcf commented 4 years ago

Actually this may be more widespread (and therefore not linked to particular types of images/MSS). I've just created pages and linked images to them for the Bent Choirbook. At first it showed us images of E-G 33/II. Then I refetched and it was OK, EXCEPT the second last image, which was replaced with another image from E-G 33/II (we have been adding images to the virtual reconstruction of that MS as well (Girona-Catalunya fragments) which may be upsetting it). I'm hoping this is just a temporary glitch linked to us working with the image links, but the original error was there before we did any tinkering with images.