Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

IIIF - possible wrong calculation of tile width

See original GitHub issue

I have this image served by IIPImage server:

{
"@context": "http://iiif.io/api/image/2/context.json",
"protocol": "http://iiif.io/api/image",
"width": 2609,
"height": 3424,
"sizes": [
{
"width": 163,
"height": 214
},
{
"width": 326,
"height": 428
},
{
"width": 652,
"height": 856
},
{
"width": 1304,
"height": 1712
}
],
"tiles": [
{
"width": 256,
"height": 256,
"scaleFactors": [
1,
2,
4,
8,
16
]
}
],
"@id": "http://iip.test/iiif/f.jp2",
"profile": [
"http://iiif.io/api/image/2/level2.json",
{
"formats": [
"jpg",
"png",
"webp"
],
"qualities": [
"native",
"color",
"gray",
"bitonal"
],
"supports": [
"regionByPct",
"regionSquare",
"sizeByForcedWh",
"sizeByWh",
"sizeAboveFull",
"sizeUpscaling",
"rotationBy90s",
"mirroring"
],
"maxWidth": 3500,
"maxHeight": 3500
}
]
}

When used with Openseadragon viewer I get these IIIF API calls:

First row:

http://iip.test/iiif/f.jp2/0,0,2048,2048/256,/0/default.jpg => 256 x 256
http://iip.test/iiif/f.jp2/2048,0,561,2048/71,/0/default.jpg => 71 x 259

Second row:

http://iip.test/iiif/f.jp2/0,2048,2048,1376/256,/0/default.jpg => 256 x 172
http://iip.test/iiif/f.jp2/2048,2048,561,1376/71,/0/default.jpg => 71 x 174

The content of the last tile column appears dropped a bit when zoomig out or on a small displays.

iip_iiif_osd

As @ruven kindly pointed out at https://github.com/ruven/iipsrv/issues/237#issuecomment-1235444340 it looks like the width parameter seems probably miscalculated resulting in bad tile height.

Issue Analytics

State:
Created a year ago
Comments:12 (11 by maintainers)

Top GitHub Comments

1reaction

azaroth42commented, Sep 7, 2022

Could you try modifying your local copy for these lines: https://github.com/openseadragon/openseadragon/blob/master/src/iiiftilesource.js#L401-L402 to use floor instead of ceil, and see if it works as expected?

(2609 - 2048) / 8 is 70.125, and the ceil makes it a width of 71. The top left of the region is correct in the second column, as is the width (561) and height (2048). So the only other parameter to manipulate is the desired size which has to be a whole number of pixels. Asking for a width of 71, when the number of pixels is 561 is 0.12656, times the height of 2048 is 259, giving you the three undesirable pixels.

That said, this has been working for a long time now so I’m surprised it hasn’t come up before

0reactions

iangilmancommented, Dec 7, 2022

Fixed by #2206.

Top Results From Across the Web

IIIF - last (non-square) tile in each row has incorrect (bigger) height ...

It's using a "tile" width of 2048 for the first IIIF request (left tile on the full-size image) and if we calculate the...

Image API 2.1.1 - IIIF

Each description gives a width, optionally a height for non-square tiles, and a set of scale factors at which tiles of those dimensions...

Wrong tile resolution when using OpenLayers to display local ...

source.IIIF({ size: [config.map.width, config.map.height], extent: extent, projection: projection, tileSize: tileSize, resolutions: [1, 2, 4], }); ...

Tiling Problem - GeeksforGeeks

Given a “2 x n” board and tiles of size “2 x 1”, count the number of ... Please write comments if you...

Defining Normal and Abnormal Fetal Growth - NCBI - NIH

In summary, an ideal definition of FGR should take into account the growth potential of the fetus, current fetal size, fetal and placental...