-
Notifications
You must be signed in to change notification settings - Fork 46
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Feature Request] allow for merge
Tileset utility to only produce root tileset json and avoid tile contents copy
#44
Comments
This certainly could make sense for some cases. We also noticed that people occasionally want to "merge" tilesets, but expect the result to be what could be achieved by first calling (And this should yield the right result in your case, if the input tilesets are "upgraded" to use There are some corner cases to consider here: Right now, most of the functionality of the
and receive a But there probably are some possible solutions for that:
|
Hi Marco, thanks a lot for the detailed feedback! Understood for the current requirement to support input/output tilesets which would either be read/written to the filesystem or self-sustained, containerized archive formats - hence the need to copy over the tile contents as well. In that case, the first For ref, we have produced a few hundreds 3D-scans (heritage sites, monuments and cities) at Iconem over the years, each consisting of poly-counts from a few hundreds of millions to a few billions primitives (polygons/points), largest being a complete multi-resolution Venice at 15B points. |
Would you be open to a PR handling that use case? Side-question: I have a tileset I created using Thanks for your help! |
Yes, sorry that there has not been any progress here. It could be a relatively low-hanging fruit in terms of the implementation effort.
A bit of context: As mentioned above, the operations like
Certainly. There are not really any detailed instructions for contributors, but smaller things (like formatting or updating the change log) could be done separately. In any case, you will have to sign the CLA at https://github.com/CesiumGS/cesium/blob/main/CONTRIBUTING.md#contributor-license-agreement-cla Based on the existing structure, your suggestion of a
I'm somehow slightly leaning towards the latter (roughly, favoring "self-contained" commands over ones that have (eventually 'many') parameters for configuration). But I don't have a strong opinion. What do you think? (BTW: If you don't want to take this up, I can allocate a bit of time for implementing that in the next days) |
Thanks for that feedback, very much appreciated! |
From quickly skimming over the changes, the description (at the CLI and the params) seems to be reversed...
It says whether to not copy the resources (when it is |
Indeed, thanks. I went ahead and followed your initial suggestion, splitting I also tried forcing line-endings to Hope this helps, and again thanks for the directions! (+ I can revert to the having the {
"endOfLine": "lf"
} |
Fixed via #140 , thanks @jo-chemla ! |
Hi there, just made a new PR since I found a better way to handle relative paths from output directory to each input tileset filename, so output merged tileset.json produced via mergeJson can be consumed directly in place. Here is the link to the PR: #142 ! |
Using the
merge
utility as presented in the docs starts by producing an outputtileset.json
root file which references the input tilesets in children directories. Then it continues by copying all the tile contents to the output folder.Since there are lots of files to copy, this copy operation can take a lot of time. Having an option to
disable tiles copying
(json + b3dm) would be great! Therefore, the output tileset would only reference the input tilesets - probably via auri
relative to the output tileset file. eg in the case above, the output tileset would look likeNote: this might be what the
combine
utility is made for, but since it does not process in my case because of aContent does not have a URI
error (properties areurl
), I could not check, see this feature request #43The text was updated successfully, but these errors were encountered: