reading foss4g presentation proposals, there is one about Brazilian Army adopting geogig
Yep Jody, @dinizime wrote that
I have talked to @DBlasby last year about that
On last year’s foss4g
Hi Folks! Currently I am trying to find out how geogig works if...
I want to use a Postgis database as the origin repo
Clone that to a local repo
edit data local As far as I am Aware I still need connection to the database...there is no possibility to edit disconnected/distributed with the postgres backend, isn't it?
There is, just like with GitHub you would eventually need to push data from one of your disconnected/distributed apps back to the origin repo.
@jodygarnett But how is the (Geo)datastore implemented locally then if there is not Postgres instance on the local machine? If I understand it right I have to Export to any Format from the cloned repo...edit for example in QGIS and then Reimport to the repo? Is this a suitable way of disconnected editing? Is this the only way?
You can choose what kind of storage to use locally.
Repo can use several different kinds of storage.
Hi all, I've got a PR here to upgrade GeoGig to the latest JTS: locationtech/geogig#446
Hi folks - I'm new to geogig but used to git, and I'm testing out geogig in an attempt to help some qgis users who aren't used to version control. One thing I find hard to explain to people are branches and they often get lost on a branch or detached state - for git, I usually recommend running git log with something like --graph --decorate --all to get an overview of where they are. I've been looking around in the manual for any similar functionality in geogig, but have been unable to find any - any suggestions? Is there something like gitk for geogig? Thanks!
Hi James, unfortunately no. reflog is on the todo list but given other pressing concerns it'd be a while until it's implemented. log --graph I'd say is doable but I don't see anyone taking the time to implement the VT output. We're currently working on a new WEB API and Qgis plugin, and also a JavaFX desktop client. I'd need to check when that'd open for general testing, I hope soon
Hi Gabriel, I assume that message was for me :) Thanks for the reply. I understand - sounds like you have plenty of stuff on your plate. I'm looking forward to trying out the Qgis plugin and desktop client once they're open for the public. That would probably be great help to those who aren't used to working with a terminal. Looking forward to seeing how GeoGig evolves, good luck to you.
can geogig store Triangulated Irregular Network (TIN) ?
Hello, no, ir can only store Feature collections like in SFS
@groldan, @DBlasby - I completed the ECA but there is still something I'm missing, hope you can help me get this PR locationtech/geogig#476 through. It updates GeoGig to being able to build against GeoTools 22. Thanks!
@vickdw a very large PR is coming in that will wipe out all deprecated APIs
For now it just removes the deprecated API that was in use
but it will be updated to remove all deprecated methods and classes, I counted 400 of them
(as per grep and wc -l, might be a bit less since some methods and classes are deprecated both as annotation and javadoc)
So, you might have to rework that PR or make another, if geogig uses any deprecated API
@aaime , thanks for the heads up. I'll look at that PR today and begin to plan for it.
Hi I'm new to the project and have been testing over the past day. One of the challenges I have to overcome is having different contributors to my repositories using different environments (postgis, gpkg). For example Editor #1 imports data from PostGIS and makes an initial commit and pushes changes to the remote. Editor #2 pulls changes, exports to geopackage and add/removes features. Editor #2 then imports the changes from then GPKG to the same repo, commits changes, and pushes to remote. I've tried this flow and everything looks good but seems like there may be a conflict on the feature id. I've event tried using the --fid-attrib flag. Any guidance/help would be appreciated.
@jdsterritt this project is not super active right now (no paid developers working on it - even though the implementation is great)
@groldan do you have any direction for @jdsterritt above?
I agree it seems the implementation is great. It is of great interest since versioning of spatial data is turning into a real challenge for us.
It is a real challenge in general, and I like the approach taken by geogig here (git workflow using features rather than rows of text).
There have been some great integrations with GoeNode and QGIS making it much easier to use. Perhaps others here could direct you towards them.
just wanted to caution you that the project is great technology, that is seeking funding / interest
I'm really struggling with getting geogig up and running and was hoping somebody could help me
with the command geogig init, I keep getting the following errors:
com.google.inject.ProvisionException: Unable to provision, see the following errors:
1) Error injecting constructor, java.lang.ExceptionInInitializerError at org.locationtech.geogig.rocksdb.RocksdbObjectDatabase.<init>(Unknown Source)
Any idea as to why?
Good day @careyshan_gitlab
The locationtech repo was recently migraed to be part of the main repository (recently like last week).
I guess that is a build problem, ...
Do you have any more information? does the stack trace continue ...
@careyshan_gitlab you might be using Java11 instead of Java8
I shoud also point out that the company/funding for this project evaporated last year. It is a really great technology and we hope it is picked up
@groldan I'm using Java8
@jodygarnett So is the project gone now?
the joy of opensource is the project is not gone
(indeed half the point of using a software foundation is so the technology has a longer shelf life then any one company)
but you should treat it as an experiment
far from gone, master is at its best, waiting for me to have enough spare time to cut a release