Rasterio version 1.0 fails silently

[jpolchlo]

GeoNotebook works using Rasterio 0.36.0, but when attempting to use with Rasterio 1.0a7, we observe an opaque error message of Warning 1: GeoTransform node does not have expected six values after adding a RasterData layer, and nothing is displayed on the map.

[aashish24]

thanks for reporting @jpolchlo. Any chance you can submit a patch that fixes this issue?

[jpolchlo]

If I could figure out the origin of the error, then yes. But after downgrading rasterio, I've got a working version again; so this bug is no longer on my radar. For now.

[danlamanna]

I tried bisecting rasterio and was unsuccessful (a number of commits fail to run outright leaving the bisect with a large pool of candidates).

Anyway, it's not that surprising that upgrading from 0.36.0 to 1.0 alpha (not sure why alpha versions are being published to pypi) would cause breakage. We should be pinning to rasterio<1.0 in setup.py until someone can take a look at how the API surface has changed.

On Mon, Aug 28, 2017 at 15:07 PM, Aashish Chaudhary [email protected] wrote:

thanks for reporting @jpolchlo. Any chance you can submit a patch that fixes this issue?

-- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenGeoscience/geonotebook/issues/139#issuecomment-325380382

[jbeezley]

I actually did the migration many months ago. It looks like I lost the branch, but it wasn't too hard at the time. I think a couple of imports moved, and the geotransform class changed, but that was about all.

[danlamanna]

not sure why alpha versions are being published to pypi Actually looking at it now, pip install rasterio only installs 0.36.0 and I'm not given the alpha unless I use the --pre option (for pre-releases). Regardless we should still pin it for when the release becomes stable.

On Wed, Aug 30, 2017 at 09:37 AM, Dan LaManna [email protected] wrote:

I tried bisecting rasterio and was unsuccessful (a number of commits fail to run outright leaving the bisect with a large pool of candidates).

Anyway, it's not that surprising that upgrading from 0.36.0 to 1.0 alpha (not sure why alpha versions are being published to pypi) would cause breakage. We should be pinning to rasterio<1.0 in setup.py until someone can take a look at how the API surface has changed.

On Mon, Aug 28, 2017 at 15:07 PM, Aashish Chaudhary [email protected] wrote:

thanks for reporting @jpolchlo. Any chance you can submit a patch that fixes this issue?

-- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenGeoscience/geonotebook/issues/139#issuecomment-325380382