-
Notifications
You must be signed in to change notification settings - Fork 63
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Can't load_remote_dataset in suite.py for OPeNDAP URL #500
Comments
I was able to execute all the commands successfully on MacOS, maybe this is isolated to windows? Does anyone else have a windows machine they can confirm on? |
I am able to reproduce on my Windows 10 laptop. |
At this stage, it's just a passthrough to netcdf's
So it seems like an upstream issue.
Should we file with unidata? |
Can you post the results of `conda list`. I'm afraid that this is a
libnetcdf issue, unrelated to compliance-checker.
Try updating libnetcdf, which should have a patch that fixes this, or force
install libnetcdf=4.4.0 (last working version).
…On May 18, 2017 1:57 PM, "Dave Foster" ***@***.***> wrote:
I am able to reproduce on my Windows 10 laptop.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#500 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AA6BL9_oYmyvRpKG973BfdhuLCW-V24Aks5r7LDNgaJpZM4NfNbq>
.
|
I see you have build 4 of libnetcdf, that is the patched one. If version
4.4.0 works I probably need to double check that patch or upstream's
solution to the problem.
…On May 18, 2017 2:02 PM, "Dave Foster" ***@***.***> wrote:
At this stage, it's just a passthrough to netcdf's Dataset() initializer
- it haves the same way in ipython:
In [4]: Dataset("http://oos.soest.hawaii.edu/thredds/dodsC/hioos/model/atm/ncep
...: _pac/NCEP_Pacific_Atmospheric_Model_best.ncd")
---------------------------------------------------------------------------
IOError Traceback (most recent call last)
<ipython-input-4-0da87aab71ef> in <module>()
----> 1 Dataset("http://oos.soest.hawaii.edu/thredds/dodsC/hioos/model/atm/ncep_pac/NCEP_Pacific_Atmospheric_Model_best.ncd")
netCDF4\_netCDF4.pyx in netCDF4._netCDF4.Dataset.__init__ (netCDF4\_netCDF4.c:13985)()
IOError: NetCDF: Invalid argument
So it seems like an upstream issue.
(ioos27) C:\Users\daf>conda list netCDF
# packages in environment at C:\Users\daf\Anaconda2\envs\ioos27:
#
libnetcdf 4.4.1.1 vc9_4 [vc9] conda-forge
netcdf4 1.2.7 np111py27_0 conda-forge
Should we file with unidata?
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#500 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AA6BL8Looel4WK50icwd4rRUetnYRp58ks5r7LH5gaJpZM4NfNbq>
.
|
@ocefpaf downgrading to |
Yep, |
I believe we can close this because it is not a |
@ocefpaf @daf So, for now if we stick with I'm also running into problems related to #497, so I'm wondering if a bug fix release of Since the prior release was only a few weeks ago and there have just been a few fixes since then, maybe not, but the cf debug output definitely is cramping my style. I've noticed it occurs for some URLs, but not all (which makes it impossible to workaround programmatically): Mixing issue content here, sorry...
(outputs cf table debug message)
(does not output cf table debug message) |
I don't think the issue is with the package in conda-forge. I got my hands on a Windows machine today and I test I can confirm that the PS: I am 👍 to a bugfix release. (Release early, release often!) Also, the issue here is unrelated to compliance-checker and there is nothing we can do besides workaround it. The only solution so far is to pin to |
FYI I have a PR to add continuous integration to the unidata netcdf-c repo in Unidata/netcdf-c#347 I added @mwengren two URLs to the test there, but you can see that the latest source does not even build on Python 2.7 (vc9) and 3.4 (vc10). Here are the results: https://ci.appveyor.com/project/ocefpaf/netcdf-c/build/1.0.23 |
@ocefpaf @daf @jbosch-noaa @kknee who can/should handle a Compliance Checker release at this point? Is everything resolved from the PRs/commits since the last release? |
I'm happy to do it, will get to it in an hour or so. |
I'll in airports most of the day but I can update the conda package as soon
as the release is out.
…On May 23, 2017 7:04 AM, "Dave Foster" ***@***.***> wrote:
I'm happy to do it, will get to it in an hour or so.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#500 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AA6BL1S46AvQ3Gyx1ojGlgNbIWAnTvBsks5r8udPgaJpZM4NfNbq>
.
|
@mwengren released 3.0.4 to pypi. @ocefpaf got started on a bump for conda-forge, conda-forge/compliance-checker-feedstock#18 |
I recently updated my Compliance Checker to 3.0.3 and am unable to run tests against OPeNDAP URLs from PacIOOS.
Here's a few examples:
Output is similar:
Apologies is this is user error or something I'm missing about running this on Windows (Python 2.7), but it does not seem able to read these OPeNDAP URLs. It is successful in the same environment with several NANOOS OPeNDAP services.
The text was updated successfully, but these errors were encountered: