Difference between revisions of "WCS NetCDF Development"
m |
|||
| Line 62: | Line 62: | ||
== Currently Existing Darcs Repositories == | == Currently Existing Darcs Repositories == | ||
| − | + | [http://webapps.datafed.net/nest/OWS Datafed] | |
| + | |||
| + | [http://htap.icg.fz-juelich.de/darcs/OWS Juelich] | ||
== Currently Live WCS Servers == | == Currently Live WCS Servers == | ||
| Line 71: | Line 73: | ||
== Feature Requests of WCS == | == Feature Requests of WCS == | ||
| + | |||
| + | == Update Log == | ||
| + | |||
| + | 2007-10-22: Development starts for Windows. | ||
| + | 2007-11-12: First Linux port. | ||
| + | 2007-11-21: First Real Dataset online. | ||
| + | 2009-06-04: After being used for some datasets at Datafed, we get the first external developer, M. Decker, to contribute to the source code. He adds Unicode support. | ||
| + | 2009-07-02: First External darcs repository for OWS at Juelich. | ||
Revision as of 14:11, July 8, 2009
Web Coverage Service Access from NetCDF Files.
The goal is to provide you an easy way to publish your data in NetCDF format.
The cross-platform toolkit from data to Web Service contains:
- Checking the CF-1.0 compatibility of your data.
- Creating CF-1.0 compatible NetCDF files from your data.
- Setting up a Web Coverage Service directly from your files with no extra configuration.
Below are sections and descriptions to all the tools we provide. Feel free to participate!
Cross Platform Python-NetCDF Library
The most feature-rich library to access NetCDF files via python is PyNIO. Unfortunately it is unix-only. That's why nc3 was created. The two purposes of nc3 are:
- Provide simple read and write access to NetCDF files.
- WCS query: Extract a subcube from a NetCDF.
This is supposed to be a small library. The subcube query is supposed to be fast, everything else is supposed to be simple. If you do a lot of data processing using NetCDF or python, we recommend PyNIO.
NetCDF Markup Language: NCML
Creating a NetCDF file via plain API is difficult, NCML allows you to describe dimensions, variables and attributes in one xml document. This simplifies creating an empty NetCDF file a great deal.
Example NCML use:
(enter sample URL here)
CF-1.0 Convention
It's not enough to have NetCDF as format, you also need to standardize the attribute names, how to store dimensions etc. To make it easy to create CF-1.x compatible files, we provide a python library cf1.
The main page for CF-1.4 documentation: http://cf-pcmdi.llnl.gov/
(enter our sample project here)
Getting the binaries for Windows
Please follow the installation instructions in Windows or Linux. It's a little bit complicated: you have to download 3d party packages yourself, but please bear it. If you see an error, please fix it or report it.
This is enough for those who just want to create NetCDF-CF files and serve them as WCS. The python code is still delivered as source, so you can change it, but if you do please get the source code the proper way.
Getting The Source for Windows
We're using darcs. It is a very flexible DVCS: Distributed Version Control System. The one big page documentation is here.
Download it from here.
If you are A windows user, get darcs from here and we recommend integrated gui TortoiseDarcs.
once you have darcs installed, issue command:
darcs get "http://webapps.datafed.net/nest/OWS"
This will create directory OWS into your current folder. For windows, we recommend C:\.
Now you are free to edit any file.
Currently Existing Darcs Repositories
Currently Live WCS Servers
List of installed WCS servers here.
Existing Features WCS
Feature Requests of WCS
Update Log
2007-10-22: Development starts for Windows. 2007-11-12: First Linux port. 2007-11-21: First Real Dataset online. 2009-06-04: After being used for some datasets at Datafed, we get the first external developer, M. Decker, to contribute to the source code. He adds Unicode support. 2009-07-02: First External darcs repository for OWS at Juelich.
