Add geometry conformance #6
Conversation
…mendation of first-last node equivalence for polygons
* Geometry_Conformance: require axis on coordinate vars Remove case sensitive requirement for geometry type, and remove recommendation of first-last node equivalence for polygons Clarify that *nodes* for polygon rings must be in a certain order. add geomery requirements
| @@ -552,6 +552,8 @@ a geometry container variable. The variable name must exist in the file. | |||
| **`node_coordinates`** attributes. | |||
| * The only legal values of geometry_type are **`point`**, **`line`**, | |||
| and **`polygon`** (case insensitive). | |||
| * For a line **`geometry_type`**, each geometry must have a minimum of two node coordinates. | |||
| * For a polygon **`geometry_type`**, each geometry must have a minimum of three node coordinates. | |||
There was a problem hiding this comment.
Shouldn't "three" be "two", in which case, we could merge two rules into one:
- For a line or polygon
geometry_type, each geometry must have a minimum of two node coordinates.
E.g. In Example 7.16. (Polygons with holes), we have only two node coordinates. (https://github.com/twhiteaker/cf-conventions/blob/master/ch07.adoc)
There was a problem hiding this comment.
I was referring to the size of the node dimension, not the number of node coordinate variables. Is this rule clearer?
- For a line
geometry_type, each geometry must have a minimum of two nodes.
There was a problem hiding this comment.
Ah - I see - my mistake! I think the original text is fine. I had confused "node coordinates" with "node coordinate variables". it is clear in chapter 7, so I think that it is OK here, too.
| the **`axis`** attribute. | ||
| * If a **`coordinates`** attribute is carried by the geometry container variable or its parent data | ||
| variable, then those coordinate variables which correspond to node coordinate variables must have a | ||
| **`bounds`** attribute that names the corresponding node coordinate. | ||
| * The geometry node coordinate variables must all have the same single dimension, | ||
| which is the total number of nodes in all the geometries. | ||
| * The nodes must be stored consecutively for each geometry and in the order of the | ||
| geometries, and within each multipart geometry the nodes must be stored | ||
| consecutively for each part and in the order of the parts. | ||
| * Nodes for polygon exterior rings must be put in anticlockwise order (viewed from above) | ||
| and polygon interior rings in clockwise order. |
There was a problem hiding this comment.
The above two rules concerning the order of nodes are not checkable, so shouldn't they be omitted from the conformance requirements? These points are already made in chapter 7.
There was a problem hiding this comment.
- The nodes must be stored consecutively for each geometry...
That one is uncheckable. I agree with omitting it.
- Nodes for polygon exterior rings must be put in anticlockwise order...
This one can be checked. Here's a checker in Python.
There was a problem hiding this comment.
That's good for me. Thanks for the link to your geometry code.
There was a problem hiding this comment.
I removed the uncheckable rule. You can find the latest version here:
https://github.com/cf-convention/Conformance/pull/6/files
conformance.adoc
Outdated
| @@ -568,9 +568,6 @@ variable, then those coordinate variables which correspond to node coordinate va | |||
| **`bounds`** attribute that names the corresponding node coordinate. | |||
There was a problem hiding this comment.
Following the conclusion of the "nodes" discussion, this check needs updating. How about:
- If a coordinate variable named by a
coordinatesattribute carried by the geometry container variable or its parent data variable has anodesattribute, then thenodesattribute must be a string whose value is a single variable name. The specified variable must be a node coordinate variable that exists in the file.
When node_count attribute is missing, require the dimension of the node coordinate variables to be one of the dimensions of the data variable.
|
Given that cf-convention/cf-conventions#158 has been merged. I'm going to go ahead and merge this too. |
Add conformance rules for geometries. See trac ticket 164 and CF conventions pull request 115 for geometries discussion.