Open alwunder opened 11 months ago
Is this mostly an issue of "Only one reference to each missing term is cited"? Because I think it could still be useful to have a report of a term being used in a table and field even if it's been reported as missing from another. For one thing, when the term or missing entry is added, the errors will go away.
Right now, the errors for each rule are appended to a list as they are found as table, field, term triples (or similar depending on the rule) and the list is converted to a set to eliminate duplicate triples. It's not that the term itself is evaluated.
But it makes sense to change the error headers to make it more clear.
Ah, I see. It's the unique triple that is being reported. I guess in previous validations I either had not noticed it or just didn't have issues with missing terms across multiple locations in the database. My typical workflow has most attributes populated as the features are created, so I wouldn't have run into this issue very often (if at all).
The database that produced this report in particular was converted from a coverage, and I have been filling in values as I evaluate the quality of the linework. So, having the report give me the locations of the missing items probably does have its advantages.
I agree with your comment to change the error header. That makes what the tool is reporting clear to the user (and it's an easy fix).
Validation script gives duplicate entries in some rules. Notable are 2.4, 2.6, 3.4, 3.6, and 3.8. I added some characters in front of a few examples in each section to make them stand out. See below: