Opened 14 years ago
Closed 14 years ago
#89 closed defect (fixed)
WKT regression due to RFC 2
| Reported by: | Norm Olsen | Owned by: | Norm Olsen |
|---|---|---|---|
| Priority: | major | Milestone: | 2.0 |
| Component: | Library | Version: | svn-trunk |
| Keywords: | WKT | Cc: |
Description
Under certain unknown (as yet) circumstances, the WKT procession in the post RFC 2 CS-MAP fails to process WKT strings which were processed correctly in the pre RFC 2 version of CS-MAP.
Change History (1)
comment:1 by , 14 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
| Version: | → svn-trunk |
Note:
See TracTickets
for help on using tickets.
This submission produced revision 1951.
It turns out that the special circumstances referenced above consist of all cases where the binary fallback technique is used to locate a match between the WKT string and the CS-MAP dictionaries. This was cauuse by a simple onnission of one line of code. It all worked prior to RFC 2 as it was the nueric contents of the datum definition that were important and the name was usually ignored. Post RFC 2, this is just the opposite. The numeric portions of a datum definition are ignored and the name become very important.