|
Re: Evolution of OGRLayer ?: msg#00124gis.gdal.devel
Jean wrote: Here is a proposition for OGRLayer... So would you be interrested in a public method on OGRLayer to get Jean, I can see your point about why this would be desirable in some circumstances. However, it isn't so obvious to me that I could correctly and efficiently implement such an operation on many of the existing formats with a reasonable amount of effort. I would add that it would be even harder to implement this efficiently if the query is anything other than "get the whole layer". I would encourage you to file this as a feature request in bugzilla, but for the time being I won't be acting on it. I assume you are interested in implementing such a grid display/editing mode for large layers where it would be impractical to hold them all in memory? If the layers to be operated on are not too huge, my first suggestion would be to load all features into a memory layer and then have the grid display operate against that. A "Memory" OGR driver was implemented a few months ago specifically for high speed access for editing in interactive applications. If the layers are likely to be too large to load into RAM, some other approach would be necessary. Depending on the format, my suggestion would be to either make one pass recording the feature id's for subsequent direct feature requests (cost 4 bytes per feature of RAM, and one pass) or to basically sequentually read through to the desired area for each display update, likely keeping a substantial "window" of features in RAM in the application to avoid having to re-read for refreshes, or small moves through the result. Hoping my english is not that bad, Very understandable. Best regards, -- ---------------------------------------+-------------------------------------- I set the clouds in motion - turn up | Frank Warmerdam, warmerdam@xxxxxxxxx light and sound - activate the windows | http://pobox.com/~warmerdam and watch the world go round - Rush | Geospatial Programmer for Rent |
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | Re: importing SRTM30: 00124, Andrey Kiselev |
|---|---|
| Next by Date: | Re: image subsetting for arbitrary polygon: 00124, Frank Warmerdam |
| Previous by Thread: | Evolution of OGRLayer ?i: 00124, Jean |
| Next by Thread: | GXF and Imagine Fixes: 00124, Frank Warmerdam |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |