Reading GIS files(.shp, .dbx, .) into ZopeDatabase
I want to Reading GIS files(.shp, .dbx, .iforget) into ZopeDatabase. Does anyone have any ideas how I could accomplish this? any existing modules? Mailing list? --------------------------------- Yahoo! FareChase - Search multiple travel sites in one click.
Am Sonntag, den 06.11.2005, 17:10 -0800 schrieb Allen Huang:
I want to Reading GIS files(.shp, .dbx, .iforget) into ZopeDatabase. Does anyone have any ideas how I could accomplish this? any existing modules? Mailing list?
Modules for python to work with GIS are usually integrated with specialized backends, such as ArcView and friends. I've never heard of anybody trying GIS with zope. Its seems you are the first here :-) I'm not saying its impossible, but you might have a lot of work, as learning the fileformats, build parsers, create transforming objects maybe using PIL, create specialized index for ZCatalog and so on. Good luck! --Tino.
Am Sonntag, den 06.11.2005, 17:10 -0800 schrieb Allen Huang:
I want to Reading GIS files(.shp, .dbx, .iforget) into ZopeDatabase. Does anyone have any ideas how I could accomplish this? any existing modules? Mailing list?
PS: Maybe http://pyogclib.sourceforge.net/ is something you can use too.
Allen Huang wrote at 2005-11-6 17:10 -0800:
I want to Reading GIS files(.shp, .dbx, .iforget) into ZopeDatabase.
What does "reading GIS files into ZODB" means? Storing them as blobs (Binary Large OBjects) is trivial (you use a "File" object for this). Understanding them might prove a bit more difficult. -- Dieter
I know it might be difficult, but I like try it. I don't really like to use multiple databases if I could fit all the data into onto zope or mutiple zope databases, maybe it could increase dataflow since I wouldn't need to transform data from one format to another. But I'm really concern on how it will stress my RAM. I made a system with zope to take in members and when I try to do some statistics with zope, I had to change from a 256mb ram to a 512mb ram. what do think about this Dieter?? Dieter Maurer <dieter@handshake.de> wrote:Allen Huang wrote at 2005-11-6 17:10 -0800:
I want to Reading GIS files(.shp, .dbx, .iforget) into ZopeDatabase.
What does "reading GIS files into ZODB" means? Storing them as blobs (Binary Large OBjects) is trivial (you use a "File" object for this). Understanding them might prove a bit more difficult. -- Dieter --------------------------------- Yahoo! FareChase - Search multiple travel sites in one click.
Am Dienstag, den 08.11.2005, 17:07 -0800 schrieb Allen Huang:
I know it might be difficult, but I like try it. I don't really like to use multiple databases if I could fit all the data into onto zope or mutiple zope databases, maybe it could increase dataflow since I wouldn't need to transform data from one format to another.
But I'm really concern on how it will stress my RAM. I made a system with zope to take in members and when I try to do some statistics with zope, I had to change from a 256mb ram to a 512mb ram.
what do think about this Dieter??
I'm not Dieter but what I can say is: avoid the use of large monolitic objects and try to split it all up into smaller chunks and work with them in small portions at once. This way you can save on memory usage.
Allen Huang wrote at 2005-11-8 17:07 -0800:
I know it might be difficult, but I like try it. I don't really like to use multiple databases if I could fit all the data into onto zope or mutiple zope databases, maybe it could increase dataflow since I wouldn't need to transform data from one format to another.
But I'm really concern on how it will stress my RAM. I made a system with zope to take in members and when I try to do some statistics with zope, I had to change from a 256mb ram to a 512mb ram.
Zope is incredibly flexible. There are huge amounts of buttons to tune and when you forget to tune or tune in the wrong way, the result will be suboptimal. For large numbers of members, e.g. it is necessary to use BTrees based data structures (they have a hiearchical tree structure and are not flat as easier data structures). As someone else already said: you should not store huge objects as a single persistent object but chunk it into smaller blocks. The built-in "File" object already does this but may not be fully adequate for GIS specific access. -- Dieter
participants (3)
-
Allen Huang -
Dieter Maurer -
Tino Wildenhain