|
POST
|
The logical layout of Personal Geodatabase (.mdb) and File Geodatabase (.gdb) is documented in the help system. Unlike the shapefile format, the binary format of .mdb/.gdb files isn't publicly documented anywhere (.mdb is managed via the Microsoft Jet engine, and .gdb is managed directly by ArcObjects). The File Geodatabase API is in the final stages of release, but the API only provides access to the data in file geodatabases (and only 10.0 FGDB at that), not the implementation details. How are these formats linked to your thesis? - V
... View more
03-29-2011
07:34 AM
|
0
|
0
|
677
|
|
POST
|
I had a similar need 11 years ago, so I wrote a 'C' utility I called 'sdeupdate' to modify the rows in an SDE table in batch mode. It was designed to work with the output of 'sdequery' with a WHERE clause that identified only the updated features. Both utilities are part of the 'se_toolkit' suite, available by FTP download (the documentation has examples of using the two piped together). I regularly INSERT, UPDATE, and DELETE rows in non-versioned layers which are also being served via AGS. If you are having SDE lock issues, it might be because you have auto-locking enabled. You can disable autolocking with: sdelayer -o alter -l table,geomcol -L OFF Of couse, today you can modify the geometry when a coordinate column is updated by using ST_GEOMETRY or a native geometry type with a trigger to execute the update (which would yield real-time update, instead of updates dependent on the frequency of batch execution). - V
... View more
03-27-2011
05:53 AM
|
0
|
2
|
1107
|
|
POST
|
Which Visual Studio? 32-bit or 64-bit? Are you using the Release or Debug libraries? Are the jobs run on the same machine with identical load? Why do the Beta-2 and Beta-3 passes have different numbers of vertices? How large are the directories before you start delete operations? What happens if you use the Beta-2 shape creation protocol with the Beta-3 API DLL (one variable test)? - V
... View more
03-24-2011
08:47 AM
|
0
|
0
|
898
|
|
POST
|
Spatial searching is still "limited to the envelope-intersects operator", so you'd need to code your own "line within polygon" filter (with or without edge touch, as appropriate). - V
... View more
03-24-2011
04:27 AM
|
0
|
0
|
337
|
|
POST
|
Comparing ST_GEOMETRY with SDEBINARY is a rigged fight. LONG RAW is much faster than the LOB family of types. The format of the points BLOB in ST_GEOMETRY and SDELOB/ SDEBINARY is exactly the same, so this isn't an "unpacking" issue, just inherent performance of the underlying RDBMS BLOB type. Given that ST_GEOMETRY often out-performs SDELOB (due to elimination of a table and the potential for inline storage), and given the instability of LONG RAW at 10gR2 (it's deprecated, so sites with difficulty have no recourse) and lack of availability at 11g, an ST_GEOMETRY default type makes good sense. I had similar optimizer issues with SDO_GEOMETRY types, and found that a view against the table with an indexing hint in the view addressed full table scan issues. I've never reproduced the optimizer issue with ST_GEOMETRY, so I don't know if the same workaround is applicable. - V
... View more
03-24-2011
04:12 AM
|
0
|
0
|
1556
|
|
POST
|
This is a chicken-egg kind of error from Oracle. A shutdown/restart won't change the metadata which generates a 29861 error. You can often delete the index by using the FORCE option of "DROP INDEX". - V
... View more
03-22-2011
06:57 AM
|
0
|
0
|
1229
|
|
POST
|
Esri never released an 8.1.1 version of ArcSDE (8.1 went directly to 8.1.2; both were retired in May 2004). What version and service pack of ArcSDE are you using? With which database? The 'shp2sde' command always allows you to specify a coordinate reference with the 'create' option. In fact, you *should* use the '-x' or '-R' option every time you use 'shp2sde' or 'sdelayer' to create a layer. The 'shp2sde' utility includes a "reject" option ('-r') to allow you to place shapes which fail validation into a location for further processing. If the data works at 100*N scale, but not at N scale, that means you will likely need to do significant processing to make it load at N scale. - V
... View more
03-22-2011
06:11 AM
|
0
|
0
|
2187
|
|
POST
|
I can't help but wonder if you can bridge the gap by installing ArcSDE in the 11g instance, capture the DDL to recreate the tables, then import them, as three discrete steps. I've done this dozens of times with SDEBINARY -> SDELOB conversion using 'exp' and 'imp', between 9i and 10g, and from 10g Windows to 10g Linux, and from 10gR2 to 11gR2. It just seems too much to ask a single utility to get the entire process perfect, which is what you're expecting of 'impdp' in this situation (and 'expdp' just can't be smart enough to organize the data in the perfect order that 'impdp' would need to make the attempt). Upgrading in place has its allures, but I find it beneficial to do a fresh database implementation with each release, so that optimization of the vector tablespaces and raster blocks can occur (this usually involves editing the DDL SQL text to change storage clauses and reloading the rasters with different tile sizes to reduce chaining [especially when going from a 16k to 8k DB_BLOCK_SIZE]). One way to achieve this is to optimize in a 10g->10g transfer, then do an 11g in-situ upgrade. - V
... View more
03-21-2011
06:26 PM
|
0
|
0
|
2787
|
|
POST
|
There's a lot a variables to be resolved before a simple "No" can be given. Is the table versioned? Does the table participate in geodatabase topologies, networks, or object behaviors? Does the layer use SDEBINARY/SDELOB storage? If so, is it in LOAD_ONLY I/O mode? In general, there are triggers to maintain referential integrity, but any of the above could preclude the safe use of DELETE. Note that the same problems would exist if you used the ArcSDE API (SE_stream_delete_from_table is just a wrapper on a SQL delete), but the I/O mode problem would probably be caught at the API. - V
... View more
03-21-2011
09:50 AM
|
0
|
0
|
368
|
|
POST
|
If all tables registered with ArcSDE have been dropped *AND* there are no tables using ST_GEOMETRY in the instance, then the easiest way to fix your ArcSDE instance is to "DROP USER sde CASCADE" and reinstall. Note that if you *DO* have outstanding tables with ST_GEOMETRY columns, then dropping SDE will corrupt your Oracle instance to the point that it too will need to be dropped and recreated. If you are uncomfortable doing SQL queries against the SDE metadata tables to assess the status of your instance, then you should probably log an incident and have a Tech Support analyst walk you through the necessary steps. - V
... View more
03-21-2011
07:51 AM
|
0
|
0
|
799
|
|
POST
|
Does an GDB_OBJECTCLASSES row that corresponds to the dropped table exist? Is there an entry in GDB_DOMAINS which corresponds to the object class ID? The easiest way to uncorrupt the geodatabase registry is to create a table with the name of the dropped table via SQL, then use Catalog to delete it. - V
... View more
03-21-2011
07:07 AM
|
0
|
0
|
799
|
|
POST
|
What database are you using? What version and service pack of ArcSDE are you using? What version and service pack of ArcGIS are you using? Is the domain referenced by a table in the GDB metadata? Does that table still exist? - V
... View more
03-21-2011
04:49 AM
|
0
|
0
|
799
|
|
POST
|
I guess it depends on how complex the original state tree is. The first step would be to identify all the branches in the state tree. Then to determine the numbers of features in each segment which are inserts and deletes along each segment. From there you can make stateN-stateM-mode exports out of the source database, and reapply the edits in the same order into the target database. This would of course be complicated by custom behaviors and object properties. It's not a pleasant task, but it probably isn't impossible (unless you require the same objectids and state numbers in the target database). In theory you could even use sdeexport/sdeimport to transfer the data (since sdeimport can specify a target version), but you'd need a way to identify the features in a form that a where clause could evaluate to make the exports (because sdexport doesn't expose access to the full range of version difference rules that the ArcSDE API and ArcObjects provide). I'd have to defer to the replication gurus on whether this is a simpler task from their standpoint, but the source database would have to be replication-ready (or a clone of the original made that way) to make that work. - V
... View more
03-19-2011
08:33 PM
|
0
|
0
|
1054
|
|
POST
|
The file geodatabase API has a separate (and different) license agreement from ArcEngine. There's no need for FGDBAPI if you're using ArcEngine (which has ArcObjects to access 10.x file geodatabases in addition to 9.x FGDBs and all the other formats and geospatial processing capabilities). Release of the file geodatabase API is akin to publishing the specification for the shapefile format -- a way to allow those without Esri software to exchange geospatial data with those who do. It's an open Beta, and the release includes samples. Lance's Welcome post includes a link to the blog announcement (which includes a link to the download page). The license is in a PDF in the "license" directory. - V
... View more
03-17-2011
04:57 AM
|
0
|
0
|
341
|
|
POST
|
I'm not on the Development side, but I think the issue would be the addition of eight more configurations (Release + Debug x 32 + 64 x vs9 + vs10 x MD + MT). Other Esri products like ArcSDE also follow the DLL distribution model. - V
... View more
03-16-2011
02:31 PM
|
0
|
0
|
722
|
| Title | Kudos | Posted |
|---|---|---|
| 2 | 10-10-2025 07:28 AM | |
| 2 | 10-07-2025 11:00 AM | |
| 1 | 08-13-2025 07:10 AM | |
| 1 | 07-17-2025 08:16 PM | |
| 1 | 07-13-2025 07:47 AM |