Latest Contributions by VinceAngelo

ArcGIS Server is part of Enterprise. - V ... View more

64-bit OID support is an existing enhancement request, as is 64-bit integer column support. - V ... View more

I haven't done any Tk with Python.  I use Python toolboxes, or just plain scripts. The key is changing the arcpy.env.workspace to use the .sde file instead of file_path (or overwrite file_path with the connection file string). - V ... View more

There's a lot of fluff unrelated to the problem in this script, but there doesn't seem to be any reason you couldn't tear off the UI prompt that is hard-coded to picking  a directory (.gdb folder) and specify a connection file path instead (.sde) You certainly don't need to copy the data to a file geodatabase to make this work. It might even just be changes to the 'askdirectory' UI calls. - V ... View more

You don't need to post the code twice, but you do need to post it legibly, using a code block in the formatting section. Indenting is of critical importance in Python, so any display mechanism which strips whitespace makes it unusable. - V   ... View more

Making a copy of the enterprise geodatabase to FGDB is the long way around the lake to get an inventory of the contents. And any problem which prevents access to the enterprise geodatabase to make an inventory would prevent data copy as well.  This sure seems to be an XY Problem (a question asking about a flawed solution instead of the root problem). Since the script should be under ten lines, you really ought to post it. From a terminology standpoint, it might help to realize there is no longer any such thing as an "SDE database" (and hasn't been since 8.0 was released). There are Oracle, SQL Server, PostgreSQL, Informix and DB2 enterprise databases, but no SDE. Each of the supported RDBMSes are accessed in the same way, with ListDatasets, ListFeatureClasses and ListTables, and GetCount. - V   ... View more

You can't use ArcGIS Server without a license. You can't enable an enterprise geodatabase in PostgreSQL without a Server license. You can't use MySQL as an enterprise geodatabase (even with a license) You can use PostgreSQL/PostGIS with Desktop (ArcMap/Pro) without a Server license, but you can't do editing or UpdateCursor or any geodatabase functionality (you can upload new tables, insert to existing tables, and query tables, plus issue SQL in ArcPy, but that's about it). All the documentation is online and in the software install. Portal is the Enterprise component that correlates to AGOL, though there are compatibility issues, since AGOL sees updates/new functionality before it is ported to Portal. - V ... View more

Hmmm, well, you should probably ask the Geodatabase team to write that blog. I've been a PostgreSQL admin, nearly exclusively, for most of the past decade, so you wouldn't want me working on it anyway. - V ... View more

That's not exactly a common use pattern, and therefore probably shouldn't be used as final arbiter for storage selection.  I've had cases where SDO_GEOMETRY was 15-20% faster, and cases where ST_GEOMETRY was 40-120% faster, though I've mostly been using PostgreSQL in the past decade, with PostGIS geometry, exclusively (which has been 1-2 orders of magnitude faster returning millions of rows against some very disadvantaged Oracle configurations). Best practice is to benchmark apples-to-apples for an entire solution, then evaluate from there, not just rely on a single datapoint. - V ... View more

More details needed: What version of ArcGIS? Using which RDBMS? If non-Oracle, is the database name going to be the same? At what release is the geodatabase? Are all the tables non-versioned? What happens after the move? (Does the old database go away?  Do they need to be kept in sync?) Note that the current term of art is "enterprise geodatabase" (hasn't been "SDE" since 8.0 was released) - V ... View more

You don't ever want to visualize a million anything -- it just takes too long. The solution is to model this with fewer features. I had a table with 74 million things where I needed to see those things, symbolized by type, at various scales. At 1:20m scale, there could be millions of features, and display took minutes. What I did was alter the features to have the 2, 5, and 10 degree grid cell as a property, then dissolved the points by grid cell and type classification into multi-point features into _grid2, _grid5, and _grid10 tables. Then I generated scale-dependent layers, so the features could display by type, tens and hundreds of thousands of features at a time in each cell.  Render time at any scale was subsecond (often <100ms), but the symbology never changed, and because the features were still present, an identify on any point still provided the full information of any feature. Something similar is possible with lines and polygons (if you model the polygons as their exterior boundary lines), intersecting the features with fishnet cells (encoded with all three cell levels) then dissolving.  If the features are dynamic, updating the features can be done as a batch, or using an enterprise geodatabase, as a trigger. - V ... View more

Keep in mind that there has not been an "in SDE" since the rewrite at 3.0.  Data resides "in" a database. All of the tools of the supported databases are available, though some of them may be incompatible with successful deployment (e.g. 64-bit integer types). The versioning  model can be incompatible with UNIQUE values, since multiple conditional edits can exist in the ADDs table, and enforcement would be delayed until much too late.  You need a proactive validation process that looks for eventual duplicates. - V ... View more

I name my tables with a numeric suffix (e.g., "project_stuff_t1") so I can make later changes ("project_stuff_t2") and keep track things like type changes.  This often manifests with revised service names ("proj_data_10" -> "proj_data_11" or "proj_data_20", depending on scope changes), allowing the "old" service to run while dashboards are updated to use the new service.  Of course, to pull this off, you don't modify the table, but create a new one from the old before archiving the old content and dropping it. - V ... View more

You can simulate large datasets ("millions" might not make the cut for "huge" -- I put it at closer to 50M points for "very large") with ten lines of SQL code.  For example, in PostgreSQL (grabbed from my post in GIS SE ) ... CREATE TABLE example1 ( idcol serial NOT NULL, geomcol geometry NULL, CONSTRAINT example1_pk PRIMARY KEY (idcol), CONSTRAINT enforce_srid CHECK (st_srid(geomcol) = 4326) ); INSERT INTO example1(geomcol) SELECT ST_SetSRID( ST_MakePoint( (random()*360.0) - 180.0, (acos(1.0 - 2.0 * random()) * 2.0 - pi()) * 90.0 / pi()), 4326) as geomcol FROM generate_series(1, 1000000) vtab; CREATE INDEX example1_spx ON example1 USING GIST (geomcol); The same is possible with ArcPy of course, as a DA InsertCursor. Finding data is generally more domain-specific than a generic request on a web forum. - V ... View more