| Commit message (Collapse) | Author | Age | Files |
| | |
|
| |
|
|
| |
The database has uses beyond the webmap.
|
| |
|
|
| |
public schema.
|
| | |
|
| |
|
|
| |
See https://www.postgresql.org/docs/current/sql-cluster.html .
|
| |
|
|
| |
Cf. https://gdal.org/en/stable/drivers/vector/pg.html#layer-creation-options
|
| |
|
|
|
| |
Cf. OGR_PG_ENABLE_METADATA=[YES/NO] from gdal ≥3.9.
https://gdal.org/en/stable/drivers/vector/pg.html#configuration-options
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
| |
Rather than putting everything in the 'postgis' schema using ':' as
hierarchy separator.
When the destination dataset is not PostgreSQL, the layers names are
prefixed using '$SCHEMA.' instead of '$SCHEMA:'
|
| |
|
|
|
|
|
| |
manually.
This assumes gdal ≥3.7, see
https://gdal.org/en/stable/api/python/raster_api.html#osgeo.gdal.Dataset.ExecuteSQL .
|
| | |
|
| |
|
|
| |
To match the layer definition found upstream.
|
| |
|
|
| |
(Based on Mozilla Firefox 140.4.0esr.)
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
When the output driver supports altering GeomFieldDefn's nullable flag
after the fact. PostgreSQL supports that. GPKG doesn't, but one can
use GEOMETRY_NULLABLE=NO in layer creation options instead.
See https://gdal.org/en/stable/development/rfc/rfc53_ogr_notnull_default.html
This commit partially reverts 2c3ee79cb434fc4cf315ee3a6a526156053d76c4.
|
| | |
|
| |
|
|
|
|
| |
Rather than CreateLayer(). Unfortunately not very helpful since
.SetNullable() appears to be a no-op; it doesn't set a NOT NULL
constraint on the geometry column.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
To avoid JSON-encoding on the Python side.
|
| |
|
|
|
| |
Recent GDAL version can set comments on columns, which is now reflected
on the schema.
|
| |
|
|
|
| |
Older GDAL version need the destination path to be a string not a
path-like argument.
|
| | |
|
| |
|
|
| |
For consistency with raster handling.
|
| |
|
|
|
|
|
|
|
|
| |
Raster data is not stored in the PostGIS database. Instead, the mtime
of the target directory is used to determine whether the COG is up to
date.
Add a new flag --metadata-compress for JSON metadata compression (which
also applies to MVT metadata), and --rasterdir for the target raster
directory.
|
| |
|
|
| |
Add rename `administrativindelning` layer group to `adm`.
|
| |
|
|
| |
been updated upstream.
|
| |
|
|
|
| |
This reverses commit 86f1e635a647e94c33e10e4d230853fd5d84c0bc. It
appears that .zip files have a consistent schema now…
|
| | |
|
| |
|
|
|
|
|
| |
“Age” is misleading as it suggests a value that increases over time as
times stretches away from a fixed epoch. A timestamp (ts) is a more
appropriate term for that field (it's rather a “daystamp” but that's a
matter of precision).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This allows client-side filtering.
The "age" attribute is a signed short (int16) expressing the number of
days since 1970-01-01. This covers the range 1880-04-15 to 2059-09-18
which should be more than enough. The source value is a Date or
Datetime and depends on the source layer.
- For vbk:*, it's the date at which the project was last saved in
Vindbrukskollen.
- For mrr:*, it's the date at which the application came to
Bergsstaten for applications, and decision date for granted permits.
- For avverk:*, it's the date at which the application came to
Skogsstyrelsen for applications, and the clearcut date (according to
Skogsstyrelsen) for completed objects.
For other layers, we don't export that attribute.
This makes the tiles a bit larger. Before (without the extra
attribute):
vbk: 1261× tiles, 599kiB uncompressed (avg=487B/t), 389kiB brotli (35%, avg=316B/t)
avverk: 3734× tiles 115MiB uncompressed (avg=32kiB/t), 72MiB brotli (37%, avg=20kiB/t)
mrr: 1324× 331kiB uncompressed (avg=257kiB/t), 289kiB brotli (13%, avg=223B/t)
→ total 121631367B uncompressed (avg=19kiB/t), 76692807B brotli (37%, avg=12kiB/t)
After (with the extra attribute):
vbk: 1261× tiles, 714kiB uncompressed (avg=580B/t), 425kiB brotli (40%, avg=345B/t)
avverk: 3734× tiles 127MiB uncompressed (avg=35kiB/t), 78MiB brotli (39%, avg=21kiB/t)
mrr: 1324× 323kiB uncompressed (avg=322kiB/t), 342kiB brotli (18%, avg=265B/t)
→ total 134274796B uncompressed (avg=21kiB/t), 82264731B brotli (39%, avg=13kiB/t)
Summary:
+12.1MiB uncompressed (+10.4%, avg=+1.95kiB/t)
+5.3MiB compressed (+7.3%, avg=+882B/t)
|
| |
|
|
|
| |
This avoids crashing when `webmap-import` is run as another user on an
homedir with restrictive permissions, for instance.
|
| |
|
|
| |
a stable attribute.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
“lst:riksintresse_*”.
|
| |
|
|
|
| |
And “sks:AvverkAnm” resp. “sks:UtfordAvverk” to “sks:avverk_anmald”
resp. “sks:avverk_utford”.
|
| |
|
|
|
| |
It's clearer that way, after all Naturvårdsavtal and Återvätningsavtal
are not part of Naturvårdsverkets Naturvårdsregistret.
|
| |
|
|
|
| |
Including k-skog, but we leave that one alone for now as the script
lacks some logic to process raster files.
|
| | |
|
| | |
|
| |
|
|
| |
Inspired from https://vbk.lansstyrelsen.se/ .
|
| | |
|