...
It's worthy at this point to note any datetime fields. Postgresql prefers to import datetime strings as YYYY-MM-DD HH24:MI:SS so some manipulation will be required to adjust these fields into the correct format. The following are two examples from a customer's Brightly Confirm installation.
Street Light Faults Field Definitions
Field Name | Data Type | Example / Notes |
Asset Id | character varying | AB/13/E |
Central Asset Id | character varying | Unique ID in Confirm |
Job Number | integer | 173243 |
Job Entry Date | datetime (DD/MM/YYYY HH24:MI:SS) GMT | 19/06/2022 11:37:28 |
Site Name | character varying | Albert Road |
Locality Name | character varying | N4 |
Feature Location | character varying | S/O No 86A Florence Road |
Job Notes | character varying | General Street Lighting Issue - Lamp on 24/7 |
Priority Name | character varying | Street Lighting - 7 days |
Status Code | character varying | JS33 |
Status Name | character varying | Job Received by Contractor |
Asset Number | decimal | 200009.01 |
Feature Type Code | character varying | LSL |
Cms Activation Date | datetime (DD/MM/YYYY HH24:MI:SS) GMT | 19/06/2022 11:37:28 |
Feat Cent East | decimal | 530981.73 |
Feat Cent North | decimal | 187613.56 |
Longitude | character varying | 51.57218637 |
Latitude | character varying | -0.12345678 |
Url | character varying | |
Defect Type | character varying | Column - Day Burner |
Defect Code | character varying | LCDB |
wkb_geometry | well-known text geometry | POINT (-0.09102906 51.58654038) |
Feature Group Code | character varying | SLCO |
Feature Group Name | character varying | SL-Street Lighting Unit |
Feature Type Name | character varying | Street Light Column |
...
Waste Assets Field Definitions
Field Name | Definition | Example |
Feature Type Name | character varying | Dog Bin |
Site Code | character varying | P421054 |
Site Name | character varying | Albany Close |
Locality Name | character varying | N15 |
Ward Name | character varying | St Anns |
Asset Number | character varying | 7,001.00 |
Feature Location | character varying | by L/C AC805H |
Central Asset Id | character varying | H5400002729 |
Material | character varying | Plastic |
Manufacturer | character varying | Unknown |
Mounting Configuration | character varying | Floor mounted |
Asset Position On Site | character varying | Back of Path |
Surface Base Type | character varying | Block Paving |
Owner | character varying | Waste Management |
Asset Id | character varying | S1 |
Survey Date | datetime (DD/MM/YYYY HH24:MI:SS) GMT | 11/24/23 0:00 |
Feat Cent East | decimal | 532047.79 |
Feat Cent North | decimal | 189119.58 |
Latitude | character varying | 51.58547145 |
Longitude | character varying | -0.09542232 |
Feature Start Date | datetime (DD/MM/YYYY HH24:MI:SS) GMT | 6/13/24 17:34 |
Feature Type Code | character varying | GRIT (for icon and/or colour coding) |
url | character varying | |
Feature Group Name | character varying | HI-Grit/Salt Bins |
Feature Group Code | character varying | GRIT |
...
The following is a Python 3 script that is used to download the data from the SFTP server. A configuration file is defined for the server details and the local location of the downloaded file.
Create the python 3 virtual environment
Ensure Python 3 is installed.
| Paste code macro |
|---|
mkdir e:\isharedata\utilities\SFTPFileDownloader cd e:\isharedata\utilities\SFTPFileDownloader c:\python3\python.exe -m venv ./venv venv\scrpts\activate.bat python -m pip install --upgrade pip pip install pysftp |
Create the SFTP file downloader script
Copy the script below and save it as E:\iShareData\Utilities\SFTPFileDownloader\sftp_file_downloader.py
| Paste code macro | ||
|---|---|---|
| ||
import sys
import configparser
import pysftp
import warnings
if __name__ == "__main__":
warnings.filterwarnings("ignore")
config = configparser.ConfigParser()
config.read(sys.argv[1])
host = config['SFTP']['host']
user = config['SFTP']['user']
password = config['SFTP']['password']
sftp_file_location = config['SFTP']['file_location']
local_file_location = config['LOCAL']['file_location']
cnopts = pysftp.CnOpts()
cnopts.hostkeys = None
with pysftp.Connection(host, username=user, password=password, cnopts=cnopts) as sftp:
try:
sftp.get(sftp_file_location, local_file_location)
except:
print(f"Download failure: {sftp_file_location}") |
Create the SFTP file downloader config file
A config file will be required for each file on the SFTP server. Adjust the following example and save under E:\iShareData\Data\<project>\sftp_<project>.config
| Paste code macro |
|---|
[SFTP] host = sftp.some.server user = some_user password = some_password file_location = path\file.csv [LOCAL] file_location = E:\iShareData\Data\path\file.csv |
Create a Program Task in Studio
Create a program task with the following details. Adjust the location of the configuration file.
| Paste code macro |
|---|
Program: E:\iShareData\Utilities\SFTPFileDownloader\venv\Scripts\python.exe
Additional Parameters: E:\iShareData\Utilities\SFTPFileDownloader\sftp_file_downloader.py E:\iShareData\Data\<project>\sftp_<project>.config |
Run the task and confirm the file has downloaded
Transfer the data to a temporary location in the Spatial Data Warehouse
Ensure that Studio supports VRT files for Spatial Data Transformations.
Inspect E:\iShareData\LIVE\config\studio\config\ConTypes.xml and check the following FileConType is defined. If not add it.
| Paste code macro |
|---|
<FileConType>
<Name>VRT File</Name>
<FormatName>VRT</FormatName>
<Direction>Both</Direction>
<FileFilter>VRT files (*.vrt)|*.vrt</FileFilter>
</FileConType> |
Create a VRT for the CSV file.
VRTs are a virtual format supported by OGR/GDAL. More information can be found here https://gdal.org/en/latest/drivers/vector/vrt.html.
Ensure that the OGRVRTLayer name matches the csv file name. It is best to use lower case and underscores and definitely no spaces for both.
The following is an example of the waste assets. There's a SrcSQL definition where is a SQL statement is present to adjust the datetime fields into a suitable format for Postgresql. The source 'table' must also match the csv file name. Note that each field name has a type e.g. string or numerical.
The geometries are defined by the GeometryType (wkbPoint), LayerSRS (EPSG:27700) and the GeometryField elements.
| Paste code macro |
|---|
<OGRVRTDataSource>
<OGRVRTLayer name="waste_assets">
<SrcDataSource>E:\iShareData\Data\<project>\waste_assets.csv</SrcDataSource>
<SrcSQL dialect="sqlite">SELECT *, (substr("Survey Date", 7, 4) || '-' || substr("Survey Date", 4, 2) || '-' || substr("Survey Date", 1, 2) || substr("Survey Date", 11, 9)) as formatted_survey_date, (substr("Feature Start Date", 7, 4) || '-' || substr("Feature Start Date", 4, 2) || '-' || substr("Feature Start Date", 1, 2) || substr("Feature Start Date", 11, 9)) as formatted_feature_start_date, replace("Feat Cent East", ",", "") as easting_decimal, replace("Feat Cent North", ",", "") as northing_decimal FROM waste_assets</SrcSQL>
<GeometryType>wkbPoint</GeometryType>
<LayerSRS>EPSG:27700</LayerSRS>
<GeometryField encoding="PointFromColumns" x="easting_decimal" y="northing_decimal"/>
<Field name="feature_type_name" src="Feature Type Name" type="String" />
<Field name="site_code" src="Site Code" type="String" />
<Field name="site_name" src="Site Name" type="String" />
<Field name="locality_name" src="Locality Name" type="String" />
<Field name="ward_name" src="Ward Name" type="String" />
<Field name="asset_number" src="Asset Number" type="String" />
<Field name="feature_location" src="Feature Location" type="String" />
<Field name="central_asset_id" src="Central Asset Id" type="String" />
<Field name="material" src="Material" type="String" />
<Field name="manufacturer" src="Manufacturer" type="String" />
<Field name="mounting_configuration" src="Mounting Configuration" type="String" />
<Field name="asset_position_on_site" src="Asset Position On Site" type="String" />
<Field name="surface_base_type" src="Surface Base Type" type="String" />
<Field name="owner" src="Owner" type="String" />
<Field name="asset_id" src="Asset Id" type="String" />
<Field name="survey_date" src="formatted_survey_date" type="DateTime" />
<Field name="feat_cent_east" src="easting_decimal" type="Real" />
<Field name="feat_cent_north" src="northing_decimal" type="Real" />
<Field name="latitude" src="Latitude" type="Real" />
<Field name="longitude" src="Longitude" type="Real" />
<Field name="feature_start_date" src="formatted_feature_start_date" type="DateTime" />
<Field name="feature_type_code" src="Feature Type Code" type="String" />
<Field name="url" src="url" type="String" />
<Field name="feature_group_name" src="Feature Group Name" type="String" />
<Field name="feature_group_code" src="Feature Group Code" type="String" />
</OGRVRTLayer>
</OGRVRTDataSource> |
It is worth testing that the VRT file has been defined correctly. You can use OGR to do this. Identify where the location of OGR by inspecting the OGR Path parameter in Studio's settings. Open a command window and navigate to that folder. Enter the following to test the VRT file. If correct OGR will report back the features found in the VRT file.
| Paste code macro |
|---|
ogrinfo.exe path\to\file.VRT |
Create the Spatial Data Transformation task
For the Source Data select 'VRT File' and enter the path to the VRT file. For output select a schema and a temporary table location for the data. Uncheck 'Projection as British National Grid' if the VRT file is not using EPSG:27700 as the SRS. If you do uncheck this option specify the destination projection by selecting 'Show expert mode' and entering the following in the Additional Parameters box.
| Paste code macro |
|---|
-t_srs "EPSG:27700" |
Run the task and confirm the data has been loaded to the destination specified by using QGIS or pgAdmin.
Check that the data has been transferred correctly
Create the Python 3 environment
| Paste code macro |
|---|
install .net 4.8
install Microsoft C++ Build Tools via https://visualstudio.microsoft.com/visual-cpp-build-tools/ --> Run exe --> Check C++ build tools --> check MSVC v142 and Windows 10 SDK --> wait for installation to finish
mkdir e:\isharedata\utilities\DataSourceChecker
cd e:\isharedata\utilities\DataSourceChecker
c:\python3\python.exe -m venv ./venv
set - this displays the environment variables
edit venv\Scripts\activate.bat
redefine PATH in activate.bat so it doesn't mention c:\python27 or ogr paths
set GDAL_DATA in activate.bat to blank if it exists as an environment variable
venv\Scripts\activate.bat
download https://github.com/cgohlke/geospatial-wheels/releases/download/v2024.9.22/GDAL-3.9.2-cp39-cp39-win_amd64.whl and copy to e:\isharedata\utilities\DataSourceChecker
pip install GDAL-3.9.2-cp39-cp39-win_amd64.whl
pip install numpy |
Create the feature_count.py script
Store the following as E:\iShareData\Utilities\DataSourceChecker\feature_count.py.
| Paste code macro | ||
|---|---|---|
| ||
import sys
import configparser
from osgeo import ogr
def count_features_in_ogr(ogr_driver, ogr_connection, ogr_layer):
# Register the VRT driver
ogr.RegisterAll()
ogr.DontUseExceptions()
# Open the datasource
if ogr_driver == 'OGR_VRT':
datasource = ogr.Open(ogr_connection)
elif ogr_driver == 'PostgreSQL':
datasource = ogr.Open(ogr_connection)
else:
driver = ogr.GetDriverByName(ogr_driver)
datasource = driver.Open(ogr_connection, 0)
if datasource is None:
print(f"Failed to open {ogr_connection}")
return 0
layerList = []
for i in datasource:
daLayer = i.GetName()
if daLayer == ogr_layer:
feature_count = i.GetFeatureCount()
return feature_count
def list_ogr_drivers():
cnt = ogr.GetDriverCount()
formatsList = [] # Empty List
for i in range(cnt):
driver = ogr.GetDriver(i)
driverName = driver.GetName()
if not driverName in formatsList:
formatsList.append(driverName)
formatsList.sort() # Sorting the messy list of ogr drivers
for i in formatsList:
print(i)
return 0
if __name__ == "__main__":
config = configparser.ConfigParser()
config.read(sys.argv[1])
source_driver = config['SOURCE']['source_driver']
source_connection = config['SOURCE']['source_connection']
source_layer = config['SOURCE']['source_layer']
source_feature_count = count_features_in_ogr(source_driver,source_connection,source_layer)
destination_driver = config['DESTINATION']['destination_driver']
destination_connection = config['DESTINATION']['destination_connection']
destination_layer = config['DESTINATION']['destination_layer']
destination_feature_count = count_features_in_ogr(destination_driver,destination_connection,destination_layer)
if source_feature_count != destination_feature_count:
print(f"feature counts do not match")
|
Each dataset that is being imported will need a configuration file. Edit accordinly to suit the source & destination dataset. Here we wish to check that the VRT file contains the same number of features as the version held by the Spatial Data Warehouse. Store the file as E:\iShareData\Data\<project>\feature_count_<project>.config.
| Paste code macro |
|---|
[SOURCE] source_driver = OGR_VRT source_connection = E:\iShareData\Data\path\file.vrt source_layer = the_layer [DESTINATION] destination_driver = PostgreSQL destination_connection = PG: host=<SDW host> user=isharedata_gis dbname=sdw password=<password> port=5432 destination_layer = the_schema.the_layer_temp |
Create a Program Task in Studio to run the script
Create a program task with the following details. Adjust the location of the configuration file.
| Paste code macro |
|---|
Program: E:\iShareData\Utilities\DataSourceChecker\venv\Scripts\python.exe
Additional Parameters: E:\iShareData\Utilities\DataSourceChecker\feature_count.py E:\iShareData\Data\<project>\feature_count_<project>.config |
Run the task and confirm the two datasets have the same number of features
Transfer the data to the 'live' location
Now we are confident that the data has been downloaded and transferred to the Spatial Data Warehouse correctly we can copy the earlier Spatial Data Transformation task and adjust the destination table so that it in the correct location for any mapsources wanting to use the data.
Create a mapsource to use the data
Now that the data is loaded into the Spatial Data Warehouse we need to create a mapsource and layer that uses the data. If any classification styling is required it is recommended to load the data into QGIS. Choose categorised styling and export the style SLD file. This can be imported into iShare. Ensure the field that is used for categorised styling is exposed as an attribute field.