Public Sector Plan members should use the following instructions to download the products from the OS Data Hub:

1. Sign in to your PSGA organisation account on OS Data Hub.

2. Click Download on the top navigation bar. You will be taken to the Premium Data Downloads catalogue.

3. Search and select the premium product you want.

4. Click the Add data package button.

5. Enter a data package name. We suggest using a name that will help you easily identify the data package later. Data package names don’t have to be unique.

   A data package is a downloadable file or group of files containing the product data you want to receive. There can be more than one version of a data package when new product updates are available to download. To download (order) data, you create a data package in the OS Data Hub for the product you want, and specify the area, format and update cycle. Once the system has fulfilled the data package, you can download it from the data package list, under Data packages on the left menu in your OS Data Hub account.

6. Choose your area. For user-defined areas, draw your area of interest (AOI) using the drawing tools or upload a polygon to the map from the polygon library.

7. Select file format. We show only the available file formats for that selection.

8. Select updates. We show only the available updates for that selection.

9. Click the Create data package button.

10. The confirmation page tells you the data package is being created, with options to Add another data package or click to view the data package status in the data package list. You will also receive a confirmation email when a data package is created.

11. The data package list displays all the data packages for your organisation. Once a data package has been fulfilled, the status will change from Available soon to Download. You will also receive an email when the data package is ready to download.

12. Click the Download button to open the data package and download the zip file(s).

GML data can be viewed and loaded into Safe FME. When AddressBase Premium is read into FME, the non-spatial tables within the product are automatically related to the spatial tables.

Note: The following instructions use FME version 2020.0, but should work in most versions.

1. Open FME Workbench.

2. Click the Reader button in the ribbon.
3. An Add Reader dialog box will appear.
4. Select GML (Geography Markup Language) as the Format.

5. Select a coordinate reference system from the Coord. System dropdown list. It’s recommended that British National Grid (EPSG:27700) is selected as the coordinate reference system; however, you can select a different coordinate reference system if required.

6. Click the three dots button next to the Dataset box. This will bring up another dialog box where you can select the GML dataset to read into FME.

7. Click OK and OK again.

8. A Select Feature Types dialog box will appear in which you can select what feature types should be read into FME. The BasicLandPropertyUnit and Street feature types also include the non-spatial table information, such as LPI and organisation.

9. Click OK.

10. The three feature types should appear as readers.

is available on the Safe Software website.

Note - The following instructions are based on PostgreSQL version 12.3 and assume that you have set-up your database with the PostGIS spatial extension in a Windows environment.

Requirements:

• A development platform for PostgreSQL (for example, pgAdmin or dBeaver)

• A PostgreSQL database

• PostGIS extension

• GDAL extension

• Access to a Command Prompt or similar

• A basic understanding of database terminology

Instructions

Note - Before you start, make sure the AddressBase Premium GML is stored in a suitable location and the file is unzipped.

1. Open the Command Prompt by clicking the Windows Start button in the bottom left-hand side of the screen and typing cmd into the search bar.
2. The Command Prompt will appear.

3. Change the directory of the Command Prompt App if necessary. The directory needs to point to the folder where the GML is stored. For this example, the GML to be loaded into PostgreSQL is stored in a C:\Temp folder.

4. In the Command Prompt, type cd followed by the directory of the location of the GML. Press Enter on the keyboard. This will change the directory.
5. Enter the following command:

   Where: <host>, <port>, <database>, <username> and <password> can be found within the subsequent database. <data_name.gml> is the AddressBase Premium GML file, for example, AddressBasePremium_AOI_2021-04- 12_001_gml.gml. <CRS> is the coordinate reference system. British National Grid (27700) is recommended but you can use alternative coordinate reference systems if required. Example:

6. Open your chosen development platform by going to the Windows Start button and selecting your chosen software.

7. In your Database Navigator, move to the database you loaded data into and expand the schema.

8. From using the code above, the GML files should be in your default schema (shown in bold). The data should load into two tables: basiclandpropertyunit and street. Data in non-spatial tables will automatically be matched to the spatial tables based on a unique identifier.

9. Once loaded, you may want to add Primary and Foreign Keys to the data. These can only be added on columns where the data values are unique. Where there are no unique data values, an index may be added which will aid searching. For the BLPU table, the UPRN provides a unique value.

Note - You can also manage your database using the PostGRES SQL Command Prompt.

The following instructions show you how to use Gawk to split the AddressBase Premium CSV files and append the Header files.

1. Group all the AddressBase Premium CSV files into an empty folder. It is very important to ensure that the folder does not contain other CSV files. The folder must contain no spaces in its file directory path name, for example, C:\AddressBaseData\AddressBase_Premium.

2. Download AddressBase Premium header files (zip) in the section below.

3. Extract the downloaded zip and you should have the following CSV files: Record_10_HEADER_Header.csv Record_11_STREET_Header.csv Record_15_STREETDESCRIPTOR_Header.csv Record_21_BLPU_Header.csv Record_23_XREF_Header.csv Record_24_LPI_Header.csv Record_28_DELIVERYPOINTADDRESS_Header.csv Record_29_METADATA_Header.csv Record_30_SUCCESSOR_Header.csv Record_31_ORGANISATION_Header.csv Record_32_CLASSIFICATION_Header.csv Record_99_TRAILER_Header.csv

4. Place the extracted AddressBase Premium Header files into the folder with the CSV files (step 1).

5. Go to the .

6. Open the AddressBasePremium_GawkSplitScript.bat file and copy the contents to a text editor such as TextPad or NotePad++. Save this text file as a .bat file in the same folder as your AddressBase Premium data and Header files.

7. In the same link in Step 5, download and extract the zip file called gawk-4.0.2-bin.zip.

8. This will extract a file called Gawk.exe. Place this file in the same folder as your AddressBase Premium data and Header files.

9. Double-click AddressBasePremium_GawkSplitScript.bat file and an MS-DOS window will appear. Once the process is complete, the window will close automatically, or you will have to press any key to continue.

   Running the .bat file creates temporary files and requires extra space in the location you are creating your files. These files can be much larger than the original CSV files. They are deleted once the process has finished, but the space is still required.

10. Running this should create additional files which will have a similar naming convention to that of the Header files. These new files should have generated in the same location as the data and headers. For example, looking in the location there should now be an ID24_LPI_RECORDS as well as a Record_24_LPI_Header.csv.

The following instructions show you how to use Python instead of Gawk to split the AddressBase Premium CSV files and append the Header files. These instructions are based on Python 2.7.

1. Group the AddressBase Premium CSV files into an empty folder. It is very important to ensure that the folder does not contain other CSV files. The folder must contain no spaces in its file directory path name, for example, C:\AddressBaseData\AddressBase_Premium.

   You do not need to download the Header files if you are using Python to split the AddressBase Premium data.

2. Go to the .

3. Open the AddressBasePremium_RecordSplitter_v2_7.py file and copy the contents to a text editor such as TextPad or NotePad++. Save that text file as a .py file to the same folder as your AddressBase Premium data.

4. Open a Command Prompt window by going to Windows Start > CMD prompt.

5. Within the Command Prompt, type cd and the directory path where you just placed the AddressBasePremium_RecordSplitter_v2_7.py file, then hit Enter on your keyboard.

6. Type the name AddressBasePremium_RecordSplitter_v2_7.py into the Command Prompt or select it using the Tab key, which will display each file within the file directory in turn. Once selected, hit the Enter key.
7. The following message should be displayed: This program will split OS AddressBase Premium Zip CSV or extracted CSV files by record identifier into new CSV files. Please type in the full path to the directory of OS AddressBase zip files: Directory Path:

8. Enter the full directory path to where you stored the AddressBase Premium data (for example, C:\AddressBaseData\AddressBase_Premium). Hit Enter on your keyboard. The process of splitting the files will then begin.

9. When the file splitting process is complete, the following message will be displayed: The program will close in 10 seconds. You may still need to close the Command Prompt window afterwards by typing Exit and hitting Enter.

10. If you navigate to the folder which contained the .py file and the AddressBase Premium CSV files, you should find new files which will have a similar naming convention to that of the Header files. These new files will contain all of your AddressBase Premium data split out by record type.

There are two methods for loading AddressBase Premium CSV data into ArcGIS Pro. We will only step through the first method in this guide. If you need guidance on the second method, please refer to the instructions in :

• The first method uses the UK Data Loader published by ESRI, which loads the data into a File Geodatabase and automatically relates between the different tables. At the end of the process, you have a fully working dataset ready for use. The step-by-step instructions in this section use this method.

• The second method must have the CSV data prepared as described in . These files are imported into a project File Geodatabase within ArcGIS Pro (see the instructions in . Only the Basic Land and Property Unit (BLPU) table and Streets table will contain geometry which can be mapped by adding XY data. All the other tables will have no geometry, but these tables can be linked together using the Relates function within ArcGIS Pro. Please refer to the AddressBase Premium structure and AddressBase Premium Islands structure <link> in the technical specifications for information on which table fields should be linked together.

Note - The following instructions are based on ArcGIS Pro version 2.3.3.

Note - These instructions assume that you have the full data Interoperability Extension installed and that it matches your ArcGIS Pro version.

1. Launch ArcGIS Pro and create a new project.

2. Select a location to save the project to, name the project and click OK. A new map project will launch with a default map backdrop supplied by ESRI. You can change this backdrop to another ESRI backdrop or use your own.

3. Select the Catalog pane on the right-hand-side of the screen.

4. Under the Project list, right-click Toolboxes and select Add Toolbox.
5. Navigate to where the UK Data Loader toolbox is held on your system, for example, C:\Data\ESRI_UK. You should obtain the latest version of the UK Data Loader toolbox from ESRI.

6. Select the UK Data Loader toolbox (.tbx) and click OK.
7. The UK Data Loader Toolbox (UKDataLoader.tbx) will now appear in the Catalog list of available toolboxes. Expand it by clicking on the arrow to the left of the entry, then click the arrow again for the OS AddressBase entry.

8. Double-click Load OS AddressBase Data script that should be listed.
9. A new Geoprocessing window will appear. There are some mandatory blank fields which you need to complete before the script can be run. For the Source Folder, enter the directory path where AddressBase Premium data is located. Select your Address Base Product from the dropdown menu, then enter a location for the Destination Workspace and Log File Folder. Other fields are optional.

10. When the required fields are completed, click the Run button at the bottom of the Geoprocessing window.
11. This can run for some time, depending upon the amount of data being processed. Wait until a message appears saying the files have been imported successfully.

12. When the import has completed, click Add Data in the main menu, then navigate to the project File Geodatabase.
13. You will see that some of the feature classes have no geometry, whilst others do. Select those required and click OK. In the example below, ABP_TestADPRB refers to geometry created from the BLPU table and ABP_TestADPRS refers to geometry created from the Street table.
14. The data will be added as new features classes in the Contents window. The feature classes containing geometry will be displayed against the map backdrop in the Map window.

You have now successfully loaded AddressBase Premium data into ArcGIS Pro.

Note - The following instructions are based on ArcGIS Desktop version 10.

1. Start ArcCatalog as a separate program, or within ArcMap if you are using version 10.
2. Connect to the folder where the AddressBase Premium data is located (for example, C:\AddressBasePremium_Data):

   • Click File, or in version 10, select Folder Connections.

   • Click Connect Folder in the top ribbon or right-click on Folder Connections > Connect To Folder, then navigate to the folder containing the data.
   • From the main window, select the folder to connect to and click OK.

3. The folder should now appear in the navigation window to the left of the screen, or within your Catalog window if you have opened it within ArcGIS Map.
4. Create a File Geodatabase (.gdb) to store the address data. Using the file tree, go to Folder Connections and navigate to the directory where you wish to create the File Geodatabase, for example, C:\AddressBase_Geodatabase\AddressBase_Premium. This may need to be set up as a new connection.

5. Right-click on the folder where you wish to store the File Geodatabase and select New > File Geodatabase.

6. It will be created and named by default as New File Geodatabase. Rename it to a name of your choice.

7. Right-click on your new File Geodatabase and select Import > Table (multiple)…

8. For Input Table, navigate to the location of the CSV files that you wish to open, that is, the folder that contains the AddressBase Premium data split into individual files by record type.

9. Select the files that you wish to add. Make sure you add the data files (which have the following naming convention: ID21_BLPU_Records.csv) and not the Header files (which have the following naming convention: Record_21_BLPU_Header.csv).

10. Click Add.

11. The Output Geodatabase option should automatically be populated by the location of the File Geodatabase that is to be updated; this should be the File Geodatabase you created in steps 4 to 6.

12. Click OK, and once the process is complete click Close. Note: The process may take some time, depending on the number of files and the amount of data being added. The window may close before the operation is complete, so if you cannot see all of your expected files under your File Geodatabase, this may mean that the data is still being loaded.
13. To create a map of the locations of the AddressBase Premium records, they need to be geocoded. To do this, double-click on the Geodatabase that the data was just imported into. The BLPU table and the Streets table are the only tables in AddressBase Premium that carry geocoded information.

14. Right-click on the Geodatabase table that was created from AddressBase Premium records with a record type of 21 or 15 (for example, ID21_BLPU_Records.csv or ID15_StreetDesc_Records.csv) and select Create Feature Class.

15. In the Create Feature Class From XY Table dialog box, you can use the dropdowns to change the X Field to either X_COORDINATE or Longitude, and the Y Field to Y_COORDINATE or Latitude. Leave the Z Field as <None>.
16. For Input Coordinates, click on Coordinate System of Input Coordinates > Select and then navigate to Projected Co-ordinate Systems > National Grids > Europe > British National Grid. Note - If you selected X and Y as 'Longitude' and 'Latitude' in the step before, then you need to select 'ETRS89 [EPSG: 4258]' for the coordinate system instead.

17. Double-click on the selected coordinate system, then click Apply and OK.

18. Click on the folder icon alongside the Output field and navigate to the location where you wish to save the output shapefile or feature class (we recommend that this be within the Geodatabase you created in steps 4 to 6). Note - If you can’t see your Geodatabase, ensure the 'Save as type' box at the bottom of the dialog box is set to 'File and Personal Geodatabase feature classes'.

19. Give the file a suitable name (for example, XYID21_BLPU_Records).

20. Click Save, then leave the Configuration keyword as DEFAULTS and click OK. Note: You may need to right-click on the Personal Geodatabase where it was saved and select 'Refresh' in order to see your new feature class.

    Now the processing has been done, the data needs to be loaded into ArcMap so that the individual tables can be related.

21. Start ArcMap if you have not been working within ArcMap version 10 already.

22. Select File > Add Data…

23. Navigate to the folder where the Geodatabase was created.

24. Double-click on the Geodatabase and select all the files inside.

25. Click Add.
26. • BLPU (ID21_BLPU_Records):

      • UPRN – Application Cross Reference (ID23_XREF_Records) UPRN

      • UPRN – LPI (ID24_LPI_Records) UPRN

      • UPRN – Delivery Point Address (ID28_DPA_Records) UPRN

      • UPRN – Organisation (ID31_Org_Records) UPRN

      • UPRN – Classification (ID32_Class_Records) UPRN

    • LPI (ID24_LPI_Records):

      • USRN – Street (ID11_Street_Records) USRN

    • Street (ID11_Street_Records):

      • USRN – Street Descriptor (ID15_StreetDesc_Records) USRN

27. To do this, select the Source tab in the left-hand navigation window and right-click on the first table that you wish to relate to another. To create the relevant relates, use the following instructions:

    • Click Joins and Relates > Relate...

    • From the first dropdown menu, select the attribute from the first table that will be used to create the relate between the two tables. (Apply the relationships as listed in step 26.)

    • From the second dropdown, select the table that is going to be related to. (Apply the relationships as listed in step 26.)

    • From the third dropdown, select the attribute from the table that is being related to. (Apply the relationships as listed in step 26.)

    • In the fourth dropdown, input a relevant name for the relate (for example, BLPU_to_LPI).
    • Click OK.

    • Repeat this process for all of the joins / relates listed in step 26.

28. Once the data has been loaded into ArcMap, you may wish to display more relevant information in the Info tool than the Esri-defined Object ID. To change this, use the following instructions:

    • Double-click on the spatial dataset that you wish to change the Display Expression of.

    • Select the Display tab.

    • Change the Field to the desired field (for example, UPRN).

    • Click OK.

Note - The following instructions are based on MapInfo Professional version 16.0.

Note - MapInfo has a size limit of 2 Gb on each table. This equates to a maximum number of approximately 4 million AddressBase Premium records.

1. Launch MapInfo.

2. Click Home > Open Table and navigate to the folder that contains the AddressBase Premium data.
3. In the Files of Type dropdown menu, select Comma delimited CSV (*.csv), then click on the AddressBase Premium data to be loaded. Click Open.
4. In the next window, tick the box Use First Line for Column Titles and select Windows US & W. Europe (“ANSI”) from the dropdown menu for File Character Set. Click OK.

   Note: When adding data this way, the field type classifications and field sizes of each column automatically try to fit the type of data that MapInfo believes is contained within the column and the largest value of that classification found within that column. This means the classifications and field sizes of some attributes may not match the field types and sizes stated in the Technical Specification. The following six instructions detail how to change these columns to match those values.

5. Select Home > Save > Save Copy As and select the AddressBase table that was loaded in. Click Save As… and name the table to be created, then click Save.
6. Open the table that was just created via Home > Open. Navigate to and select the copy of the table you just named. Click Open.

7. Navigate to Table > Maintenance > Table > Modify Structure and select the table to be edited. Click OK.
9. Type and Width should be changed for all attributes, apart from the following due to software-specific dependencies:

   • UPRN, which should be classified as Float

   • All attributes that have a Field Type of Date in the Technical Specification, which should be classified as Character with a length of 10
10. After all changes have been made, click OK.

11. To create a map of the location of the AddressBase Premium records, they need to be geocoded. Ensure the table of records that you wish to geocode is open, then navigate to Spatial > Create Points.
12. Select the table you wish to geocode from the Create Points for Table dropdown menu.

13. Expand the Get X Coordinates from Column dropdown menu and select either X_Coordinate or Longitude.

14. Expand the Get Y Coordinates from Column dropdown menu and select either Y_Coordinate or Latitude.

15. Click on the Projection button, then select the British Coordinate Systems option from the Category dropdown menu. Select British National Grid [EPSG: 27700], or if you selected Longitude and Latitude in the steps above, then you should select ETRS89 [EPSG: 4258].
16. Click OK to close that window and OK again to close the next window.

17. Finally, click Map > Map (New Map Window) to view the loaded geocoded points.

Note - The following instructions are based on QGIS version 3.1.

1. Launch QGIS and click Settings > Options.

2. Select CRS from the left-hand menu and check the coordinate reference system (CRS) is set to British National Grid within the Default CRS for new projects section and the CRS for new layers section.

   If the coordinate reference system is not already set, click the Select… button in the bottom right-hand corner of each section. A new dialog box will appear. In the Coordinate Reference System Selector dialog box, type 27700 into the Filter box to find and select British National Grid. Alternatively, if you intend to use Latitude and Longitude columns, select ETRS89 [EPSG: 4258]. Click OK to close the Coordinate Reference System Selector dialog box.
3. Click OK to close the Options CRS dialog box.

4. • BLPU (ID21_BLPU_Records):

     • UPRN – Application Cross Reference (ID23_XREF_Records) UPRN

     • UPRN – LPI (ID24_LPI_Records) UPRN

     • UPRN – Delivery Point Address (ID28_DPA_Records) UPRN

     • UPRN – Organisation (ID31_Org_Records) UPRN

     • UPRN – Classification (ID32_Class_Records) UPRN

   • LPI (ID24_LPI_Records):

     • USRN – Street (ID11_Street_Records) USRN

   • Street (ID11_Street_Records):

     • USRN – Street Descriptor (ID15_StreetDesc_Records) USRN

5. Back in the QGIS window, in the top ribbon, select Layer > Add Layer > Add Delimited Text Layer. The Data Source Manager – Delimited Text dialog box will appear.

   Note - The following steps explain how the joins are made for the first relationship given in the list above, that is, between the BLPU (ID21_BLPU_Records) and the Application Cross Reference (ID23_XREF_Records). This process will need to be repeated for all subsequent joins you make.

7. In the Layer name box, you can keep the default layer name or change it to a new one.

8. Ensure the First record has field names option is ticked.

9. Ensure the Detect field types option is ticked.

10. For Geometry Definition, select Point coordinates.

11. You should now be able to select the X_Coordinate field for the X Field dropdown and the Y_Coordinate for the Y Field dropdown if this was not done automatically. Alternatively, if you wish to use the Latitude and Longitude columns, Longitude needs to be inserted into X field dropdown, and Latitude needs to be inserted into the Y field dropdown.
12. Click Add. So far, you have loaded the BLPU record information.

13. In the QGIS window, in the top ribbon, select Layer > Add Layer > Add Delimited Text Layer. The Data Source Manager – Delimited Text dialog box will appear.
15. In the Layer name box, you can keep the default layer name or change it to a new one.

16. Ensure the First record has field names option is ticked.

17. Ensure the Detect field types option is ticked.

18. For Geometry definition, choose No Geometry (attribute only table).

19. Click Add. This table will now be added to QGIS, but it will not be viewable as we have added it as an attribute table only.

20. Next, right-click on your BLPU layer and select Properties.

21. Go to the Joins tab found on the left-hand side.
22. Click the green plus button in the bottom left-hand corner.

23. Select your Join layer. For this example, it is: Application Cross reference (ID23_XREF_RECORDS).

24. Select the Join field. For all BLPU links, this will be the UPRN (see the joins listed in step 4).

25. The Target field will also be the UPRN for this example.

26. Click OK. You should now have a join listed in your Joins window.

27. Click OK to return to your main QGIS mapping screen.

28. If you now select one of your BLPU records in the main mapping window, you will see the BLPU attributes and the relevant Application Cross References for that record.
29. You can now repeat steps 4 to 28 for all the additional joins (listed in step 4) you wish to make.

Loading CSV using CadCorp Address Loader

These instructions show you how to load AddressBase Premium CSV into a database using the CadCorp Address Loader. Note - The CadCorp Address Loader requires a separate license to the CadCorp SIS Desktop version.

1. Open CadCorp Address Loader. A dialog box will appear:
2. Click the Browse button and navigate to the location of the AddressBase Premium CSV file (zipped or unzipped) that you want to load into CadCorp.

3. In the Workspace Folder section, select a folder in which to store the log file and unzipped data. This can be a folder of your choosing or the application data folder.

4. Click Next.

5. Another dialog box will appear where you can select which address files to load. You can choose to load all or selected files.
6. Click Next.

7. The next dialog box to appear will allow you to select a database into which you can save the AddressBase Premium data. Select your database of choice, then click Next.

   Note - The following instructions will relate to a PostGIS database.

8. In the next dialog box, connect to an existing database by entering information into the following fields:

   • Host

   • Port Number

   • Database

   • User name

   • Password Note - These details can be found within the parameters of your database. It is not possible to create a database via this method, and one will have had to have been created prior to this stage.
9. Click Next.

10. A schema dialog box will appear showing the tables that the data will be loaded into:
11. At this stage, it is possible to create a gazetteer view or an interface table where a concatenated LPI address field can be built.

12. Click Next in that dialog box, then Start in the following dialog box to begin loading the data:
13. Once finished, the tables will be loaded into the database.

Loading CSV into CadCorp SIS Desktop via a database

1. Open a new or existing CadCorp SIS Desktop project.

2. In the top ribbon, select Add Overlay.
3. On the left, select Databases, then select your database of choice from the list on the right (Note: A PostGIS database will be referred to in the instructions from this point on).
4. Click Next.

5. Connect to your existing database. To connect to a saved database, select a database under Saved connections. Alternatively, you can manually enter the database details. Click Next.
6. In the next dialog box, select the type of database connection you want to make. The Simple connection option is effective in this instance. Click Next.
7. Select the tables to load into CadCorp, then click Finish.

The following instructions describe how to load AddressBase Premium into a PostgreSQL database using the text files created using the CSV file merge utility. The instructions are based on PostgreSQL version 1.12.3 and assume that you have set-up your database with the PostGIS spatial extension. It is recommended that you have a basic understanding of database terminology before following these instructions.

2. Check that there are no carriage returns (extra rows) at the end of the CSV output file as this will result in errors. To check this, open the CSV file and hit End on your keyboard. Your cursor should now be at the end of the last line, and not on any extra line below. If it is on the line below, hit Delete to remove the extra empty row.

3. Open the PGAdmin tool (this can be found on the Windows Start Menu > pgAdmin).

4. On the left-hand side, under Browser, you have the option to connect to either your existing databases or a new one. To connect to a new database, right-click on Databases, then select Create > Database…. It is recommended that the encoding is set to UTF-8.

5. Open the public schema (although in a production environment, it is advised to use a different schema) and create the tables using the following steps:

   • Open the SQL query tool.

   • This SQL file can be opened in a text editor, and the SQL scripts within can be copied and pasted into the SQL query tool within PostgreSQL. To open a new SQL file, select Tools in the ribbon, then select Query Tool from the dropdown.
6. Copy and paste the SQL code from the GitHub link given in step 5 into the SQL Query Tool (as shown in the following screenshot).
7. The following tables should be created (you can alter table names as you wish):

   • BLPU

   • Classification

   • Cross reference

   • Delivery Point Address

   • LPI

   • Organisation

   • Streets

   • Street Descriptor

   • Successor Records

8. Once the table has been created, the data can be loaded into each table using SQL COPY. Adding the CSV option as the first line contains a header record for each table. Note - The path and file name may need to be changed to reflect your data:
9. Once loaded, you may want to add Primary and Foreign Keys to the data. These can only be added on columns where the data values are unique. Where there are no unique data values, an index may be added which will aid searching. For the BLPU table, the UPRN provides a unique value. Primary Keys are added using the following steps:

   • Right-click on the table name and select Properties.

   • Select the Constraints tab.

   • Click the + symbol to add a new Primary Key.

   • Enter a name to call the key under the General tab (for example, Key1).

   • Under the Definition tab, select UPRN or any other unique value from the dropdown under columns.

   • Click Save.

10. Repeat the procedure for the Streets table and USRN. However, in the other tables these columns may contain duplicate values. In this case, use the table key (for example, LPI_Key as the primary identifier). Alternative Object Identifiers (OID) can be added to each table (these are also required to use the data in some GIS, including QGIS and MapInfo.) The following SQL can be used for this:

11. • Right-click on the table name and select Properties.

    • Select the Constraints tab.

    • Click the + symbol to add a new Foreign Key.

    • Enter a name to call the key under the General tab (for example, Key1).

    • Under the Definition tab, select UPRN or any other unique value from the dropdown under columns.

    • Click Save.

12. You can also index the data by using the following steps:

    • Right-click on the table name and select Create > Index.

    • Under the General tab, enter a name (for example, Idx1).

    • Under Definition tab > Columns, click the + symbol.

    • Select the UPRN for example, or any other unique value.

    • Click Save.

Converting coordinates to geometry

A PostGIS extension is required to create geometries. The AddressBase Premium products contain both British National Grid (BNG) and ETRS89 coordinates. The SQL below shows how to create a column for BNG, but it can be altered to utilise the ETRS89 data.

1. Add a geometry column to make the data usable in a GIS:

2. Load the data into your new geometry column:

   This sets the geom column in the BLPU table to equal the values from the X_coordinate and Y_coordinate columns, with the spatial reference defined as 27700.

3. Create a spatial index on the data using:

   This adds the index name idx_abp_geom to the same table on the geom column.

Note - The following instructions assume a basic knowledge of Oracle databases and SQLLDR (the package used to load CSV files into the database). Other options are available for loading data into Oracle databases.

Note - When using SQLLDR, it is not necessary to merge all the AddressBase files into a single file as it can load the data directly from the provided file as long as it has been unzipped first.

Note - The following steps describe one method for loading a full supply of the data. Sections of text within <…> (a pair of less than and greater than symbols) denote where changes will need to be made to accommodate local file naming.

1. Copy the data files from the disk to an appropriate location. It is worth noting that the files will need to be unzipped; therefore, you will need in the region of 40 Gb of free space to load AddressBase Premium and 1 Gb of free space to load AddressBase Premium Islands.

2. Once the data is copied, the next stage is to unzip the *.zip files to extract the *.csv files. This can be done using a package such as Winzip or 7Zip. Please see the Data supply page for more information.

3. Now that all the files are unzipped, the latter stages will be easier if you create a file listing all the CSV files to be loaded. This can be done using a batch file that writes all the files out to a text file. Copy the following into a text editor and save it as a .bat file in the same directory as the AddressBase Premium data:

   This file will form the basis for loading the control file in a later step.

5. Open the AddressBase_Premium_and_Islands folder. Open the file ending CreateTables.sql in a text editor.

6. Within that SQL there are references to <TablespaceName>, which need to be changed to the tablespace name that you are working in. When these are changed, copy and paste the SQL into Oracle to create the tables.

7. Next, a SQLLDR control file needs to be created. An example of one of these files is Oracle_AddressBasePremium_Control.ctl, provided in the AddressBase_Premium_and_Islands folder in the OS GitHub repository linked in step 4. Open this file in a text editor.

8. Populate the INFILE lines with the file listing that was created in step 3; use one INFILE command for each file. This tells the process to open each of the files and carry out the other tasks listed below it.

9. The rest of the file tells the tool how to interpret the files that it is reading in. The INTO statement at the top of each of the tables tests the first column (01) of the row in the file that it is looking at. If it meets the criteria, the structure of the table that the line is to be loaded into is described below it. Save the completed file with the extension *.ctl.

10. Once this file is created, it can be called from a .bat file to run it on the box that holds the database rather than a remote machine. If you wish to run it from a remote machine, contact your Oracle Administrator who will be able to advise on the best way to do this within your environment. The contents of the .bat file should be similar to the following:

11. When the .bat file has been run, the data is loaded. Errors with the load and/or any records that do not meet the expected structure are recorded in the and *.log and *.bad files, respectively. These are written out to the same drive location as the control file that is being used to load the data. It is strongly recommended that the log file is checked once the load is completed to verify that all of the data has loaded correctly before continuing.

12. After loading, the indexes need to be built in order to be able to carry out spatial queries and other queries where the relationship between the tables need to be built. For example, in order to return all the Delivery Point Addresses within a county, there needs to be a spatial index on the BLPU table which contains the geometry, as well as the UPRN in both the BLPU and Delivery Point Address tables. The SQL statements to create the indexes can be found in the GitHub repository link that was referenced in step 4.

13. Again, you can copy and paste the SQL statements from a text editor into Oracle in order to create the rest of the indexes. The table name provided may be different to yours and therefore might need changing before use.

14. Once the indexes are complete, the data loading process is complete and the data is ready to use.

Note - There are many ways to load AddressBase products into Microsoft SQL Server; this is just one suggested method for guidance.

1. Open the SQL Server Management Studio (SSMS).

2. Right-click on the database you are loading into and select Properties.

3. Select Options on the left-hand side menu.

4. Expand the dropdown box for Recovery model and select Bulk-logged. This minimises the logfile size, otherwise the default logging for Microsoft SQL Server can cause logfiles to grow over 20 Gb which, in turn, can cause issues with loading. Click OK.
5. Open the SQL Server Management Studio (SSMS) and right-click your database from the left-hand panel.

6. Navigate to Tasks and click Import Data. This will open the SQL Server Import and Export Wizard.

7. Click Next.

8. On the next screen, change your Data Source to Flat File Source.

9. Use the Browse button to select your CSV file. Each table needs to be added separately (for example, ID21_BLPU_Records.csv). If you cannot see your files, ensure that the bottom right dropdown box has 'CSV files (*.csv)' selected.

10. Click Open.

12. Check that the Text Qualifier is set to a double quote (“). This ensures the quotations in the raw data supply are removed upon loading, but that the data remains intact.

13. On the left-hand side of this screen, select Columns and check that the Column delimiter is set to Comma.

14. On the left-hand side of the screen, select Advanced.

16. Once you have changed the column types to match the technical specification, click Next.

17. Check that your table is going to be imported into the correct database and click Next.

18. On this screen, you can edit the default table name that Microsoft SQL Server has chosen by clicking in the destination box. For example, for AddressBase Premium, renaming to [dbo].[BLPU_TABLE].

19. Select Edit Mappings in the bottom right-hand corner.

21. Click Next. On this screen, you can check that the Source column and Destination column are correct.

22. Click Next. A summary of your import will appear. If you want to continue, click Finish.

23. A report will be generated as your data is imported. Success should appear at the top once complete.

24. You may need to right-click on your database and click Refresh to see your new table listed.

Setting Primary and Foreign Keys

To create a Primary Key field you can run an SQL statement, such as the following example.

Note - the columns you are creating these constraints on cannot be null or allowed to be null.

Primary Key

Foreign Key

Creating the point geometry

You can also create point geometry using the X and Y coordinates or the Latitude and Longitude coordinate values. This is achieved by running the following SQL statement:

Note - This is using British National Grid coordinates, with 27700 representing the spatial reference of the data. To use the Latitude and Longitude coordinate, the spatial reference should be set to 4258 for ETRS89.

The following instructions are based on ArcGIS Pro version 2.7.1. It is recommended that you have a basic understanding of ArcGIS Pro tools and terminology before following these instructions.

Note: The file must have the file extension .gpkg (not case sensitive) to recognise it as an OGC GKPG. Make sure the AddressBase Premium GeoPackage is stored in a suitable location and the file is unzipped.

1. Launch ArcGIS Pro.

2. To establish a connection to the folder where the GKPG file is stored, navigate to the Insert tab and click Add Folder. Find and select the folder, then click OK.
3. Navigate to the View tab and open the Catalog Pane. In the pane that appears, find the now-connected folder where the GKPG is stored, then expand the GKPG to show the available layers.
4. Right-click the main.blpu features class and click Add To Current Map (or Add To New if a map frame has not yet been created for this project). Repeat this step for all of the layers in the GKPG.
5. Note that the other tables can be added to the map frame, but they will only act as standalone tables as they have no spatial element to visualise. The BLPU and Streets tables will be visible in the current Map view as they contain geometry. Once all of the tables have been added to the map frame, it should look similar to the following screenshot:
6. These standalone tables can be related to the spatial tables in order to give them a spatial component. The following relates are required:

   • BLPU (ID21_BLPU_Records):

     • UPRN – Application Cross Reference (ID23_XREF_Records) UPRN

     • UPRN – LPI (ID24_LPI_Records) UPRN

     • UPRN – Delivery Point Address (ID28_DPA_Records) UPRN

     • UPRN – Organisation (ID31_Org_Records) UPRN

     • UPRN – Classification (ID32_Class_Records) UPRN

   • LPI (ID24_LPI_Records):

     • USRN – Street (ID11_Street_Records) USRN

   • Street (ID11_Street_Records):

     • USRN – Street Descriptor (ID15_StreetDesc_Records) USRN

7. To perform a relate, right-click on the main.blpu feature class, then click Joins and Relates > Add Relate.
8. In the Add Relate dialog box, change the parameters as below:

   • Layer Name or Table View: The table the relate is being added to (for example, main.blpu).

   • Relate Table: The table the relate is being taken from (for example, main.application.xref).

   • Input Relate Field and Output Relate Field: The field to relate the two tables on (for example, uprn).

   • Relate Name: Something simple that reflects the tables being related (for example, blpu to application_xref).
9. Once the parameters are set, click OK.

10. Repeat steps 7, 8 and 9 for all the relates listed in step 6. You may wish to confirm the relates have been successful by right-clicking on a layer and navigating to Properties > Relates.
11. Once the data has been loaded into ArcGIS, you may wish to display more relevant information in the Explore tool than the Esri-defined Object ID. To change this, use the following instructions:

    • Click on the layer that you wish to change the label field of.

    • Select the Labeling tab (under Feature Layer).

    • Change the Field to the desired field (for example, UPRN).

Note - The following instructions are based on ArcGIS Desktop version 10.

Note: The file must have the file extension .gpkg (not case sensitive) to recognise it as an OGC GKPG. Make sure the AddressBase Premium GeoPackage is stored in a suitable location and the file is unzipped.

1. Start ArcCatalog as a separate program, or within ArcMap if you are using version 10.

2. To establish a connection to the folder where the GKPG file is stored, right-click on Folder Connections > Connect To Folder, then navigate to the folder containing the data. Select that folder and click OK.
3. Navigate to the newly connected folder where the GKPG is stored. Open the GKPG to view its contents.

4. Drag the data with spatial elements (that is, main.street and main.blpu, which are designated by the point symbol) into the Map view. From here, they can be visualised.
5. Add the standalone tables to the map. The tables have no spatial element so cannot be visualised, but they can be related to the data with a spatial element. Select the standalone tables and drag them into the Map view. When prompted, select ID as the Unique Identifier Field(s) before clicking Finish. Note - The ‘Toggle Contents Panel’ can be used to open a window where you can select multiple tables at once.
6. You need to create the following joins/relates between the tables, as stated in in the .

   • BLPU (ID21_BLPU_Records):

     • UPRN – Application Cross Reference (ID23_XREF_Records) UPRN

     • UPRN – LPI (ID24_LPI_Records) UPRN

     • UPRN – Delivery Point Address (ID28_DPA_Records) UPRN

     • UPRN – Organisation (ID31_Org_Records) UPRN

     • UPRN – Classification (ID32_Class_Records) UPRN

   • LPI (ID24_LPI_Records):

     • USRN – Street (ID11_Street_Records) USRN

   • Street (ID11_Street_Records):

     • USRN – Street Descriptor (ID15_StreetDesc_Records) USRN

7. To create the joins / relates between tables, select the Source button in the left-hand navigation window and right-click on the first table that you wish to relate to another. To create the relevant relates, use the following instructions:

   • Click Joins and Relates > Relate...
   • From the first dropdown menu, select the attribute from the first table that will be used to create the relate between the two tables. (Apply the relationships as listed in step 5.)

   • From the second dropdown, select the table that is going to be related to. (Apply the relationships as listed in step 5.)

   • From the third dropdown, select the attribute from the table that is being related to. (Apply the relationships as listed in step 5.)

   • In the fourth box, input a relevant name for the relate (for example, BLPU_to_LPI).
   • Click OK.

   • Repeat this process for all of the joins / relates listed in step 5.

8. Once the data has been loaded into ArcMap, you may wish to display more relevant information in the Info tool than the Esri-defined Object ID. To change this, use the following instructions:

   • Double-click on the spatial dataset that you wish to change the Display Expression of.

   • Select the Display tab.

   • Change the Field to the desired field (for example, UPRN).

   • Click OK.

Note - These instructions were created using QGIS version 3.14. Other versions of QGIS can be used, from version 2.10.1 onward.

1. Launch QGIS.

2. Navigate to the GKPG location in the QGIS Browser panel. If the GKPG is not listed when this container is expanded, then right-click on the GeoPackage location and select New Connection.

3. In the File Explorer window that launches, navigate to the location where you saved the AddressBase Premium GKPG and select it.

4. In the QGIS Browser panel, click on the drop-down arrow next to your AddressBase Premium GKPG to view the contents of the GKPG file. Drag the data with spatial elements (i.e. main.street and main.blpu, which are designated by the point symbol) into the map. From here, they can be visualised.
5. Next, add the standalone tables to the map. These tables have no spatial element so cannot be visualised, but they can be related to the data with a spatial element.

6. You need to create the following joins/relates between the tables, as stated in in in the

   • BLPU (ID21_BLPU_Records):

     • UPRN – Application Cross Reference (ID23_XREF_Records) UPRN

     • UPRN – LPI (ID24_LPI_Records) UPRN

     • UPRN – Delivery Point Address (ID28_DPA_Records) UPRN

     • UPRN – Organisation (ID31_Org_Records) UPRN

     • UPRN – Classification (ID32_Class_Records) UPRN

   • LPI (ID24_LPI_Records):

     • USRN – Street (ID11_Street_Records) USRN

   • Street (ID11_Street_Records):

     • USRN – Street Descriptor (ID15_StreetDesc_Records) USRN

7. To create the joins / relates between the tables, right-click on your BLPU layer and select Properties...
8. Go to the Joins tab found on the left-hand side.

9. Click the green plus button in the bottom left-hand corner.

10. Select your Join layer. For this example, it is Application Cross reference (application_xref).

11. Select the Join field. For all BLPU links, this will be the UPRN (see the joins listed in step 4 above).

12. The Target field will also be the UPRN for this example.
13. Click OK. You should now have a join listed in your Joins window.
14. Click OK to return to your main QGIS mapping screen.

15. If you now select one of your BLPU records in the main mapping window, you will see the BLPU attributes and the relevant Application Cross References for that record.
16. You can now repeat steps 7 to 14 for all the additional joins (listed in step 6) you wish to make.

Note - The following step-by-step instructions are based on PostgreSQL version 12.3 and assume that you have set-up your database with the PostGIS spatial extension.

Requirements:

• A development platform for PostgreSQL (for example, pgAdmin or dBeaver)

• A PostgreSQL database

• PostGIS extension

• GDAL extension

• Access to a Command Prompt or similar

• A basic understanding of database terminology

Instructions

Note - Make sure the AddressBase Premium GKPG is stored in a suitable location and the file is unzipped.

1. Open Command Prompt by clicking the Windows Start button in the bottom left-hand side of the screen and typing cmd into the search bar.
2. The Command Prompt will appear.

3. Change the directory of the Command Prompt App if necessary. The directory needs to point to the folder where the GKPG is stored. For this example, the GKPG to be loaded into PostgreSQL is stored in a C:\Temp folder.

4. In the Command Prompt, type cd followed by the directory of the location of the GKPG. Press Enter on the keyboard. This will change the directory.
5. Enter the following command:

   Where: <username>, <password>, <database> and <host> can be found within the subsequent database. <data_name> is the AddressBase Premium GKPG file, for example, AddressBasePremium_AOI_2021-04-12_001_gpkg.gpkg. Example:

6. Open your chosen development platform (for example, dBeaver) by going to the Windows Start button and selecting your chosen software.

7. In your Database Navigator, move to the database you loaded data into and expand the schema. Using the code above, the GKPG should be in your default schema.

8. The GKPG will now appear as new tables / a new table in the schema nominated as default.

9. Once loaded, you may want to add Primary and Foreign Keys to the data. These can only be added on columns where the data values are unique. Where there are no unique data values, an index may be added which will aid searching. For the BLPU table, the UPRN provides a unique value.