Changes to interface use of PxWeb databases on 8 June – change interface queries as instructed
NewsThe structure of px files in statistical databases will change on 8 June 2026. In order for queries made via the API interface to function after the change, they need to be adjusted. Instructions for the necessary changes have now been published.
The changes concern tables with a four-character ID code, e.g. 12tg. The changes have an effect on the StatFin database and chargeable databases, for example. No changes are made to the StatFin archive database at this stage.
The changes aim at improving the retrievability and understandability of table contents in the PxWeb interface, in search engine use and in AI usage. The changes also help improve the consistency of table metadata, as well as the management of change situations
and the life cycle of tables, and communication related to these.
What changes will there be in the tables?
There will be three changes:
1. The file name, TABLEID ID and MATRIX ID are shortened. This applies to all active databases.
2. A variable code has been added to all variables. Previously, the name of the variable was used in API as the variable identifier but in future, the variable code will be in its place.
3. In some cases, a prefix must be added to the value codes of the analysis variable. This change concerns all active databases, but not all variables.
Shortening of identifiers
As part of the change, the data of the TABLEID and MATRIX fields will be shortened to consist only of the ID code. After the change, the file name, TABLEID and MATRIX will be identical.
For example, the file name, TABLEID and MATRIX in a table of the employment service statistics will change as follows:
before the change | after the change | |
file name | statfin_tyonv_pxt_12tg.px | 12tg.px |
TABLEID | TABLEID="statfin_tyonv_pxt_12tg” | TABLEID="12tg” |
MATRIX | MATRIX="010_12tg_2026" | MATRIX="12tg" |
In some cases, the file name will not only be shortened, but a specifying version identifier will also be added at the end of the ID. This change mainly applies to old archive tables of chargeable databases. No additional elements will be added to the file names in StatFin tables. While such suffixes exist in the StatFin archive database, the file names in that database will not be changed or shortened at this stage.
Variable code
A variable-level code is added to each table variable to serve as the variable identifier in API queries in place of the current variable name after the change. Below is a list of the variable codes to be added to different variables.
Content variables
VARIABLECODE contentscode is added to the content variable “Data”:
VARIABLECODE(“Data”)=“contentscode”;
Time variables
VARIABLECODE according to the cycle of the time variable is added to the time variable. For example, for annual tables timeperiod_y:
VARIABLECODE(“Year”)=“timeperiod_y”;
Classifying variables
VARIABLECODE, which is the identifier of the value-level codes of the classification in
question, is added to classifying variables. For example, for a region variable often alue_23_20260101:
VARIABLECODE(“Region”)=“alue_23_20260101”;
Prefix of an analysis variable
In some cases, a prefix will be added to the value-level codes of an analysis variable. The analysis variable refers to the variable “Data”.
For example, in StatFin table 111e concerning road traffic accidents, the data variable “kuol” is changed to “ton-kuol”.
These changes will not appear in all tables.
Specifying instructions
The instructions include, as an annex, detailed before-and-after listings of the changes at both file and variable level, as well as examples of the changes to px files and interface queries.
Contact information
Kim Huuhko
Web Service Specialist
+358 29 551 2354