The validation report consists of a Header section and an Error Listing section. The Header contains validation process metadata and a general overview of the results of the validation. The Error Listing section contains the details of all unique errors detected.
Header of the report
Total number of errors found
Number of all error occurrences encountered during the validation process. Note: The validation services apply a cap of 2000 individual occurrences; the service concludes and produces a report once it reaches 20001 errors. This means that after the correction and resubmission of the dataset further errors may be identified. If the error cap is reached, this is indicated via a message in the Header.
Errors per type
Total of all error occurrences broken down by error type. The breakdown is based on Error Code and Message ID. Error Codes refer to a wider classification of possible errors (e.g. technical issues, validation related issues), while Message IDs refer to a specific type of error (e.g. unexpected code, incorrect data format). The aggregation in this section groups errors with identical Message IDs together, meaning the subtotal represented by a line item may stem from a number of different root causes. The purpose of the grouping here is to provide general information on the nature of errors identified, not a detailed breakdown. Note: A single root cause my trigger multiple error types, and these will be listed separately (e.g. a code is unexpected and also violates a length constraint). In case the dataset does not contain errors, the Errors Per Type section has no entries.
Dataset validated
Name of the input dataset
SDMX Converted version
Version of the validation service engine used.
Generated on
Date stamp.
Structure information
DSD name and version.
Figure 62 – Example error report header, with errors detected
Figure 63 – Example error report header, with no errors detected
Error Listing
The Error Listing contains the list of specific errors detected by the validation service, in the order of detection. In case of no errors are detected, this section of the validation report remains empty.
Error Code
High level error classification, first level of error typology. Serves diagnostic purposes.
Message ID
Error type, second level of error typology. Serves diagnostic purposes.
Concept Name
Error header. Name of concept affected by error.
Concept Type
Error header. Type of concept affected by error. Value may be: DimensionAttributeMeasure.
Concept Value
Error header. Value of concept affected by error. Possibly confidential.
Error in Concept
Name of concept affected by error.
Number of Occurrences
Value refers to this unique error (total).
Error Description
Error message, describing the error. Possibly confidential.
Detail
Additional information on the error and its resolution.
Series Key (column view)
Location of the error in the dataset, per dimension name and values of the series key.
Series key (horizontal view)
Location of the error in the dataset, per dimension values of the series key. The values and sequence of dimensions presented are identical to those of the column view.
Grouping and display of errors
Grouping of errors
Errors are grouped under separate sub-headings per the unique root cause they are generated by. The errors are identified as unique according to the logic presented in Figure 64 – Grouping of errors
Display of errors
The validation report contains all detected errors (unless the 2000 cap is reached, see above). Error messages and the dimension list are only displayed for the first error occurrence. For any further occurrences only the series key is listed, with a corresponding header with the names of dimensions available. In case of business need the number of error occurrences displayed in the report may be limited (e.g. 3 per group).
Filtering reports due to confidentiality constraints
Eurostat policy prohibits the inclusion of confidential data in validation reports distributed to external parties, including national statistical institutes. Statistical domains collecting and processing confidential data will receive a report where elements with the possibility of the presence of confidential data are filtered out. The filtering:
...
Figure 65 – Example of error listing without filtering
Figure 66 – Example of error listing with a measure filtered for confidentiality.
The concept value and error description fields are removed