api_v1.meta_dimensions
Atlas view
api_v1.meta_dimensions
api_v1.meta_dimensionsOne row per source and coded dimension, explaining what that column's codes mean.
First 5 rows ·
api_v1.meta_dimensionssnapshot 2026-05-12
| source_id | code | meaning | value_format | notes |
|---|---|---|---|---|
| bufdir-barnefattigdom | category_format | Workbook column Tallformat (second category axis), normalised lowercase | antall or prosent | Paired with category_unit — four tuples per workbook × geography when all rows exist upstream. |
| bufdir-barnefattigdom | category_unit | Workbook column Enhet (first category axis), normalised lowercase | barn or husholdning | Mirrors the paired rows in each Data sheet for child vs household units. |
| bufdir-barnefattigdom | indicator_api_id | Stable surrogate indicator key for ZIP workbook rows (not Strapi) | bf_zip_<24 lowercase hex> | Derived as SHA-256 fingerprint of the `Indikator_*.xlsx` filename stem; breaks if Bufdir renames workbook files inside the ZIP. |
| bufdir-barnefattigdom | region_code | Geographic code from workbook column Region | 0 (land), 2-digit fylke, 4-digit kommune, or longer codes for bydel/delbydel | Rows follow whatever geography levels each indicator publishes. |
| bufdir-barnefattigdom | values_json | Parsed year→value map for the geography and Enhet/Tallformat tuple | JSON object with string year keys; suppressed cells as null | Matches prior raw semantics (full spine per slice); mart reads scalar `value` for the keyed year row. |
Built from
This Atlas view joins data from the following upstream sources. Click through for provenance, methodology, and citation.
No direct dbt parents recorded.
Columns
| Column | Description |
|---|---|
| source_id | Source the dimension belongs to. Joins to `mart_meta_sources.source_id` and `_sources_manifest.source_id`. |
| code | Upstream's own dimension code (e.g. `Region`, `Tid`, `ContentsCode` for SSB; `GEO`, `AAR`, `KJONN`, `MEASURE_TYPE` for FHI). Verbatim from the source's API/export — not translated to canonical Atlas vocabulary. |
| meaning | Short human-readable description of what the dimension represents. Hand-authored editorial content from each source's `manifest.yml` `dimensions:` block. |
| value_format | Encoding of values within the dimension (e.g. "4-digit year as text", "1- to 6-character numeric code"). Empty string when no specific format applies. |
| notes | Editorial notes — code-list size, semantic gotchas, downstream filtering decisions. Empty string when no notes apply. |
Sample query
Live PostgREST query that returns the first 5 rows. URL adapts to the current host — local dev shows api-atlas.localhost; the deployed site shows the configured API base.
https://api-atlas.urbalurba.com/meta_dimensions?limit=5
About this view
One row per source and coded dimension, explaining what that column's codes mean.
Per-source × per-dimension catalogue. Backs the "what does this
column mean" panel on `/data/sources/[source_id]` (customer
frontend, PLAN-007 phase 4) and exposes editorial semantics
shoppers would otherwise have to leave Atlas to find on the
upstream's docs site.
v1 scope (this PR): editorial pass-through over the
`_sources_dimensions` seed (4 columns: `meaning`, `value_format`,
`notes`). The `cardinality`, `example_values`, and `null_count`
columns from PLAN-007 phase 3.4's full spec require introspecting
`raw.<source>` tables with per-source column-name rules — deferred
to a follow-up PR. See `mart_meta_dimensions.sql` header for the
open question (Jinja loop vs separate Python extract script vs
per-source `column_name:` field in manifest.yml).
Row count = same as `_sources_dimensions` (~216, growing with
every new source × upstream-dimension).