#84: R5/R13 catalogue — contours, peaks, buildings and airports now have a line per country, and ten import issues are filed #111

Merged
art-bot merged 1 commit from issue-84-terrain-catalogue into main 2026-08-13 04:46:00 +00:00
Owner

Closes #84


Issue #84 — Terrain, buildings and airports: the catalogue lines, and the ten issues they turn into

Branch: issue-84-terrain-catalogue · docs-only change (plus a stale comment in
conf/fimap.yaml), no code, no data, no DB.

What changed

Each country doc now carries one line per layer for the four Iceland-only layers —
contours, peaks, buildings, airports — stating either yes and here is the source and
the size
or no and here is why, per the maintainer's direction on the issue (contours
should exist everywhere, they are heavy, and the Sweden–Finland border match is of
interest):

  • docs/country-sweden.md — new "Terrain, buildings and airports" section. All four
    are in the already-negotiated topo10 Geotorget order (each a files: entry away,
    no new approval): contours + peaks from tema Höjd (hojdlinje 5 m ekvidistans +
    hojdpunkt), buildings from tema Byggnadsverk (byggnad, objekttypnr 2061–2067,
    ≥ 15 m²), airports from tema Anläggningsområde (flygplatsomrade IATA/ICAO +
    start_landningsbana, 2853). Theme facts verified against Lantmäteriet's own
    documentation (GEODOK/51). The themes' exact GB is not public — the catalogue says so
    and the first-fetch recording is part of the filed issues (R5).
  • docs/country-poland.md — new section, same shape. All four are in the BDOT10k
    catalogue but none among the 17 fetched classes — the class names are read from the
    2021 schema (Dz.U. 2021 poz. 1412 załącznik 2, the regulation text), not the fetch
    list, which is exactly the R6 lesson of #85. Contours = OT_RTLW_L (rodzaj
    poziomica), peaks = OT_RTPW_P, buildings = OT_BUBD_A, airport areas = KU*
    OT_KUKO_A/P (lotnisko lub lądowisko). Measured on disk: OIKM_P (already fetched)
    carries 436 lądowisko dla helikopterów nationally and 0 pas startowy rows; PTKM
    has no RODZAJ column, so cover cannot split airports out.
  • docs/country-finland.md — the half-written catalogue is finished: verdicts
    revisited per the maintainer's comment (contours and buildings are wanted; the "out
    of scope for Sweden parity" note in conf/fimap.yaml is withdrawn), airports are
    already on disk (lentokenttaalue, 835 polygons in the fetched MTK-muut), and peaks
    is the one genuine "no via MTK today": the elevation-point class korkeuspiste
    (52210) exists but the 2025 delivery carries 0 rows (measured on disk) — issue
    #108 picks the source (DEM-derived vs the register's named fells). MTK file sizes
    re-measured against Paituli (112 GB total, was 107 GB).
  • docs/workstream-a.md — the Sweden/Poland catalogue home updated to match (the
    hojd/byggnadsverk/anlaggningsomrade bullets, the BDOT10k not-on-disk list now names
    RT*, the work list points at the issues).
  • conf/fimap.yaml — the skip-list comment rewritten (contours/buildings wanted and
    filed; bogs/rock/edges stay out for the reasons the doc already gives).

The ten issues (the acceptance's "own issue per worth-importing layer")

#101 Sweden contours+peaks (tema Höjd) · #102 Sweden buildings (Byggnadsverk) · #103
Sweden airports (Anläggningsområde) · #104 Poland contours+peaks (RT*) · #105 Poland
buildings (BUBD_A) · #106 Poland airports (KU*) · #107 Finland contours
(MTK-korkeussuhteet) · #108 Finland peaks (source decision) · #109 Finland buildings
(MTK-rakennus) · #110 Finland airports (lentokenttaalue).

Contours and peaks are grouped per country where they share one source theme (SE tema
Höjd, PL RT* group) — the project's natural fetch unit, matching how the workstream-a
work list already grouped them. Each issue carries the source, the size where known,
and the first step (measure the GB on first fetch; the R6 poziomica-population check for
Poland).

Verification

  • Class/theme facts: Lantmäteriet GEODOK/51 (hojd/byggnadsverk/anlaggningsomrade
    layers, objekttyper, attributes); BDOT10k 2021 regulation text (RT*/BUBD*/KU* codes
    and rodzaj domains); the fetched BDOT10k county data (national OIKM_P rodzaj counts,
    PTKM schema); MML Lueminut.txt + the on-disk MTK-muut (lentokenttaalue 835,
    korkeuspiste 0); Paituli file sizes re-measured via HTTP.
  • go test ./... in importer/ (incl. TestRealConfigsValidate over the edited
    fimap.yaml) and mapsrv/ — pass.
  • Nothing imported, no DB touched — the issue is documentation + backlog.

Left undone

  • Poland's poziomica county population (R6 check) and Sweden's exact theme GBs are
    first-fetch steps of the filed issues, deliberately not measured here (would need a
    source-data download, out of scope for a catalogue pass).
Closes #84 --- # Issue #84 — Terrain, buildings and airports: the catalogue lines, and the ten issues they turn into **Branch:** `issue-84-terrain-catalogue` · docs-only change (plus a stale comment in `conf/fimap.yaml`), no code, no data, no DB. ## What changed Each country doc now carries one line per layer for the four Iceland-only layers — contours, peaks, buildings, airports — stating either *yes and here is the source and the size* or *no and here is why*, per the maintainer's direction on the issue (contours should exist everywhere, they are heavy, and the Sweden–Finland border match is of interest): - **docs/country-sweden.md** — new "Terrain, buildings and airports" section. All four are in the already-negotiated `topo10` Geotorget order (each a `files:` entry away, no new approval): contours + peaks from tema Höjd (`hojdlinje` 5 m ekvidistans + `hojdpunkt`), buildings from tema Byggnadsverk (`byggnad`, objekttypnr 2061–2067, ≥ 15 m²), airports from tema Anläggningsområde (`flygplatsomrade` IATA/ICAO + `start_landningsbana`, 2853). Theme facts verified against Lantmäteriet's own documentation (GEODOK/51). The themes' exact GB is not public — the catalogue says so and the first-fetch recording is part of the filed issues (R5). - **docs/country-poland.md** — new section, same shape. All four are in the BDOT10k catalogue but none among the 17 fetched classes — the class names are read from the 2021 schema (Dz.U. 2021 poz. 1412 załącznik 2, the regulation text), not the fetch list, which is exactly the R6 lesson of #85. Contours = OT_RTLW_L (rodzaj `poziomica`), peaks = OT_RTPW_P, buildings = OT_BUBD_A, airport areas = KU* OT_KUKO_A/P (`lotnisko lub lądowisko`). Measured on disk: OIKM_P (already fetched) carries 436 `lądowisko dla helikopterów` nationally and 0 `pas startowy` rows; PTKM has no RODZAJ column, so `cover` cannot split airports out. - **docs/country-finland.md** — the half-written catalogue is finished: verdicts revisited per the maintainer's comment (contours and buildings are wanted; the "out of scope for Sweden parity" note in `conf/fimap.yaml` is withdrawn), airports are already on disk (`lentokenttaalue`, 835 polygons in the fetched MTK-muut), and peaks is the one genuine "no via MTK today": the elevation-point class `korkeuspiste` (52210) exists but the 2025 delivery carries **0 rows** (measured on disk) — issue #108 picks the source (DEM-derived vs the register's named fells). MTK file sizes re-measured against Paituli (112 GB total, was 107 GB). - **docs/workstream-a.md** — the Sweden/Poland catalogue home updated to match (the hojd/byggnadsverk/anlaggningsomrade bullets, the BDOT10k not-on-disk list now names RT*, the work list points at the issues). - **conf/fimap.yaml** — the skip-list comment rewritten (contours/buildings wanted and filed; bogs/rock/edges stay out for the reasons the doc already gives). ## The ten issues (the acceptance's "own issue per worth-importing layer") #101 Sweden contours+peaks (tema Höjd) · #102 Sweden buildings (Byggnadsverk) · #103 Sweden airports (Anläggningsområde) · #104 Poland contours+peaks (RT*) · #105 Poland buildings (BUBD_A) · #106 Poland airports (KU*) · #107 Finland contours (MTK-korkeussuhteet) · #108 Finland peaks (source decision) · #109 Finland buildings (MTK-rakennus) · #110 Finland airports (lentokenttaalue). Contours and peaks are grouped per country where they share one source theme (SE tema Höjd, PL RT* group) — the project's natural fetch unit, matching how the workstream-a work list already grouped them. Each issue carries the source, the size where known, and the first step (measure the GB on first fetch; the R6 poziomica-population check for Poland). ## Verification - Class/theme facts: Lantmäteriet GEODOK/51 (hojd/byggnadsverk/anlaggningsomrade layers, objekttyper, attributes); BDOT10k 2021 regulation text (RT*/BUBD*/KU* codes and rodzaj domains); the fetched BDOT10k county data (national OIKM_P rodzaj counts, PTKM schema); MML Lueminut.txt + the on-disk MTK-muut (lentokenttaalue 835, korkeuspiste 0); Paituli file sizes re-measured via HTTP. - `go test ./...` in importer/ (incl. TestRealConfigsValidate over the edited fimap.yaml) and mapsrv/ — pass. - Nothing imported, no DB touched — the issue is documentation + backlog. ## Left undone - Poland's `poziomica` county population (R6 check) and Sweden's exact theme GBs are first-fetch steps of the filed issues, deliberately not measured here (would need a source-data download, out of scope for a catalogue pass).
Every country doc now answers the maintainer's question for each of the four
Iceland-only layers — yes-and-where-and-how-big, or no-and-why. This is the R13
parity-gap pass from #84: documentation only, no code, no data, no DB.

Per country, the verdicts (all sources verified against the providers' own
documentation or measured on disk — nothing from memory):

**Sweden** — all four are in the already-negotiated `topo10` Geotorget order, each a
`files:` entry away: contours + peaks from tema Höjd (`hojdlinje` 5 m ekvidistans,
`hojdpunkt`, GEODOK/51), buildings from tema Byggnadsverk (`byggnad`, objekttypnr
2061–2067, ≥ 15 m²), airports from tema Anläggningsområde (`flygplatsomrade` with
IATA/ICAO + `start_landningsbana`). Issues #101/#102/#103. The themes' exact GB is
recorded on first fetch (the delivery listing is not public; for scale, `mark` is
6.4 GB zipped / 13.4 GB on disk).

**Poland** — all four are in the BDOT10k catalogue but none among the 17 fetched
classes, named from the 2021 schema (Dz.U. 2021 poz. 1412): contours OT_RTLW_L
(rodzaj poziomica) and peaks OT_RTPW_P (both with `wysokosc`; the poziomica population
still needs the R6 check on first fetch, NMT 1 m as the national fallback), buildings
OT_BUBD_A, airport areas OT_KUKO_A/P (lotnisko lub lądowisko — the KU* overlay, same
design problem as sankmark; the fetched OIKM_P already carries 436 helipads, measured
across all 380 counties). Issues #104/#105/#106.

**Finland** — the half-written catalogue is finished and the verdicts revisited per
the maintainer's comment: contours (MTK-korkeussuhteet, 34.5 GB) and buildings
(MTK-rakennus, 4.4 GB) are wanted — the old "out of scope for Sweden parity" note in
conf/fimap.yaml is withdrawn; airports (lentokenttaalue, 835 polygons) are already on
disk in the fetched MTK-muut; peaks is the one "no via MTK today" — the elevation-point
class korkeuspiste (52210) exists but the 2025 delivery carries 0 rows (measured on
disk), so issue #108 picks the source (DEM-derived spot heights vs the register's named
fells). Issues #107–#110. The MTK file sizes in the catalogue are re-measured against
Paituli (112 GB total, was 107 GB).

Verified: the class/theme facts against GEODOK/51 (Lantmäteriet), the BDOT10k 2021
regulation text and the fetched county data (helipad counts, PTKM having no RODZAJ
column), Lueminut.txt and the on-disk MTK-muut (lentokenttaalue 835, korkeuspiste 0);
importer `TestRealConfigsValidate` over the edited fimap.yaml, `go test ./...` in
importer/ and mapsrv/. No layer data was touched, so no DB verification applies.

The workstream-a catalogue is updated to match (the Sweden/Poland catalogue home).
art-bot force-pushed issue-84-terrain-catalogue from bc332a3c0f to 7ea7295407 2026-08-13 04:45:56 +00:00 Compare
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
art/ismap!111
No description provided.