Workstream A: catalogue what our downloads actually contain versus what we import #16

Closed
opened 2026-08-01 19:12:55 +00:00 by art-bot · 3 comments
Collaborator

Every country's layer list is a proof of concept of the importer rather than a
considered map, and we have never systematically looked at what is already sitting on disk.
Much of this is free — we already pay the download cost and read the files.

Two parts:

  1. What's already on disk. Enumerate every layer/table in the GeoPackages, shapefiles
    and GML we already download: LMÍ IS 50V (this subsumes the standing "enumerate remaining
    LMÍ GeoPackage layers and decide which to import" item), Lantmäteriet Topo 10/50 themes,
    the SverigepaketTP columns we ignore (speed limit, width, ADT, road authority…), and
    BDOT10k's full feature-class list.
  2. What else the same sources offer. Other themes in the same Geotorget order, other
    Lastkajen packages, other BDOT10k products, other LMÍ datasets. The örnefni
    place-name dataset keeps coming up as the fix for several open items at once — unnamed
    peaks, unnamed waterfalls and springs, and Iceland's unlabelled water.

Output should be a per-country table of available → imported → worth importing, which
is what turns "feature parity" from an aspiration into a work list. Sweden probably wins:
full Topo 10 access is the deepest data we have.

This is an exploration job, not an implementation one — the deliverable is the table.
Background: docs/unification.md, "Parallel workstreams".

Every country's layer list is a proof of concept of the *importer* rather than a considered map, and we have never systematically looked at what is already sitting on disk. Much of this is free — we already pay the download cost and read the files. Two parts: 1. **What's already on disk.** Enumerate every layer/table in the GeoPackages, shapefiles and GML we already download: LMÍ IS 50V (this subsumes the standing "enumerate remaining LMÍ GeoPackage layers and decide which to import" item), Lantmäteriet Topo 10/50 themes, the `SverigepaketTP` columns we ignore (speed limit, width, ADT, road authority…), and BDOT10k's full feature-class list. 2. **What else the same sources offer.** Other themes in the same Geotorget order, other Lastkajen packages, other BDOT10k products, other LMÍ datasets. The **örnefni** place-name dataset keeps coming up as the fix for several open items at once — unnamed peaks, unnamed waterfalls and springs, and Iceland's unlabelled water. **Output should be a per-country table of *available → imported → worth importing***, which is what turns "feature parity" from an aspiration into a work list. Sweden probably wins: full Topo 10 access is the deepest data we have. This is an exploration job, not an implementation one — the deliverable is the table. Background: `docs/unification.md`, "Parallel workstreams".
Author
Collaborator

night-shift — released, not worked.

The worker for this issue was cut off by a session limit before it committed anything; the branch night/issue-16-download-catalogue has no commits. No obstacle was found, so this goes back in the ready queue for the next run rather than being marked blocked.

**night-shift** — released, not worked. The worker for this issue was cut off by a session limit before it committed anything; the branch `night/issue-16-download-catalogue` has no commits. No obstacle was found, so this goes back in the ready queue for the next run rather than being marked blocked.
Owner

Queue note (2026-08-04): first in the run — #29 (place names) is gated behind this catalogue, so this issue is the unlock. Deliverable is the available → imported → worth-importing table per country.

Queue note (2026-08-04): first in the run — #29 (place names) is gated behind this catalogue, so this issue is the unlock. Deliverable is the available → imported → worth-importing table per country.
Owner

PR #51#51

PR #51 — https://git.blahonga.org/art/ismap/pulls/51
art closed this issue 2026-08-05 00:18:14 +00:00
Sign in to join this conversation.
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#16
No description provided.