Part of a Data Solution

Postal Codes Solution
Every worldwide postal-code dataset in one bundle — the ultimate global reference for precise postal codes.
Explore →

Premium

You're viewing a free sample

Get the complete dataset — all rows and files — delivered instantly after checkout, with lifetime access to the latest version.

Secure checkout via Stripe
Instant download after payment
Lifetime access to the latest version

20% off

$99.90$79.92

one-time payment

Download sample · CSV · 7.49 kB Report an Issue

API Access

Access dataset files directly from scripts, code, or AI agents.

Browse dataset files

Dataset Files

Each file has a stable URL (r-link) that you can use directly in scripts, apps, or AI agents. These URLs are permanent and safe to hardcode.

/logistics/postal-codes-ca/

├ .licensing-notes.md

https://datahub.io/logistics/postal-codes-ca/_r/-/.licensing-notes.md

├ ENRICHMENT_PLAN.md

https://datahub.io/logistics/postal-codes-ca/_r/-/ENRICHMENT_PLAN.md

├ README.md

https://datahub.io/logistics/postal-codes-ca/_r/-/README.md

└ datapackage.yml

https://datahub.io/logistics/postal-codes-ca/_r/-/datapackage.yml

Key Files

Start with these files — they give you everything you need to understand and access the dataset.

datapackage.yml— metadata & schema

https://datahub.io/logistics/postal-codes-ca/_r/-/datapackage.yml

README.md— documentation

https://datahub.io/logistics/postal-codes-ca/_r/-/README.md

Typical Usage

1. Fetch datapackage.yml to inspect schema and resources
2. Download data resources listed in datapackage.yml
3. Read README.md for full context

Data Files

Postal Codes Data Resource for Canada, CA

Field	Description	Title
country_code	ISO 3166-1 alpha-2 code for the country — always CA.	Country Code
postal_code	Six-character postal code, formatted 'A1A 1A1'. Primary key — one row per postal code.	Postal Code
place_name	Name of the city, town, or place.	Place Name
admin_name1	Province or territory name (e.g., Ontario).	Administrative Name 1
admin_code1	Province or territory two-letter code (e.g., ON).	Administrative Code 1
admin_name2	County-tier administrative division — Census Division name (e.g., Peel, Division No. 13).	Administrative Name 2
admin_code2	Four-digit Census Division UID (SGC 2021) matching the county-tier division.	Administrative Code 2
admin_name3	Community name — settlement-level locality below the county tier.	Administrative Name 3
admin_code3	Code for the tertiary administrative division (reserved).	Administrative Code 3
latitude	Latitude coordinate of the place (WGS84, decimal degrees).	Latitude
longitude	Longitude coordinate of the place (WGS84, decimal degrees).	Longitude
accuracy	Accuracy level of the latitude and longitude coordinates (1–6).	Accuracy
alternative_city_name	Alternative name(s) for the city or place, pipe-separated.	Alternative City Name

Download

Download sample CSV

About

Last updated: 17 July 2026
Total rows: ...
Format: CSV
File size: 7.49 kB
Source: Multiple sources

About this dataset

Canada (CA) — internal licensing notes

Status: sold product is clean. The R2 bucket is not. Two separate findings below — don't conflate them.

The problem (datahub-next#941 §1.1, opened 2026-08-08)

StatCan's Postal Code Conversion File (PCCF) End-use Licence has three clauses that each independently prohibit commercial use: Clause 1 (no distribution/making public), Clause 3 (no deriving a product for commercial sale — bites on derivation, so PCCF can't even sit open beside a build as a validation reference), Clause 8 (publication capped at a 1% extract; a nationwide crosswalk over 899,779 codes is not that). GeoNames CA_full was inspected during that review and found to have admin2_nonempty = 0, admin3_nonempty = 0 across all 900,676 rows — meaning 100% of the admin2 content in the pre-rebuild product was PCCF-derived.

What was done — the clean-room rebuild (issue comment, 2026-08-11)

"Canada is rebuilt and live on r2. two open sources only — GeoNames postal
codes (CC BY 4.0) + StatCan 2021 boundary files (OGL–Canada, permits
commercial sale). PCCF never opened, not even to validate."

Coverage: 899,779 codes, 13/13 provinces, 293/293 census divisions, 3,724/5,161 subdivisions (the rest have no postal delivery). Fill: admin2 99.98%, admin3 99.96% — up from the old PCCF build's 94.2%/79.3%.

I independently re-verified this claim on 2026-08-19 (before I knew this issue existed) while auditing script reproducibility across the postal-codes product: found two competing pipelines existed (scripts/ca/ on origin/main — PCCF-dependent; local ca/scripts/ — explicitly a "clean-room rebuild" per its own docstring, which states it "must never read, join against, or 'check itself' with" PCCF). Confirmed the clean-room build is what's actually live — ca/ca.csv and ca/consolidated/latest/ca.csv (the path the paid product's premium_source_key actually points to) both match the local clean-room output on byte size. BUILD_LOG.md documents fetch SHA-256s and a cross-check against 6 known cities.

What the issue comment does NOT mention — found 2026-08-19

The raw, licence-gated PCCF file itself is still sitting in the same R2 bucket this paid product ships from: ca/pccf/canada_pccf.csv, 65,761,853 bytes, dated 2026-07-17 (before the 2026-08-11 rebuild). ca/scripts/ publish_r2.py has an --include-pccf deletion flag built specifically for this, evidently never invoked. PCCF's Clause 1 ("shall not lend, rent, lease, sublicense, distribute, make public, transfer or sell any part") does not distinguish between "the sold product" and "a file sitting in the same bucket" — having it there at all is a live gap the rebuild didn't close.

This deletion was NOT executed — it's irreversible (the only copy of the file) and not mine to trigger unilaterally. Also left unresolved: the superseded ca/consolidated/latest/* and dated snapshot paths that the same flag was meant to clean up (a storage-cost issue, not a licensing one, since they now hold clean-room data).

Open questions the issue itself flagged, still unanswered

The 5-province StatCan supplement's own origin/licence was never traced
GeoNames' own upstream source for full 6-character Canadian codes is unclear — their published source list says Canada is "FSA only" (~1,600 FSAs) but the file ships 900,676 full codes
What, if anything, the two existing customers (orders shipped 2026-07-30, 2026-08-04) need to be told

Source: datahub-next#941 (opened by @anuveyatsu, 2026-08-08), comment by @amautadev, 2026-08-11; independent re-verification and the PCCF-file finding, 2026-08-19 (this session).

CLOSED (attribution only) — 2026-08-24 (datahub-next#941 §4)

Added standalone ca/ATTRIBUTION.txt (GeoNames CC BY 4.0 + Open Government Licence – Canada, with an explicit note that this dataset was never validated against PCCF) — now in R2 and inside ca.zip. package.py updated to write and zip it on every run; publish_r2.py updated to upload it. Also found and fixed a stale _meta/catalog.json CA entry: it still pointed at the pre-rebuild file sizes from 2026-07-30 (ca.csv 73,965,760 bytes) instead of the clean-room rebuild live since 2026-08-11 (ca.csv 81,550,696 bytes) — rebuilt from the actual R2 listing.

This does NOT close the PCCF finding above. ca/pccf/canada_pccf.csv is still sitting in the bucket, still not deleted — that decision was already flagged as irreversible and not mine to trigger unilaterally, and remains open.