Hello all,
I’ve been looking into improving geocoding coverage for India and wanted to get some feedback from the community before going any further. In particular, I’d like to start a discussion around postcode-level geocoding support for India in Nominatim.
The Government of India’s Open Government Data (OGD) platform publishes an All India Pincode Directory, which maps PIN codes to post offices along with associated locality, district, and state information. The dataset is openly available via the OGD platform and provides nationwide coverage at the post-office level (pincode/postcode) & includes circlename, regionname, divisionname, officename, pincode, officetype, delivery, district, statename, latitude, longitude.
This effectively maps each PIN code to a post office with associated administrative context (district, state) and a central point coordinate.
My intention is not to claim street- or address-level accuracy, but rather to explore whether this dataset could be used as a supplementary gazetteer layer for searching PIN codes directly & improving locality-level fallback results & reverse geocoding results.
One possible approach would be to import this data as external postcode data, with appropriate ranking, similar in spirit to how other national postcode datasets are handled, while keeping it clearly separate from street-level addressing.
I can understand that support for external datasets is not a core priority for Nominatim and that such work generally only progresses when it is either contributor-driven or funded. I’m not expecting this to be taken up as a priority task, but would like to check whether the proposed approach aligns with the project’s direction.
So, I wanted to ask for feedback from maintainers on whether this kind of openly available official dataset fits Nominatim’s data model and import philosophy? If it seems appropriate, I’d be more than happy to provide the dataset in an importable format and help in the implementation.
Thanks.