Skip to content

Sweep hardcoded *.ocortez.com data URLs to HF resolve URLs (doc generators, analysis examples, index.html) #39

Description

@oskrgab

What to build

Point every consumer-facing and example data URL at the HF resolve base instead of *.ocortez.com, so copy-paste queries work against the new data host (ADR-0005). Prefer deriving URLs from BASE_URL over hardcoding.

Surfaces to update:

  • scripts/export/petrobras_3w/schema_doc_generator.py (access-pattern example queries / source_url docs)
  • scripts/export/force_2020/schema_doc_generator.py (BASE_URL = https://dev-petrodb.ocortez.com/force_2020)
  • scripts/analysis/force_2020/plot_well_logs.py (BASE_URL = https://dev-petrodb.ocortez.com/force_2020/wells)
  • scripts/analysis/argentina/production_analysis_remote.py and ..._imperial.py (BASE_URL = https://petrodb.ocortez.com/argentina)
  • parquet/index.html — example DuckDB queries / data links

Schema docs/example queries must use manifest-based discovery (_files.json) against HF, never .../*.parquet globbing (ADR-0004). Leave the constants.py dev fallback host as-is (local Caddy dev is unchanged).

Acceptance criteria

  • No data-fetching URL in the listed files points at *.ocortez.com; all use the HF resolve base (ideally via BASE_URL)
  • Generated schema docs (schema.md) example queries copy-paste into DuckDB and succeed against HF
  • parquet/index.html example queries reference HF resolve URLs and work
  • Manifest-based discovery is used in examples (no *.parquet globbing)

Blocked by

Context: ADR-0005, ADR-0004 (manifest is the discovery contract). The dev-only Caddy fallback in constants.py stays unchanged.

Metadata

Metadata

Assignees

No one assigned

    Labels

    needs-triageMaintainer needs to evaluate this issue

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions