Backend Code Map¶

Source repository: core-stack-org/core-stack-backend

The backend is a Django project with several apps. For pipeline work, the center of gravity is computing/, backed by utilities/, public_api/, gee_computing/, and GeoServer helpers.

Common Request Path¶

flowchart TB
    A[Client request<br/>/api/v1/...] --> B[nrm_app/urls.py<br/>mounts Django apps]
    B --> C[computing/urls.py<br/>matches compute API path]
    C --> D[computing/api.py<br/>parses payload]
    D --> E[Celery task<br/>queue: nrm]
    E --> F[Pipeline module<br/>runs processing]
    F --> G[Earth Engine, GCS,<br/>local files, or PostGIS]
    G --> H[GeoServer publication<br/>and Layer metadata]

    D -. computing-api-run .-> E
    F -. export or sync .-> G
    H -. reusable output .-> I[Explorer, APIs,<br/>QGIS, notebooks]

Modules You Will Need¶

App or module	Why it matters
`nrm_app/urls.py`	Mounts all API apps under `/api/v1/` and exposes Swagger/ReDoc.
`computing/urls.py`	Lists computation API paths such as LULC, hydrology, SWB, terrain, NREGA, and STAC.
`computing/api.py`	Parses request payloads and triggers tasks or helper functions.
`computing/` submodules	Hold the pipeline logic: `lulc/`, `mws/`, `misc/`, `terrain_descriptor/`, `surface_water_bodies/`, and others.
`computing/models.py`	Stores `Dataset` and `Layer` records for generated outputs.
`computing/utils.py`	Publishes vectors to GeoServer, writes layer metadata, updates sync flags, and fixes geometries.
`utilities/gee_utils.py`	Initializes GEE, builds asset paths, exports rasters/vectors, polls tasks, syncs GCS, and makes assets public.
`utilities/geoserver_utils.py`	Wraps GeoServer REST operations.
`gee_computing/models.py`	Stores encrypted `GEEAccount` credentials used by GEE-backed pipelines.
`public_api/`	Exposes read-oriented public data APIs and generated layer URLs.

Common Helper Functions¶

The call graph points to a small set of heavily reused helpers. Learn these early:

Helper	File	Job
`valid_gee_text()`	`utilities/gee_utils.py`	Normalizes names for GEE paths and layer names.
`ee_initialize()`	`utilities/gee_utils.py`	Initializes Earth Engine from a stored `GEEAccount`.
`is_gee_asset_exists()`	`utilities/gee_utils.py`	Checks whether an asset already exists before export or reuse.
`check_task_status()`	`utilities/gee_utils.py`	Polls Earth Engine export tasks.
`get_gee_asset_path()` / `get_gee_dir_path()`	`utilities/gee_utils.py`	Builds deterministic asset paths.
`export_vector_asset_to_gee()`	`utilities/gee_utils.py`	Exports an `ee.FeatureCollection` to an Earth Engine asset.
`export_raster_asset_to_gee()`	`utilities/gee_utils.py`	Exports an `ee.Image` to an Earth Engine asset.
`make_asset_public()`	`utilities/gee_utils.py`	Makes a GEE asset readable by public users.
`sync_raster_to_gcs()`	`utilities/gee_utils.py`	Stages a raster through Google Cloud Storage.
`sync_raster_gcs_to_geoserver()`	`utilities/gee_utils.py`	Publishes a GCS-backed raster to GeoServer.
`sync_fc_to_geoserver()`	`computing/utils.py`	Publishes a feature collection to GeoServer.
`save_layer_info_to_db()`	`computing/utils.py`	Creates or updates the `Layer` metadata row.
`update_layer_sync_status()`	`computing/utils.py`	Marks GeoServer or STAC sync completion.

Current Pipelines¶

Family	Main examples	Typical modules
Boundary and MWS	admin boundary, MWS, centroids, connectivity, ZOI	`computing/misc/admin_boundary.py`, `computing/mws/`
Hydrology	precipitation, runoff, ET, delta G, well depth	`computing/mws/`
Land use	LULC, LULC vector, change detection, crop grid, cropping intensity	`computing/lulc/`, `computing/change_detection/`, `computing/cropping_intensity/`
Water bodies	SWB, ponds, wells, SWB-pond merge	`computing/surface_water_bodies/`
Terrain and drainage	terrain raster, terrain clusters, CLART, drainage, stream order, slope, catchment, depressions	`computing/terrain_descriptor/`, `computing/clart/`, `computing/misc/`
Enrichment overlays	NREGA, facilities, aquifer, SOGE, agroecology, mining, CSR, green credit, LCW conflict	`computing/misc/`
Vegetation and tree health	NDVI time series, HLS interpolation, canopy density, canopy height, tree-cover change	`computing/misc/`, `computing/tree_health/`
Plantation and restoration	plantation suitability, restoration opportunity	`computing/plantation/`, `computing/misc/restoration_opportunity.py`
Catalog and delivery	STAC collections and items, layer status, missing layers	`computing/STAC_specs/`, `computing/views.py`

For how these become working pipelines, continue to Develop New Pipelines.