etlplus 0.5.2__py3-none-any.whl → 0.9.1__py3-none-any.whl

etlplus/api/README.md

@@ -7,8 +7,8 @@ paginated REST endpoints.
 - Supports page-, offset-, and cursor-based pagination via `PaginationConfig`
 - Simple bearer-auth credentials via `EndpointCredentialsBearer`
 - Convenience helpers to extract records from nested JSON payloads
-- Returns the shared `JSONRecords` alias (a list of `JSONDict`) for paginated responses, matching
-  the rest of the library.
+- Returns paginated JSON payloads (lists of record dictionaries) consistent with the rest of the
+  library.
 
 Back to project overview: see the top-level [README](../../README.md).
 
@@ -29,7 +29,6 @@ import requests
 from etlplus.api import (
   EndpointClient,
   EndpointCredentialsBearer,
-  JSONRecords,
 )
 
 auth = EndpointCredentialsBearer(
@@ -54,12 +53,12 @@ client = EndpointClient(
 
 # Page-based pagination
 pg: PaginationConfig = {"type": "page", "page_size": 100}
-rows: JSONRecords = client.paginate("list", pagination=pg)
+rows = client.paginate("list", pagination=pg)
 for row in rows:
   print(row)
 ```
 
-### Overriding rate limits per call
+### Overriding Rate Limits Per Call
 
 When a client is constructed with ``rate_limit`` metadata you can still tweak the pacing for
 individual calls by passing ``rate_limit_overrides`` to ``paginate``/``paginate_iter``. The
@@ -101,10 +100,10 @@ If the API responds like this:
 
 If the response is a list at the top level, you can omit `records_path`.
 
-## Cursor-based pagination example
+## Cursor-Based Pagination Example
 
 ```python
-from etlplus.api import EndpointClient, PaginationConfig, JSONRecords
+from etlplus.api import EndpointClient, PaginationConfig
 
 client = EndpointClient(
     base_url="https://api.example.com/v1",
@@ -125,7 +124,7 @@ pg: PaginationConfig = {
     # "start_cursor": "abc123",
 }
 
-rows: JSONRecords = client.paginate("list", pagination=pg)
+rows = client.paginate("list", pagination=pg)
 for row in rows:
     process(row)
 ```
@@ -192,36 +191,35 @@ client = EndpointClient(
 providers can fall back to their own defaults. If you already possess a static token, attach it to a
 `requests.Session` manually rather than instantiating `EndpointCredentialsBearer`.
 
-## Errors and rate limiting
+## Errors and Rate Limiting
 
 - Errors: `ApiRequestError`, `ApiAuthError`, and `PaginationError` (in `etlplus/api/errors.py`)
   include an `as_dict()` helper for structured logs.
-- Rate limiting: `RateLimiter` and its `resolve_sleep_seconds` helper (in
-  `etlplus/api/rate_limiter.py`) derives fixed sleeps or `max_per_sec` windows. The paginator now
-  builds a `RateLimiter` whenever the effective delay comes from
-  `rate_limit`/`rate_limit_overrides`, so each page fetch sleeps before making another HTTP call.
-  Passing `rate_limit_overrides` to `paginate*` lets you momentarily speed up or slow down a single
-  request without mutating the client-wide defaults.
-
-## Types and transport
-
-- Types: pagination config helpers live in `etlplus/api/paginator.py`; retry helpers (including
-  `RetryPolicy`) live in `etlplus/api/retry_manager.py`; rate-limit helpers live in
-  `etlplus/api/rate_limiter.py`. These are all re-exported from `etlplus.api` for convenience.
+- Rate limiting: `RateLimiter` (in `etlplus/api/rate_limiting/rate_limiter.py`) derives fixed sleeps
+  or `max_per_sec` windows. The paginator now builds a `RateLimiter` whenever the effective delay
+  comes from `rate_limit`/`rate_limit_overrides`, so each page fetch sleeps before making another
+  HTTP call. Passing `rate_limit_overrides` to `paginate*` lets you momentarily speed up or slow
+  down a single request without mutating the client-wide defaults.
+
+## Types and Transport
+
+- Types: pagination config helpers live in `etlplus/api/pagination/paginator.py`; retry helpers
+  (including `RetryPolicy`) live in `etlplus/api/retry_manager.py`; rate-limit helpers live in
+  `etlplus/api/rate_limiting/rate_limiter.py`. These are all re-exported from `etlplus.api` for
+  convenience.
 - Transport/session: `etlplus/api/transport.py` contains the HTTP adapter helpers and
   `etlplus/api/request_manager.py` wraps `requests` sessions plus retry orchestration. Advanced
   users may consult those modules to adapt behavior.
 
-## Supporting modules
+## Supporting Modules
 
 - `etlplus.api.types` collects friendly aliases such as `Headers`, `Params`, `Url`, and
   `RateLimitOverrides` (whose values accept numeric override inputs) so endpoint helpers share the
   same type vocabulary.
-- `etlplus.utils` exposes lightweight helpers used across the project, including CLI-friendly
-  functions like `json_type`/`print_json` plus numeric coercion utilities (`to_float`,
-  `to_positive_int`, etc.).
+- `etlplus.utils` exposes lightweight helpers used across the project, including `print_json` and
+  numeric coercion utilities (`to_float`, `to_positive_int`, etc.).
 
-## Minimal contract
+## Minimal Contract
 
 - Inputs
   - `base_url: str`, `endpoints: dict[str, str]`