airbyte-cdk 6.34.1.dev0__py3-none-any.whl → 6.34.1.dev1__py3-none-any.whl

airbyte_cdk/sources/declarative/declarative_component_schema.yaml

@@ -40,6 +40,8 @@ properties:
     "$ref": "#/definitions/Spec"
   concurrency_level:
     "$ref": "#/definitions/ConcurrencyLevel"
+  api_budget:
+    "$ref": "#/definitions/HTTPAPIBudget"
   metadata:
     type: object
     description: For internal Airbyte use only - DO NOT modify manually. Used by consumers of declarative manifests for storing related metadata.
@@ -80,7 +82,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - "{{ record['updates'] }}"
           - "{{ record['MetaData']['LastUpdatedTime'] }}"
@@ -794,7 +795,7 @@ definitions:
         description: This option is used to adjust the upper and lower boundaries of each datetime window to beginning and end of the provided target period (day, week, month)
         type: object
         required:
-        - target
+          - target
         properties:
           target:
             title: Target
@@ -1365,6 +1366,170 @@ definitions:
       $parameters:
         type: object
         additional_properties: true
+  HTTPAPIBudget:
+    title: HTTP API Budget
+    description: >
+      Defines how many requests can be made to the API in a given time frame. `HTTPAPIBudget` extracts the remaining
+      call count and the reset time from HTTP response headers using the header names provided by
+      `ratelimit_remaining_header` and `ratelimit_reset_header`. Only requests using `HttpRequester`
+      are rate-limited; custom components that bypass `HttpRequester` are not covered by this budget.
+    type: object
+    required:
+      - type
+      - policies
+    properties:
+      type:
+        type: string
+        enum: [HTTPAPIBudget]
+      policies:
+        title: Policies
+        description: List of call rate policies that define how many calls are allowed.
+        type: array
+        items:
+          anyOf:
+            - "$ref": "#/definitions/FixedWindowCallRatePolicy"
+            - "$ref": "#/definitions/MovingWindowCallRatePolicy"
+            - "$ref": "#/definitions/UnlimitedCallRatePolicy"
+      ratelimit_reset_header:
+        title: Rate Limit Reset Header
+        description: The HTTP response header name that indicates when the rate limit resets.
+        type: string
+        default: "ratelimit-reset"
+      ratelimit_remaining_header:
+        title: Rate Limit Remaining Header
+        description: The HTTP response header name that indicates the number of remaining allowed calls.
+        type: string
+        default: "ratelimit-remaining"
+      status_codes_for_ratelimit_hit:
+        title: Status Codes for Rate Limit Hit
+        description: List of HTTP status codes that indicate a rate limit has been hit.
+        type: array
+        items:
+          type: integer
+        default: [429]
+    additionalProperties: true
+  FixedWindowCallRatePolicy:
+    title: Fixed Window Call Rate Policy
+    description: A policy that allows a fixed number of calls within a specific time window.
+    type: object
+    required:
+      - type
+      - period
+      - call_limit
+      - matchers
+    properties:
+      type:
+        type: string
+        enum: [FixedWindowCallRatePolicy]
+      period:
+        title: Period
+        description: The time interval for the rate limit window.
+        type: string
+      call_limit:
+        title: Call Limit
+        description: The maximum number of calls allowed within the period.
+        type: integer
+      matchers:
+        title: Matchers
+        description: List of matchers that define which requests this policy applies to.
+        type: array
+        items:
+          "$ref": "#/definitions/HttpRequestRegexMatcher"
+    additionalProperties: true
+  MovingWindowCallRatePolicy:
+    title: Moving Window Call Rate Policy
+    description: A policy that allows a fixed number of calls within a moving time window.
+    type: object
+    required:
+      - type
+      - rates
+      - matchers
+    properties:
+      type:
+        type: string
+        enum: [MovingWindowCallRatePolicy]
+      rates:
+        title: Rates
+        description: List of rates that define the call limits for different time intervals.
+        type: array
+        items:
+          "$ref": "#/definitions/Rate"
+      matchers:
+        title: Matchers
+        description: List of matchers that define which requests this policy applies to.
+        type: array
+        items:
+          "$ref": "#/definitions/HttpRequestRegexMatcher"
+    additionalProperties: true
+  UnlimitedCallRatePolicy:
+    title: Unlimited Call Rate Policy
+    description: A policy that allows unlimited calls for specific requests.
+    type: object
+    required:
+      - type
+      - matchers
+    properties:
+      type:
+        type: string
+        enum: [UnlimitedCallRatePolicy]
+      matchers:
+        title: Matchers
+        description: List of matchers that define which requests this policy applies to.
+        type: array
+        items:
+          "$ref": "#/definitions/HttpRequestRegexMatcher"
+    additionalProperties: true
+  Rate:
+    title: Rate
+    description: Defines a rate limit with a specific number of calls allowed within a time interval.
+    type: object
+    required:
+      - limit
+      - interval
+    properties:
+      limit:
+        title: Limit
+        description: The maximum number of calls allowed within the interval.
+        type: integer
+      interval:
+        title: Interval
+        description: The time interval for the rate limit.
+        type: string
+        examples:
+          - "PT1H"
+          - "P1D"
+    additionalProperties: true
+  HttpRequestRegexMatcher:
+    title: HTTP Request Matcher
+    description: >
+      Matches HTTP requests based on method, base URL, URL path pattern, query parameters, and headers.
+      Use `url_base` to specify the scheme and host (without trailing slash) and
+      `url_path_pattern` to apply a regex to the request path.
+    type: object
+    properties:
+      method:
+        title: Method
+        description: The HTTP method to match (e.g., GET, POST).
+        type: string
+      url_base:
+        title: URL Base
+        description: The base URL (scheme and host, e.g. "https://api.example.com") to match.
+        type: string
+      url_path_pattern:
+        title: URL Path Pattern
+        description: A regular expression pattern to match the URL path.
+        type: string
+      params:
+        title: Parameters
+        description: The query parameters to match.
+        type: object
+        additionalProperties: true
+      headers:
+        title: Headers
+        description: The headers to match.
+        type: object
+        additionalProperties: true
+    additionalProperties: true
   DefaultErrorHandler:
     title: Default Error Handler
     description: Component defining how to handle errors. Default behavior includes only retrying server errors (HTTP 5XX) and too many requests (HTTP 429) with an exponential backoff.
@@ -1549,7 +1714,6 @@ definitions:
         anyOf:
           - "$ref": "#/definitions/JsonDecoder"
           - "$ref": "#/definitions/XmlDecoder"
-          - "$ref": "#/definitions/CompositeRawDecoder"
       $parameters:
         type: object
         additionalProperties: true
@@ -1611,7 +1775,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - "/products"
           - "/quotes/{{ stream_partition['id'] }}/quote_line_groups"
@@ -1661,7 +1824,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - |
             [{"clause": {"type": "timestamp", "operator": 10, "parameters":
@@ -1679,7 +1841,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - sort_order: "ASC"
             sort_field: "CREATED_AT"
@@ -1700,7 +1861,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - Output-Format: JSON
           - Version: "{{ config['version'] }}"
@@ -1717,7 +1877,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - unit: "day"
           - query: 'last_event_time BETWEEN TIMESTAMP "{{ stream_interval.start_time }}" AND TIMESTAMP "{{ stream_interval.end_time }}"'
@@ -2072,7 +2231,6 @@ definitions:
         interpolation_context:
           - config
           - record
-          - stream_state
           - stream_slice
       new:
         type: string
@@ -2086,7 +2244,6 @@ definitions:
         interpolation_context:
           - config
           - record
-          - stream_state
           - stream_slice
       $parameters:
         type: object
@@ -2133,23 +2290,6 @@ definitions:
       $parameters:
         type: object
         additionalProperties: true
-  GzipJsonDecoder:
-    title: GzipJson Decoder
-    description: Use this if the response is Gzip compressed Json.
-    type: object
-    additionalProperties: true
-    required:
-      - type
-    properties:
-      type:
-        type: string
-        enum: [GzipJsonDecoder]
-      encoding:
-        type: string
-        default: utf-8
-      $parameters:
-        type: object
-        additionalProperties: true
   ZipfileDecoder:
     title: Zipfile Decoder
     description: Decoder for response data that is returned as zipfile(s).
@@ -2157,19 +2297,19 @@ definitions:
     additionalProperties: true
     required:
       - type
-      - parser
+      - decoder
     properties:
       type:
         type: string
         enum: [ZipfileDecoder]
-      parser:
+      decoder:
         title: Parser
         description: Parser to parse the decompressed data from the zipfile(s).
         anyOf:
-          - "$ref": "#/definitions/GzipParser"
-          - "$ref": "#/definitions/JsonParser"
-          - "$ref": "#/definitions/JsonLineParser"
-          - "$ref": "#/definitions/CsvParser"
+          - "$ref": "#/definitions/CsvDecoder"
+          - "$ref": "#/definitions/GzipDecoder"
+          - "$ref": "#/definitions/JsonDecoder"
+          - "$ref": "#/definitions/JsonlDecoder"
   ListPartitionRouter:
     title: List Partition Router
     description: A Partition router that specifies a list of attributes where each attribute describes a portion of the complete data set for a stream. During a sync, each value is iterated over and can be used as input to outbound API requests.
@@ -2753,7 +2893,6 @@ definitions:
           - stream_interval
           - stream_partition
           - stream_slice
-          - stream_state
         examples:
           - "{{ record['created_at'] >= stream_interval['start_time'] }}"
           - "{{ record.status in ['active', 'expired'] }}"
@@ -2847,25 +2986,35 @@ definitions:
         enum: [RequestPath]
   RequestOption:
     title: Request Option
-    description: Specifies the key field and where in the request a component's value should be injected.
+    description: Specifies the key field or path and where in the request a component's value should be injected.
     type: object
     required:
       - type
-      - field_name
       - inject_into
     properties:
       type:
         type: string
         enum: [RequestOption]
       field_name:
-        title: Request Option
-        description: Configures which key should be used in the location that the descriptor is being injected into
+        title: Field Name
+        description: Configures which key should be used in the location that the descriptor is being injected into. We hope to eventually deprecate this field in favor of `field_path` for all request_options, but must currently maintain it for backwards compatibility in the Builder.
         type: string
         examples:
           - segment_id
         interpolation_context:
           - config
           - parameters
+      field_path:
+        title: Field Path
+        description: Configures a path to be used for nested structures in JSON body requests (e.g. GraphQL queries)
+        type: array
+        items:
+          type: string
+        examples:
+          - ["data", "viewer", "id"]
+        interpolation_context:
+          - config
+          - parameters
       inject_into:
         title: Inject Into
         description: Configures where the descriptor should be set on the HTTP requests. Note that request parameters that are already encoded in the URL path will not be duplicated.
@@ -2992,79 +3141,39 @@ definitions:
         description: Component decoding the response so records can be extracted.
         anyOf:
           - "$ref": "#/definitions/CustomDecoder"
+          - "$ref": "#/definitions/CsvDecoder"
+          - "$ref": "#/definitions/GzipDecoder"
           - "$ref": "#/definitions/JsonDecoder"
           - "$ref": "#/definitions/JsonlDecoder"
           - "$ref": "#/definitions/IterableDecoder"
           - "$ref": "#/definitions/XmlDecoder"
-          - "$ref": "#/definitions/GzipJsonDecoder"
-          - "$ref": "#/definitions/CompositeRawDecoder"
           - "$ref": "#/definitions/ZipfileDecoder"
       $parameters:
         type: object
         additionalProperties: true
-  CompositeRawDecoder:
-    description: "(This is experimental, use at your own risk)"
+  GzipDecoder:
     type: object
     required:
       - type
-      - parser
+      - decoder
     properties:
       type:
         type: string
-        enum: [CompositeRawDecoder]
-      parser:
-        anyOf:
-          - "$ref": "#/definitions/GzipParser"
-          - "$ref": "#/definitions/JsonParser"
-          - "$ref": "#/definitions/JsonLineParser"
-          - "$ref": "#/definitions/CsvParser"
-  #    PARSERS
-  GzipParser:
-    type: object
-    required:
-      - type
-      - inner_parser
-    properties:
-      type:
-        type: string
-        enum: [GzipParser]
-      inner_parser:
+        enum: [GzipDecoder]
+      decoder:
         anyOf:
-          - "$ref": "#/definitions/JsonLineParser"
-          - "$ref": "#/definitions/CsvParser"
-          - "$ref": "#/definitions/JsonParser"
-  JsonParser:
-    title: JsonParser
-    description: Parser used for parsing str, bytes, or bytearray data and returning data in a dictionary format.
-    type: object
-    required:
-      - type
-    properties:
-      type:
-        type: string
-        enum: [JsonParser]
-      encoding:
-        type: string
-        default: utf-8
-  JsonLineParser:
-    type: object
-    required:
-      - type
-    properties:
-      type:
-        type: string
-        enum: [JsonLineParser]
-      encoding:
-        type: string
-        default: utf-8
-  CsvParser:
+          - "$ref": "#/definitions/CsvDecoder"
+          - "$ref": "#/definitions/GzipDecoder"
+          - "$ref": "#/definitions/JsonDecoder"
+          - "$ref": "#/definitions/JsonlDecoder"
+  CsvDecoder:
     type: object
     required:
       - type
     properties:
       type:
         type: string
-        enum: [CsvParser]
+        enum: [CsvDecoder]
       encoding:
         type: string
         default: utf-8
@@ -3192,24 +3301,24 @@ definitions:
         description: Component decoding the response so records can be extracted.
         anyOf:
           - "$ref": "#/definitions/CustomDecoder"
+          - "$ref": "#/definitions/CsvDecoder"
+          - "$ref": "#/definitions/GzipDecoder"
           - "$ref": "#/definitions/JsonDecoder"
           - "$ref": "#/definitions/JsonlDecoder"
           - "$ref": "#/definitions/IterableDecoder"
           - "$ref": "#/definitions/XmlDecoder"
-          - "$ref": "#/definitions/GzipJsonDecoder"
-          - "$ref": "#/definitions/CompositeRawDecoder"
           - "$ref": "#/definitions/ZipfileDecoder"
       download_decoder:
         title: Download Decoder
         description: Component decoding the download response so records can be extracted.
         anyOf:
           - "$ref": "#/definitions/CustomDecoder"
+          - "$ref": "#/definitions/CsvDecoder"
+          - "$ref": "#/definitions/GzipDecoder"
           - "$ref": "#/definitions/JsonDecoder"
           - "$ref": "#/definitions/JsonlDecoder"
           - "$ref": "#/definitions/IterableDecoder"
           - "$ref": "#/definitions/XmlDecoder"
-          - "$ref": "#/definitions/GzipJsonDecoder"
-          - "$ref": "#/definitions/CompositeRawDecoder"
           - "$ref": "#/definitions/ZipfileDecoder"
       $parameters:
         type: object
@@ -3571,12 +3680,6 @@ interpolation:
     - title: stream_slice
       description: This variable is deprecated. Use stream_interval or stream_partition instead.
       type: object
-    - title: stream_state
-      description: The current state of the stream. The object's keys are defined by the incremental sync's cursor_field the and partition router's values.
-      type: object
-      examples:
-        - created_at: "2020-01-01 00:00:00.000+00:00"
-        - updated_at: "2020-01-02 00:00:00.000+00:00"
   macros:
     - title: now_utc
       description: Returns the current date and time in the UTC timezone.

airbyte_cdk/sources/declarative/declarative_stream.py

@@ -138,7 +138,9 @@ class DeclarativeStream(Stream):
         """
         :param: stream_state We knowingly avoid using stream_state as we want cursors to manage their own state.
         """
-        if stream_slice is None or stream_slice == {}:
+        if stream_slice is None or (
+            not isinstance(stream_slice, StreamSlice) and stream_slice == {}
+        ):
             # As the parameter is Optional, many would just call `read_records(sync_mode)` during testing without specifying the field
             # As part of the declarative model without custom components, this should never happen as the CDK would wire up a
             # SinglePartitionRouter that would create this StreamSlice properly

airbyte_cdk/sources/declarative/decoders/__init__.py

@@ -10,10 +10,8 @@ from airbyte_cdk.sources.declarative.decoders.composite_raw_decoder import (
 )
 from airbyte_cdk.sources.declarative.decoders.decoder import Decoder
 from airbyte_cdk.sources.declarative.decoders.json_decoder import (
-    GzipJsonDecoder,
     IterableDecoder,
     JsonDecoder,
-    JsonlDecoder,
 )
 from airbyte_cdk.sources.declarative.decoders.noop_decoder import NoopDecoder
 from airbyte_cdk.sources.declarative.decoders.pagination_decoder_decorator import (
@@ -27,9 +25,7 @@ __all__ = [
     "CompositeRawDecoder",
     "JsonDecoder",
     "JsonParser",
-    "JsonlDecoder",
     "IterableDecoder",
-    "GzipJsonDecoder",
     "NoopDecoder",
     "PaginationDecoderDecorator",
     "XmlDecoder",

airbyte_cdk/sources/declarative/decoders/composite_raw_decoder.py

@@ -1,5 +1,6 @@
 import csv
 import gzip
+import io
 import json
 import logging
 from abc import ABC, abstractmethod
@@ -130,11 +131,15 @@ class CompositeRawDecoder(Decoder):
     """
 
     parser: Parser
+    stream_response: bool = True
 
     def is_stream_response(self) -> bool:
-        return True
+        return self.stream_response
 
     def decode(
         self, response: requests.Response
     ) -> Generator[MutableMapping[str, Any], None, None]:
-        yield from self.parser.parse(data=response.raw)  # type: ignore[arg-type]
+        if self.is_stream_response():
+            yield from self.parser.parse(data=response.raw)  # type: ignore[arg-type]
+        else:
+            yield from self.parser.parse(data=io.BytesIO(response.content))

airbyte_cdk/sources/declarative/decoders/json_decoder.py

@@ -10,21 +10,24 @@ from typing import Any, Generator, List, Mapping, MutableMapping, Optional
 import orjson
 import requests
 
+from airbyte_cdk.sources.declarative.decoders import CompositeRawDecoder, JsonParser
 from airbyte_cdk.sources.declarative.decoders.decoder import Decoder
 
 logger = logging.getLogger("airbyte")
 
 
-@dataclass
 class JsonDecoder(Decoder):
     """
     Decoder strategy that returns the json-encoded content of a response, if any.
+
+    Usually, we would try to instantiate the equivalent `CompositeRawDecoder(parser=JsonParser(), stream_response=False)` but there were specific historical behaviors related to the JsonDecoder that we didn't know if we could remove like the fallback on {} in case of errors.
     """
 
-    parameters: InitVar[Mapping[str, Any]]
+    def __init__(self, parameters: Mapping[str, Any]):
+        self._decoder = CompositeRawDecoder(parser=JsonParser(), stream_response=False)
 
     def is_stream_response(self) -> bool:
-        return False
+        return self._decoder.is_stream_response()
 
     def decode(
         self, response: requests.Response
@@ -32,25 +35,16 @@ class JsonDecoder(Decoder):
         """
         Given the response is an empty string or an emtpy list, the function will return a generator with an empty mapping.
         """
+        has_yielded = False
         try:
-            body_json = response.json()
-            yield from self.parse_body_json(body_json)
-        except requests.exceptions.JSONDecodeError:
-            logger.warning(
-                f"Response cannot be parsed into json: {response.status_code=}, {response.text=}"
-            )
+            for element in self._decoder.decode(response):
+                yield element
+                has_yielded = True
+        except Exception:
             yield {}
 
-    @staticmethod
-    def parse_body_json(
-        body_json: MutableMapping[str, Any] | List[MutableMapping[str, Any]],
-    ) -> Generator[MutableMapping[str, Any], None, None]:
-        if not isinstance(body_json, list):
-            body_json = [body_json]
-        if len(body_json) == 0:
+        if not has_yielded:
             yield {}
-        else:
-            yield from body_json
 
 
 @dataclass
@@ -69,43 +63,3 @@ class IterableDecoder(Decoder):
     ) -> Generator[MutableMapping[str, Any], None, None]:
         for line in response.iter_lines():
             yield {"record": line.decode()}
-
-
-@dataclass
-class JsonlDecoder(Decoder):
-    """
-    Decoder strategy that returns the json-encoded content of the response, if any.
-    """
-
-    parameters: InitVar[Mapping[str, Any]]
-
-    def is_stream_response(self) -> bool:
-        return True
-
-    def decode(
-        self, response: requests.Response
-    ) -> Generator[MutableMapping[str, Any], None, None]:
-        # TODO???: set delimiter? usually it is `\n` but maybe it would be useful to set optional?
-        #  https://github.com/airbytehq/airbyte-internal-issues/issues/8436
-        for record in response.iter_lines():
-            yield orjson.loads(record)
-
-
-@dataclass
-class GzipJsonDecoder(JsonDecoder):
-    encoding: Optional[str]
-
-    def __post_init__(self, parameters: Mapping[str, Any]) -> None:
-        if self.encoding:
-            try:
-                codecs.lookup(self.encoding)
-            except LookupError:
-                raise ValueError(
-                    f"Invalid encoding '{self.encoding}'. Please check provided encoding"
-                )
-
-    def decode(
-        self, response: requests.Response
-    ) -> Generator[MutableMapping[str, Any], None, None]:
-        raw_string = decompress(response.content).decode(encoding=self.encoding or "utf-8")
-        yield from self.parse_body_json(orjson.loads(raw_string))

airbyte_cdk/sources/declarative/extractors/record_selector.py

@@ -41,6 +41,7 @@ class RecordSelector(HttpSelector):
     _name: Union[InterpolatedString, str] = field(init=False, repr=False, default="")
     record_filter: Optional[RecordFilter] = None
     transformations: List[RecordTransformation] = field(default_factory=lambda: [])
+    transform_before_filtering: bool = False
 
     def __post_init__(self, parameters: Mapping[str, Any]) -> None:
         self._parameters = parameters
@@ -104,9 +105,17 @@ class RecordSelector(HttpSelector):
         Until we decide to move this logic away from the selector, we made this method public so that users like AsyncJobRetriever could
         share the logic of doing transformations on a set of records.
         """
-        filtered_data = self._filter(all_data, stream_state, stream_slice, next_page_token)
-        transformed_data = self._transform(filtered_data, stream_state, stream_slice)
-        normalized_data = self._normalize_by_schema(transformed_data, schema=records_schema)
+        if self.transform_before_filtering:
+            transformed_data = self._transform(all_data, stream_state, stream_slice)
+            transformed_filtered_data = self._filter(
+                transformed_data, stream_state, stream_slice, next_page_token
+            )
+        else:
+            filtered_data = self._filter(all_data, stream_state, stream_slice, next_page_token)
+            transformed_filtered_data = self._transform(filtered_data, stream_state, stream_slice)
+        normalized_data = self._normalize_by_schema(
+            transformed_filtered_data, schema=records_schema
+        )
         for data in normalized_data:
             yield Record(data=data, stream_name=self.name, associated_slice=stream_slice)