classifyre-cli 0.4.35__tar.gz → 0.4.37__tar.gz

classifyre_cli-0.4.37/.turbo/turbo-build.log

@@ -0,0 +1,3 @@
+$ uv sync
+Resolved 276 packages in 336ms
+Checked 51 packages in 1ms

{classifyre_cli-0.4.35 → classifyre_cli-0.4.37}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: classifyre-cli
-Version: 0.4.35
+Version: 0.4.37
 Summary: Classifyre CLI — scan and classify unstructured data sources
 License: MIT
 Keywords: data,ingestion,metadata,pii,secrets,unstructured

{classifyre_cli-0.4.35 → classifyre_cli-0.4.37}/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@classifyre/cli",
-  "version": "0.4.35",
+  "version": "0.4.37",
   "private": true,
   "scripts": {
     "build": "uv sync",

{classifyre_cli-0.4.35 → classifyre_cli-0.4.37}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "classifyre-cli"
-version = "0.4.35"
+version = "0.4.37"
 description = "Classifyre CLI — scan and classify unstructured data sources"
 readme = "README.md"
 requires-python = ">=3.12"
@@ -51,6 +51,8 @@ dev = [
     "datamodel-code-generator>=0.57.0",
     "pytest>=9.0.3",
     "pytest-asyncio>=0.24.0",
+    # Real SQL engine for the lakehouse source tests (sampling over Parquet).
+    "duckdb>=1.1.0",
 ]
 privacy = [
     "presidio-analyzer>=2.2.362",
@@ -134,6 +136,9 @@ file-processing = [
     "openpyxl>=3.1.5",
     "chardet>=7.4.3",
     "pyarrow>=18.0.0",
+    # Table layer: Iceberg metadata resolution + SQL over Parquet without Spark.
+    "pyiceberg>=0.9.0",
+    "duckdb>=1.1.0",
 ]
 transcription = [
     # CPU audio/video transcription. faster-whisper decodes media via bundled
@@ -202,37 +207,23 @@ youtube = [
     "yt-dlp>=2025.1.0",
     "youtube-transcript-api>=1.0.0",
 ]
-spark = [
-    # Shared PySpark runtime for the lakehouse sources. Requires a JDK (Java 21
-    # LTS) on the host; format JARs (Delta/Iceberg/Hudi) are resolved at runtime
-    # via spark.jars.packages (see SPARK_* env vars). Pinned to the Spark 4.1
-    # minor so the format-JAR coordinates below stay version-matched.
-    #
-    # The `connect` extra pulls in pandas>=2.2, pyarrow, and grpcio, which the
-    # Spark Connect client (Spark Catalog via sc:// URLs) hard-requires at session
-    # build time. Classic/local Spark (Delta/Hudi/Iceberg) does not need them, but
-    # they share this group, so we ship the client deps once for all of them.
-    "pyspark[connect]>=4.1,<4.2",
-]
 delta-lake = [
-    { include-group = "spark" },
-    "delta-spark>=4.0",
-]
-hudi = [
-    # Hudi Spark integration ships as Maven JARs configured via
-    # SPARK_JARS_PACKAGES; only the shared PySpark runtime is needed here.
-    { include-group = "spark" },
-]
-spark-catalog = [
-    { include-group = "spark" },
+    # Pure-Python/Rust Delta reader (delta-rs) — no Spark, no JVM. Table metadata
+    # and data-file lists come from deltalake; rows are sampled through DuckDB
+    # (file-processing group) over the table's Parquet files.
+    { include-group = "file-processing" },
+    "deltalake>=0.25.0",
+    "boto3>=1.43.10,<2.0.0",
 ]
 iceberg = [
-    # Inspected through Spark (iceberg-spark-runtime JARs via SPARK_JARS_PACKAGES),
-    # sharing the JVM the other lakehouse sources already require.
-    { include-group = "spark" },
+    # PyIceberg resolves table metadata and finds Parquet files without Spark;
+    # DuckDB (file-processing group) samples rows from those files.
+    { include-group = "file-processing" },
+    "boto3>=1.43.10,<2.0.0",
 ]
 kafka = [
-    "kafka-python>=2.0",
+    # librdkafka-based client (bundled wheels, no JVM).
+    "confluent-kafka>=2.5.0",
 ]
 otel = [
     "opentelemetry-sdk>=1.42.0",
@@ -353,6 +344,12 @@ module = [
     "re2.*",
     "faster_whisper",
     "faster_whisper.*",
+    "pyiceberg.*",
+    "deltalake.*",
+    "duckdb.*",
+    "confluent_kafka.*",
+    "boto3.*",
+    "botocore.*",
 ]
 ignore_missing_imports = true
 

{classifyre_cli-0.4.35 → classifyre_cli-0.4.37}/src/models/generated_input.py

@@ -47,8 +47,6 @@ class AssetType(StrEnum):
     YOUTUBE = 'YOUTUBE'
     DELTA_LAKE = 'DELTA_LAKE'
     ICEBERG = 'ICEBERG'
-    HUDI = 'HUDI'
-    SPARK_CATALOG = 'SPARK_CATALOG'
     KAFKA = 'KAFKA'
     ELASTICSEARCH = 'ELASTICSEARCH'
     OPENSEARCH = 'OPENSEARCH'
@@ -353,8 +351,6 @@ class Type(StrEnum):
     YOUTUBE = 'YOUTUBE'
     DELTA_LAKE = 'DELTA_LAKE'
     ICEBERG = 'ICEBERG'
-    HUDI = 'HUDI'
-    SPARK_CATALOG = 'SPARK_CATALOG'
     KAFKA = 'KAFKA'
     ELASTICSEARCH = 'ELASTICSEARCH'
     OPENSEARCH = 'OPENSEARCH'
@@ -2868,8 +2864,6 @@ class Type19(StrEnum):
     YOUTUBE = 'YOUTUBE'
     DELTA_LAKE = 'DELTA_LAKE'
     ICEBERG = 'ICEBERG'
-    HUDI = 'HUDI'
-    SPARK_CATALOG = 'SPARK_CATALOG'
     KAFKA = 'KAFKA'
     ELASTICSEARCH = 'ELASTICSEARCH'
     OPENSEARCH = 'OPENSEARCH'
@@ -3117,17 +3111,6 @@ class NotionInput(CoreInput):
     resources: ResourceOverrides | None = None
 
 
-class IcebergCatalogType(StrEnum):
-    """
-    PyIceberg catalog backend type
-    """
-
-    REST = 'REST'
-    HIVE = 'HIVE'
-    GLUE = 'GLUE'
-    SQL = 'SQL'
-
-
 class KafkaSecurityProtocol(StrEnum):
     """
     Kafka client security protocol
@@ -3153,351 +3136,41 @@ class DeltaLakeRequired(BaseModel):
     model_config = ConfigDict(
         extra='forbid',
     )
-    warehouse_path: str = Field(
+    bucket: str = Field(
         ...,
-        description='Root storage location holding Delta tables (e.g. s3a://lake/warehouse, file:///data)',
-    )
-
-
-class DeltaLakeMasked(BaseModel):
-    """
-    Optional object-store credentials for the warehouse location.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    s3_access_key_id: str | None = Field(
-        None, description='S3 access key id for object-store warehouses'
-    )
-    s3_secret_access_key: str | None = Field(None, description='S3 secret access key')
-    s3_session_token: str | None = Field(None, description='Optional S3 session token')
-
-
-class DeltaLakeOptionalConnection(BaseModel):
-    """
-    Delta Lake connection and storage options.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    metastore_uri: str | None = Field(
-        None,
-        description='Hive Metastore thrift URI; enables catalog-based table discovery',
-    )
-    endpoint_url: str | None = Field(
-        None, description='Custom S3-compatible endpoint URL'
+        description='Bucket holding the tables — AWS S3, MinIO, Cloudflare R2, Backblaze B2, Garage, and other S3-compatible endpoints',
     )
-    region: str | None = Field(None, description='Object-store region')
 
 
 class DeltaLakeOptionalScope(BaseModel):
     """
-    Delta Lake database and table selection scope.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    database: str | None = Field(
-        None, description='Single database/namespace to scan (catalog mode)'
-    )
-    include_all_databases: bool | None = Field(
-        False, description='Scan all visible databases except excluded system databases'
-    )
-    exclude_databases: list[str] | None = Field(
-        ['information_schema', 'sys'], description='Database denylist (exact names)'
-    )
-    include_tables: list[str] | None = Field(
-        None,
-        description='Optional table allowlist. Accepted forms: table or database.table',
-    )
-    table_limit: int | None = Field(
-        None, description='Optional cap on number of table assets per database', ge=1
-    )
-    table_paths: list[str] | None = Field(
-        None,
-        description='Explicit Delta table locations to scan when no metastore is configured',
-    )
-
-
-class DeltaLakeOptional(BaseModel):
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    connection: DeltaLakeOptionalConnection | None = None
-    scope: DeltaLakeOptionalScope | None = None
-
-
-class DeltaLakeInput(CoreInput):
-    type: Literal['DELTA_LAKE'] | None = Field(
-        None, description='Type of the asset or source'
-    )
-    required: DeltaLakeRequired
-    masked: DeltaLakeMasked | None = None
-    optional: DeltaLakeOptional | None = None
-    detectors: list[Detector] | None = Field(
-        None, description='Detectors to run on ingested content'
-    )
-    custom_detectors: list[CustomDetectorSelection] | None = Field(
-        None,
-        description='Reusable custom detector IDs selected from the custom detector catalog.',
-    )
-    sampling: SamplingConfig
-    resources: ResourceOverrides | None = None
-
-
-class HudiRequired(BaseModel):
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    warehouse_path: str = Field(
-        ..., description='Root storage location holding Hudi tables'
-    )
-
-
-class HudiMasked(BaseModel):
-    """
-    Optional object-store credentials for the warehouse location.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    s3_access_key_id: str | None = Field(
-        None, description='S3 access key id for object-store warehouses'
-    )
-    s3_secret_access_key: str | None = Field(None, description='S3 secret access key')
-    s3_session_token: str | None = Field(None, description='Optional S3 session token')
-
-
-class HudiOptionalConnection(BaseModel):
-    """
-    Hudi connection and storage options.
+    Delta Lake table selection scope within the bucket.
     """
 
     model_config = ConfigDict(
         extra='forbid',
     )
-    metastore_uri: str | None = Field(
-        None,
-        description='Hive Metastore thrift URI; enables catalog-based table discovery',
-    )
-    endpoint_url: str | None = Field(
-        None, description='Custom S3-compatible endpoint URL'
-    )
-    region: str | None = Field(None, description='Object-store region')
-
-
-class HudiOptionalScope(BaseModel):
-    """
-    Hudi database and table selection scope.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    database: str | None = Field(
-        None, description='Single database/namespace to scan (catalog mode)'
-    )
-    include_all_databases: bool | None = Field(
-        False, description='Scan all visible databases except excluded system databases'
-    )
-    exclude_databases: list[str] | None = Field(
-        ['information_schema', 'sys'], description='Database denylist (exact names)'
-    )
-    include_tables: list[str] | None = Field(
+    prefix: str | None = Field(
         None,
-        description='Optional table allowlist. Accepted forms: table or database.table',
-    )
-    table_limit: int | None = Field(
-        None, description='Optional cap on number of table assets per database', ge=1
+        description='Key prefix to search for Delta Lake tables (e.g. warehouse/). Tables are auto-discovered by their _delta_log/ directory.',
     )
     table_paths: list[str] | None = Field(
         None,
-        description='Explicit Hudi table locations to scan when no metastore is configured',
-    )
-
-
-class HudiOptional(BaseModel):
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    connection: HudiOptionalConnection | None = None
-    scope: HudiOptionalScope | None = None
-
-
-class HudiInput(CoreInput):
-    type: Literal['HUDI'] | None = Field(
-        None, description='Type of the asset or source'
-    )
-    required: HudiRequired
-    masked: HudiMasked | None = None
-    optional: HudiOptional | None = None
-    detectors: list[Detector] | None = Field(
-        None, description='Detectors to run on ingested content'
-    )
-    custom_detectors: list[CustomDetectorSelection] | None = Field(
-        None,
-        description='Reusable custom detector IDs selected from the custom detector catalog.',
-    )
-    sampling: SamplingConfig
-    resources: ResourceOverrides | None = None
-
-
-class SparkCatalogRequired(BaseModel):
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    connect_url: str = Field(
-        ...,
-        description='Spark Connect endpoint (sc://host:15002) or classic master (spark://host:7077)',
-    )
-
-
-class SparkCatalogMasked(BaseModel):
-    """
-    Optional Spark Connect authentication.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    token: str | None = Field(
-        None, description='Bearer token for Spark Connect authentication'
-    )
-
-
-class SparkCatalogOptionalScope(BaseModel):
-    """
-    Spark catalog and table selection scope.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    catalog: str | None = Field(
-        None, description='Spark catalog name to scan (defaults to the session catalog)'
-    )
-    database: str | None = Field(None, description='Single database/namespace to scan')
-    include_all_databases: bool | None = Field(
-        False, description='Scan all visible databases except excluded system databases'
-    )
-    exclude_databases: list[str] | None = Field(
-        ['information_schema', 'sys'], description='Database denylist (exact names)'
-    )
-    include_tables: list[str] | None = Field(
-        None,
-        description='Optional table allowlist. Accepted forms: table or database.table',
+        description='Explicit Delta Lake table root keys or s3:// URIs. When set, auto-discovery under prefix is skipped.',
     )
     table_limit: int | None = Field(
-        None, description='Optional cap on number of table assets per database', ge=1
+        None, description='Optional cap on number of table assets', ge=1
     )
 
 
-class SparkCatalogOptional(BaseModel):
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    scope: SparkCatalogOptionalScope | None = None
-
-
-class SparkCatalogInput(CoreInput):
-    type: Literal['SPARK_CATALOG'] | None = Field(
-        None, description='Type of the asset or source'
-    )
-    required: SparkCatalogRequired
-    masked: SparkCatalogMasked | None = None
-    optional: SparkCatalogOptional | None = None
-    detectors: list[Detector] | None = Field(
-        None, description='Detectors to run on ingested content'
-    )
-    custom_detectors: list[CustomDetectorSelection] | None = Field(
-        None,
-        description='Reusable custom detector IDs selected from the custom detector catalog.',
-    )
-    sampling: SamplingConfig
-    resources: ResourceOverrides | None = None
-
-
 class IcebergRequired(BaseModel):
     model_config = ConfigDict(
         extra='forbid',
     )
-    catalog_type: IcebergCatalogType
-    catalog_uri: str | None = Field(
-        None,
-        description='Catalog URI (REST endpoint, Hive metastore thrift URI, or SQL DSN). Not required for GLUE.',
-    )
-    warehouse: str = Field(
-        ..., description='Warehouse location root (e.g. s3://bucket/warehouse)'
-    )
-
-
-class IcebergMasked(BaseModel):
-    """
-    Optional Iceberg catalog/storage credentials.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    token: str | None = Field(None, description='Bearer token for a REST catalog')
-    aws_access_key_id: str | None = Field(
-        None, description='AWS access key id (Glue/S3)'
-    )
-    aws_secret_access_key: str | None = Field(
-        None, description='AWS secret access key (Glue/S3)'
-    )
-
-
-class IcebergOptionalScope(BaseModel):
-    """
-    Iceberg namespace and table selection scope.
-    """
-
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    namespace: str | None = Field(
-        None, description='Single namespace to scan (dotted form supported)'
-    )
-    include_all_namespaces: bool | None = Field(
-        False, description='Scan all visible namespaces'
-    )
-    include_tables: list[str] | None = Field(
-        None,
-        description='Optional table allowlist. Accepted forms: table or namespace.table',
-    )
-    table_limit: int | None = Field(
-        None, description='Optional cap on number of table assets per namespace', ge=1
-    )
-
-
-class IcebergOptional(BaseModel):
-    model_config = ConfigDict(
-        extra='forbid',
-    )
-    scope: IcebergOptionalScope | None = None
-
-
-class IcebergInput(CoreInput):
-    type: Literal['ICEBERG'] | None = Field(
-        None, description='Type of the asset or source'
-    )
-    required: IcebergRequired
-    masked: IcebergMasked | None = None
-    optional: IcebergOptional | None = None
-    detectors: list[Detector] | None = Field(
-        None, description='Detectors to run on ingested content'
-    )
-    custom_detectors: list[CustomDetectorSelection] | None = Field(
-        None,
-        description='Reusable custom detector IDs selected from the custom detector catalog.',
+    bucket: str = Field(
+        ...,
+        description='Bucket holding the tables — AWS S3, MinIO, Cloudflare R2, Backblaze B2, Garage, and other S3-compatible endpoints',
     )
-    sampling: SamplingConfig
-    resources: ResourceOverrides | None = None
 
 
 class NoAuthentication(BaseModel):
@@ -3881,6 +3554,60 @@ class MeilisearchInput(CoreInput):
     resources: ResourceOverrides | None = None
 
 
+class LakehouseStorageConnection(BaseModel):
+    """
+    S3-compatible storage connection options (AWS S3, MinIO, Cloudflare R2, Backblaze B2, Garage, ...). Mirrors the S3 Compatible Storage source connection settings.
+    """
+
+    model_config = ConfigDict(
+        extra='forbid',
+    )
+    region_name: str | None = Field(
+        None,
+        description='Region (recommended for AWS; required by some S3-compatible providers)',
+    )
+    endpoint_url: AnyUrl | None = Field(
+        None,
+        description='Custom endpoint URL for MinIO/R2/B2/Garage and other S3-compatible providers',
+    )
+    request_timeout_seconds: float | None = Field(
+        30,
+        description='Network timeout in seconds for storage list/read operations',
+        ge=1.0,
+        le=300.0,
+    )
+    max_keys_per_page: int | None = Field(
+        1000,
+        description='Maximum objects requested per provider list API call during table discovery',
+        ge=1,
+        le=1000,
+    )
+    verify_ssl: bool | None = Field(
+        True, description='TLS certificate verification toggle'
+    )
+
+
+class IcebergOptionalScope(BaseModel):
+    """
+    Apache Iceberg table selection scope within the bucket.
+    """
+
+    model_config = ConfigDict(
+        extra='forbid',
+    )
+    prefix: str | None = Field(
+        None,
+        description='Key prefix to search for Apache Iceberg tables (e.g. warehouse/). Tables are auto-discovered by their metadata/ directory.',
+    )
+    table_paths: list[str] | None = Field(
+        None,
+        description='Explicit Apache Iceberg table root keys or s3:// URIs. When set, auto-discovery under prefix is skipped.',
+    )
+    table_limit: int | None = Field(
+        None, description='Optional cap on number of table assets', ge=1
+    )
+
+
 class YouTubeInput(CoreInput):
     type: Literal['YOUTUBE'] | None = Field(
         None, description='Type of the asset or source'
@@ -3899,6 +3626,58 @@ class YouTubeInput(CoreInput):
     resources: ResourceOverrides | None = None
 
 
+class DeltaLakeOptional(BaseModel):
+    model_config = ConfigDict(
+        extra='forbid',
+    )
+    connection: LakehouseStorageConnection | None = None
+    scope: DeltaLakeOptionalScope | None = None
+
+
+class DeltaLakeInput(CoreInput):
+    type: Literal['DELTA_LAKE'] | None = Field(
+        None, description='Type of the asset or source'
+    )
+    required: DeltaLakeRequired
+    masked: S3CompatibleStorageMasked | None = None
+    optional: DeltaLakeOptional | None = None
+    detectors: list[Detector] | None = Field(
+        None, description='Detectors to run on ingested content'
+    )
+    custom_detectors: list[CustomDetectorSelection] | None = Field(
+        None,
+        description='Reusable custom detector IDs selected from the custom detector catalog.',
+    )
+    sampling: SamplingConfig
+    resources: ResourceOverrides | None = None
+
+
+class IcebergOptional(BaseModel):
+    model_config = ConfigDict(
+        extra='forbid',
+    )
+    connection: LakehouseStorageConnection | None = None
+    scope: IcebergOptionalScope | None = None
+
+
+class IcebergInput(CoreInput):
+    type: Literal['ICEBERG'] | None = Field(
+        None, description='Type of the asset or source'
+    )
+    required: IcebergRequired
+    masked: S3CompatibleStorageMasked | None = None
+    optional: IcebergOptional | None = None
+    detectors: list[Detector] | None = Field(
+        None, description='Detectors to run on ingested content'
+    )
+    custom_detectors: list[CustomDetectorSelection] | None = Field(
+        None,
+        description='Reusable custom detector IDs selected from the custom detector catalog.',
+    )
+    sampling: SamplingConfig
+    resources: ResourceOverrides | None = None
+
+
 class SourceInput(
     RootModel[
         SlackInput
@@ -3926,8 +3705,6 @@ class SourceInput(
         | YouTubeInput
         | DeltaLakeInput
         | IcebergInput
-        | HudiInput
-        | SparkCatalogInput
         | KafkaInput
         | ElasticsearchInput
         | OpenSearchInput
@@ -3960,8 +3737,6 @@ class SourceInput(
         | YouTubeInput
         | DeltaLakeInput
         | IcebergInput
-        | HudiInput
-        | SparkCatalogInput
         | KafkaInput
         | ElasticsearchInput
         | OpenSearchInput