classifyre-cli 0.4.33__tar.gz → 0.4.35__tar.gz

{classifyre_cli-0.4.33 → classifyre_cli-0.4.35}/.gitignore

@@ -63,3 +63,6 @@ dmypy.json
 
 # Local training artifacts
 checkpoints/
+
+# Spark local warehouse (created during dev/testing of lakehouse sources)
+spark-warehouse/

classifyre_cli-0.4.35/.turbo/turbo-build.log

@@ -0,0 +1,3 @@
+$ uv sync
+Resolved 271 packages in 177ms
+Checked 50 packages in 1ms

{classifyre_cli-0.4.33 → classifyre_cli-0.4.35}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: classifyre-cli
-Version: 0.4.33
+Version: 0.4.35
 Summary: Classifyre CLI — scan and classify unstructured data sources
 License: MIT
 Keywords: data,ingestion,metadata,pii,secrets,unstructured

{classifyre_cli-0.4.33 → classifyre_cli-0.4.35}/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@classifyre/cli",
-  "version": "0.4.33",
+  "version": "0.4.35",
   "private": true,
   "scripts": {
     "build": "uv sync",

{classifyre_cli-0.4.33 → classifyre_cli-0.4.35}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "classifyre-cli"
-version = "0.4.33"
+version = "0.4.35"
 description = "Classifyre CLI — scan and classify unstructured data sources"
 readme = "README.md"
 requires-python = ">=3.12"
@@ -202,6 +202,38 @@ 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" },
+]
+iceberg = [
+    # Inspected through Spark (iceberg-spark-runtime JARs via SPARK_JARS_PACKAGES),
+    # sharing the JVM the other lakehouse sources already require.
+    { include-group = "spark" },
+]
+kafka = [
+    "kafka-python>=2.0",
+]
 otel = [
     "opentelemetry-sdk>=1.42.0",
     "opentelemetry-exporter-otlp-proto-http>=1.27.0",