classifyre-cli 0.7.3__py3-none-any.whl → 0.7.4__py3-none-any.whl

{classifyre_cli-0.7.3.dist-info → classifyre_cli-0.7.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: classifyre-cli
-Version: 0.7.3
+Version: 0.7.4
 Summary: Classifyre CLI — scan and classify unstructured data sources
 License: MIT
 Keywords: data,ingestion,metadata,pii,secrets,unstructured

{classifyre_cli-0.7.3.dist-info → classifyre_cli-0.7.4.dist-info}/RECORD

@@ -191,7 +191,7 @@ src/sources/sqlite/source.py,sha256=e7A4vurxKEwb4oxpgEK8vNvCabxxFfDZZY4DvgcgnWM,
 src/sources/tableau/__init__.py,sha256=Di1Bh9s9SYrUYg4xZhhAhl5HfoTt0B6WQB5SxVrQhbY,63
 src/sources/tableau/source.py,sha256=zTOCwobJgE4E-3RBV-twKL_Nxrm60mBvwdI32dIDwgs,37139
 src/sources/trino/__init__.py,sha256=GFQDJ82MG8VACM8e2XzHIj-MyXBAwoKSXPdOJx8-NW4,59
-src/sources/trino/source.py,sha256=WrTIYOoc5yFCf7JJZBHvjpcBq_YwIzWndd9P3MRvs0k,12650
+src/sources/trino/source.py,sha256=3s9oP4Hn7o3rJYfCvKQERmfhvcthA7CC7WT2eMZ-Z4s,12890
 src/sources/wordpress/__init__.py,sha256=T7v2TGUYiKzARpYk9bynxChY0P9GDDxpdbBz34x9g6o,67
 src/sources/wordpress/source.py,sha256=Bhyopqhq-AuQFTsQzaLUrG4P8oX-gcAA26jLyLs6d0A,29739
 src/sources/youtube/__init__.py,sha256=TqLfewxoisr752G4shsgslPFxc0LSf6q7-EAbQP8oGI,63
@@ -208,7 +208,7 @@ src/utils/file_parser.py,sha256=oiWZ8SEqz1HvnUl0qoQnfE8qW6CVIN9aSsftN2hErY8,1027
 src/utils/file_to_images.py,sha256=rU9uATckg8JDxHZLobjOJdwkrC2CZxegMS4hv76txTI,4782
 src/utils/hashing.py,sha256=T0GL1FWtG32QJID5RN3X-TzHHkolnNK6EfADvX6sdlE,2490
 src/utils/lakehouse_layout.py,sha256=EilfOlOfrTAF2fn0S6v2_EM_eoYEtqt5Ja-1E7QSM24,4706
-src/utils/lakehouse_storage.py,sha256=vAKjpvmMBvr1aFOCIgyfQF-Tls6HBfJSLdh6SIb5oiQ,7712
+src/utils/lakehouse_storage.py,sha256=r8ltJypHVoZpr7Jry32oW_R-20nF2zW5qle5YPUrzRo,8115
 src/utils/lakehouse_tables.py,sha256=Hb4gkoV4jDFmGSeCTXv968wedtpZPM-Sta2pQwu01R8,19751
 src/utils/legacy_office.py,sha256=LCDiU7E-ZUxHgtQieihQWCXNXOtjiUpJEExWjme-dPI,9205
 src/utils/payload.py,sha256=ewvSABOLSiIGnrq5A9wfOeFZnVJ_TXtSmEOxDYcE58U,6159
@@ -226,7 +226,7 @@ src/utils/urn.py,sha256=gqSl0H0FBZZ3cJbpDWX1ewGcyZz2MrNVTIQdoFrQMM0,12622
 src/utils/uv_sync.py,sha256=VZ0TZ7t3_pSjxOb20BxUs2XLVjigXnVqa5y69XWlnE8,9307
 src/utils/validation.py,sha256=5NHsgyZwyEONRu9_ADxUAQNPvbMnpMhKL-hpgFEVMm8,3837
 src/utils/video_processing.py,sha256=B6Br9ybFEa_ioglJGo0P9Fmp1e2B95C24joAktwrjww,11293
-classifyre_cli-0.7.3.dist-info/METADATA,sha256=Hxaiv_5iV9RbxFD7tUm_0v66yOUEpBJJRq5IWiCyZso,5850
-classifyre_cli-0.7.3.dist-info/WHEEL,sha256=2iZrJkDIkeDh0k9SikKMqFuRP-ua6mb4Ir1AsER4BEw,87
-classifyre_cli-0.7.3.dist-info/entry_points.txt,sha256=edxllyMiIFfajgCZ3dTRACXYGC34ZX1AYTezMSR5E_8,45
-classifyre_cli-0.7.3.dist-info/RECORD,,
+classifyre_cli-0.7.4.dist-info/METADATA,sha256=z7IMi9fyOUP4vwTdkKIiaiV7sZxLJNIJOx1l1enKlbc,5850
+classifyre_cli-0.7.4.dist-info/WHEEL,sha256=2iZrJkDIkeDh0k9SikKMqFuRP-ua6mb4Ir1AsER4BEw,87
+classifyre_cli-0.7.4.dist-info/entry_points.txt,sha256=edxllyMiIFfajgCZ3dTRACXYGC34ZX1AYTezMSR5E_8,45
+classifyre_cli-0.7.4.dist-info/RECORD,,

src/sources/trino/source.py

@@ -32,13 +32,15 @@ _SKIPPED_CATALOGS = {"system"}
 _SKIPPED_SCHEMAS = {"information_schema", "system"}
 
 # Trino connector name -> the table format its tables are stored in.
+# A catalog's connector names the format of every table in it. Anything else,
+# Hive's plain tables included, is ``native`` (as the Hive source reports them).
+# The ``lakehouse`` connector mixes formats per table, so it is ``native`` here
+# until its tables' own ``type`` property is read.
 _CONNECTOR_FORMATS = {
     "iceberg": "iceberg",
     "delta_lake": "delta",
     "delta-lake": "delta",
     "hudi": "hudi",
-    "hive": "hive",
-    "lakehouse": "lakehouse",
 }
 
 

src/utils/lakehouse_storage.py

@@ -18,6 +18,7 @@ What each provider supports (see the lakehouse docs page for the tested matrix):
 
 from __future__ import annotations
 
+import os
 from pathlib import Path
 from urllib.parse import urlsplit
 
@@ -118,6 +119,11 @@ def azure_table_storage(
     object_store: dict[str, str] = {"azure_storage_account_name": account}
     if emulated:
         object_store |= {"azure_storage_use_emulator": "true", "allow_http": "true"}
+        # In emulator mode the Rust object_store readers (Lance, Hudi, and the
+        # Delta kernel inside DuckDB, which turns emulator mode on by itself for
+        # the Azurite account) take the endpoint only from this variable, and
+        # otherwise go to 127.0.0.1:10000 whatever account_url says.
+        os.environ["AZURITE_BLOB_STORAGE_URL"] = f"{parsed.scheme}://{parsed.netloc}"
     elif not blob_endpoint.endswith(".blob.core.windows.net"):
         object_store["azure_storage_endpoint"] = blob_endpoint  # OneLake, sovereign clouds