parq_blockmodel.schema.utils

Pure utility functions for schema handling and validation.

Functions in this module are stateless and can be used independently of any ParquetBlockModel or SchemaService instance.

Functions

build_parquet_compression_kwargs	Build Parquet writer kwargs from a resolved compression policy.
build_schema_metadata	Build combined metadata dict for Parquet schema (geometry + schema YAML + compression).
df_eval_operations_from_schema	Extract df-eval operations from a pandera schema.
dump_schema_yaml	Serialize a pandera schema to YAML text.
extract_column_aliases_from_schema	Extract alias mappings from schema df-eval metadata.
extract_required_columns_from_schema	Extract the set of column names that are required (should be persisted).
extract_schema_compression_policy	Read compression policy from schema metadata if present.
load_embedded_compression	Extract compression policy metadata embedded in a Parquet file.
load_embedded_schema	Extract and load a schema embedded in Parquet metadata.
load_schema	Load and return a pandera DataFrameSchema.
resolve_active_compression_policy	Resolve a simple user-facing compression choice for active writes.
resolve_archive_compression_policy	Resolve archive compression using schema metadata or explicit policy.
schema_uses_strict_filter	Return True when schema root-level strict mode is set to "filter".
select_df_eval_operations	Recursively select df-eval operations and their dependencies.
validate_chunk	Validate and coerce a single DataFrame chunk against a schema.