Optional row-level diagnostics produced by read_json_with_row_diagnostics.
More...
#include <cudf/io/json.hpp>
Classes | |
| struct | schema_mismatch_rows |
| Row indices for one top-level column that had schema mismatches. More... | |
Public Attributes | |
| std::vector< std::string > | top_level_columns_with_schema_mismatch |
Top-level column names whose value tree contained at least one JSON node whose category (NC_STRUCT / NC_LIST / NC_VAL) did not match the requested schema type. Empty when the reader is invoked without a user-supplied schema or when no mismatches were observed. | |
| std::vector< schema_mismatch_rows > | top_level_columns_with_schema_mismatch_rows |
| Row indices grouped by top-level output column for schema mismatches. More... | |
Optional row-level diagnostics produced by read_json_with_row_diagnostics.
Reserved for reader-specific diagnostics whose collection has additional cost and whose payload is not part of the ABI of json_reader_diagnostics.
| std::vector<schema_mismatch_rows> cudf::io::json_reader_row_diagnostics::top_level_columns_with_schema_mismatch_rows |
Row indices grouped by top-level output column for schema mismatches.
Each entry names a top-level output column and lists the rows in that column where at least one descendant JSON node's category did not match the requested schema type. The indices are sorted, unique, and relative to the final returned table, not to an internal reader batch. Entries are ordered consistently with top_level_columns_with_schema_mismatch.
1.9.1