Summary
Adds column_mapping support for tabular comparisons.
Key message
Reconlify can now compare semantically equivalent datasets even when source and target use different column names. This is especially useful for migration validation, cross-system reconciliation, and exported reports from different tools. Column mapping uses logical/source-side column names throughout the config, while target-side physical columns are resolved via column_mapping. This behavior is now also reflected in the report metadata under details.column_mapping.
Example
type: tabular
source: trades_erp.csv
target: trades_ledger.csv
keys:
- trade_id
column_mapping:
trade_id: id
amount: total_amountNotes
-
all other config sections still use logical/source-side column names
-
no target-side transforms
-
no many-to-one mapping
-
report includes effective details.column_mapping when configured