awkreader

File Reading with Pre-Filtering, Pattern Searching, and Distributed Files

Provides high-performance tools for out-of-core text processing and data ingestion by leveraging system 'AWK' utilities. Allows users to count records, filter rows, and compute streaming aggregations—such as group-by means, streaming medians, standard deviations, and correlations—directly on disk prior to reading data into R. By delegating line-by-line filtering and summarization to system-level 'AWK' commands and streaming results back through data.table::fread(), the package significantly reduces memory footprint and execution times when working with large individual files or multi-file directory structures.

Versions across snapshots

Dependencies (latest)