CSV Engine vs csvkit

CSV Engine is a web tool for cleaning and validating uploaded spreadsheets before export. It maps source columns to reusable templates, sanitizes values, detects duplicates, previews row-level failures, and can export only records that pass configured rules. Uploads use drag and drop, with automatic detection of delimiters, encodings, and headers. Column mapping supports aliases and fuzzy matching, and mappings can be saved per template. Listed cleanup includes whitespace trimming, casing normalization, email checks, phone parsing, and date fixes. Duplicate matching can use multiple fields and merge strategies, with conflicts reviewed before commit. Templates preserve schemas, field types, and validation rules for reuse. One credit processes up to 1,000 rows, including validation, deduplication, and export. The Free plan provides five credits per month and three active templates; the Starter plan is $15 a month billed annually. One-off credit packs are also available, beginning at $8, paid once, with credits that never expire. Pro includes email support with a 24-hour response; its developer API is listed as coming soon.

csvkit is a free suite of command-line tools for converting, inspecting and transforming CSV data on Linux, macOS and Windows. Its tools handle common file and table tasks: in2csv converts formats such as Excel, JSON and fixed-width files; csvcut selects, removes or reorders columns; and csvlook and csvstat display data and summary statistics. Other commands filter matching rows, sort records, join related data or stack subsets. For database workflows, csvsql can query CSV data with SQL and import it into PostgreSQL, while sql2csv exports query results. csvjson converts CSV to JSON or GeoJSON. Commands can read standard input and write standard output, making them suitable for repeatable pipelines. By default, csvkit infers format from the first 1024 bytes and infers cell types, with options to turn off either behavior. Installation instructions include pip and Homebrew and recommend a virtual environment. The tutorial assumes basic command-line familiarity, but not prior data-processing experience. The project notes that larger files or faster processing may exceed its limits and suggests SQL or other tools in that case.