Install ffe with Homebrew, apt, MacPorts

Project links

• Homepage
• Package manager page

Commands and aliases

• ffe

Project history

The official project page lists Timo Savinen as the author and records the first public version, 0.1.0, on March 1, 2006. The tool was developed in a GNU/Linux environment, distributed under GPL-2.0-or-later, built with GNU autotools, and intended to build on many Unix-like systems.

The release history shows steady, practical evolution rather than a platform pivot: early 0.1 releases added filtering, expression, replacement, and Windows executable support; 0.2 releases added regex record selection, binary fixed-length parsing, raw output, field-level output selection, and loose mode; 0.3 releases added structured output helpers, preprocessing, performance improvements, dynamic record length, anonymization, and printf-style formatting.

Adoption history

ffe stayed in the small-tools lane, but the official site points to SourceForge source and Windows downloads, Debian packages, and a FreeBSD port. That is the adoption pattern of many durable Unix data tools: modest upstream visibility, distro packaging, and use by people who need a specific kind of batch conversion or validation.

Its niche is especially clear from the homepage's use cases: extracting fields or records, converting CSV to fixed length, verifying flat-file structure, testing flat-file development, displaying records in human-readable form, modifying structures, and mapping flat-file fields to other formats.

How it is used

ffe requires a configuration file that describes input structures and output formats. The manual says the default configuration file is ~/.fferc on Unix and ffe.rc on Windows, with -c available to choose another file. Because it reads standard input and writes standard output by default, it fits into shell pipelines.

The practical workflow is to define record types, fixed or separated fields, expressions, and output directives, then run ffe against flat files to produce fixed-length, separated, tokenized, XML, SQL, raw, or other configured output. That makes it useful when the format is structured enough for a schema but too local or old-fashioned for a mainstream parser.

Why package nerds care

ffe matters to package nerds as a representative of the quiet SourceForge-era Unix utilities that solve one unglamorous data problem well. Its value is not broad ecosystem gravity; it is that a packaged, documented, autotools-based CLI can keep old flat-file workflows reproducible across systems.

It is also a reminder that data tooling is not only JSON, CSV, and databases. Fixed-length records, mixed record types, binary fields, and configurable output formats still appear in institutional data exchanges, and ffe packages that logic as a small command instead of a one-off script.

Timeline

• 2006: First version 0.1.0 released.
• 2007: 0.2 series begins, adding regular-expression record selection and other parsing features.
• 2008: 0.2.7 adds fixed-length binary parsing, raw output, and an info option.
• 2010: 0.3 series adds command-line expression fixes, case comparison, record levels, and structured output helpers.
• 2015: 0.3.6 adds filter-program formatting and dynamic record length support.
• 2017: 0.3.7 adds simple data anonymization.
• 2018: 0.3.8 adds printf-style output-field formatting; 0.3.9 fixes raw-data printing with variable-length records.

Related projects

• ffe sits near awk, cut, sed, csv conversion tools, fixed-width parsers, XML or SQL export scripts, and older flat-file validation utilities, but uses its own configuration language to describe record structure and output.

Sources

• https://ff-extractor.sourceforge.net/
• https://ff-extractor.sourceforge.net/ffe.html
• https://ff-extractor.sourceforge.net/ffe.1.html