Docfd

Online Demo

TUI multiline fuzzy document finder

Think interactive grep for text files, PDFs, DOCXs, etc, but word/token based instead of regex and line based, so you can search across lines easily.

Docfd aims to provide good UX via integration with common text editors and PDF viewers, so you can jump directly to a search result with a single key press.

Interactive use

Non-interactive use

Features

Why Docfd might be for you

You want a standalone, offline TUI search tool that allows you to immediately start searching without any complicated setup.

Docfd only starts processing the current directory or specified directories/files upon start. Hashing is used to pick out files that have not been indexed yet.

There is no need to wait for a background indexer to refresh before you get up-to-date results.

You don't want to move everything into a central storage, and want to just keep your current folder structure

There are no strings attached with using Docfd. Docfd does not require you to import your files into any special storage system, so you can continue mix and match tools to best handle your files.

You want to script or record your search

Docfd comes with a simple scripting language, which is already used to capture your actions in the TUI.

Finally found what you need after many steps? Save the session as a script with Ctrl+S! Then open it next time with Ctrl+O.

Why Docfd might not be for you

Docfd is not all-encompassing

Docfd does not try to be a full blown document management system such as Paperless-ngx. While there may be significant overlaps in terms of the search functionality, Docfd will fall short for almost any other kind of features, such as storage management, tagging, web interface, OCR, email ingestion.

Docfd is not a "proper" search engine

Docfd is a search engine in the sense that it uses the same fundamental principles, i.e. inverted indices, but it lacks features that you would expect from a "proper" search engine solution, e.g. Apache Lucene, Tantivy, Lnx.

Here are some of the fundamental features which I think are crucial to a proper search engine, but Docfd lacks:

Some of these shortcomings are fundamental to the goals of Docfd. For instance, Docfd is primarily a standalone desktop TUI tool with quick startup and should not impact other desktop applications. As such, some performance related engineering choices typical for a proper search engine are difficult to accommodate as they require longer startup and significantly more memory usage.

Other shortcomings are due to limited time and limited return on efforts - if one is to push Docfd so much to reach the feature parity and performance of a proper search engine, then one might as well just use an existing search engine to begin with.

If your notes are consistently very short, and you only want to do simple searches, then there are better options

If you follow note taking methodologies such as Zettelkasten, where each note consists of very few lines, then using a combination of grep and file preview tool can yield a much faster search experience by skipping out on indexing and consideration of word proximity.

Docfd does not "stream" its search results

One user feedback received was that searching felt slow when Docfd is still conducting the search as UI is not updated result by result. By comparison, fzf felt faster as results start to immediately pop into the screen.

It is fundamentally more difficult to implement this streaming behavior nicely in Docfd, as Docfd operates with snapshots in mind (e.g. allowing you to undo/redo commands), while fzf does not. More specifically, it is much easier to wait for all search results to be ready, and finalize as a snapshot before presenting onto Docfd UI.

So while possible to implement in Docfd, it is unclear if the effort is worthwhile with the additional system complexity in mind.

Installation

Statically linked binaries for Linux and macOS are available via GitHub releases.

Docfd is also packaged on the following platforms for Linux:

The only way to use Docfd on Windows right now is via WSL.

Notes for packagers: Outside of the OCaml toolchain for building (if you are packaging from source), Docfd also requires the following external tools at run time for full functionality:

Basic usage

The typical usage of Docfd is to either cd into the directory of interest and launch docfd directly, or specify the paths as arguments:

docfd [PATH]...

The list of paths can contain directories. Each directory in the list is scanned recursively for files with the following extensions by default:

You can change the file extensions to use via --exts and --single-line-exts, or add onto the list of extensions via --add-exts and --single-line-add-exts.

If the list PATHs is empty, then Docfd defaults to scanning the current directory . unless any of the following is used: --paths-from, --glob, --single-line-glob.

Documentation

See GitHub Wiki for more examples/cookbook, and technical details.

Changelog

CHANGELOG

Limitations

Acknowledgement