Tagalyzer 0.3 is released!

This is exciting for me because it’s the first version that I feel feels like a CLI tool. Release 0.2.0 included the functionality to parse text files and return word counts, but it’s this release that brought the interface up to par with other tools. There’s two major changes that made that true.

Passing Files like a CLI

Version 0.2 could handle multiple files, provided that you passed the correct and literal filepaths. For example, these all would have worked:

tagalyzer file.md
tagalyzer $(ls *.md)
tagalyzer $(ls **/*.md)
tagalyzer $(find . -name "*.md")

However, the program relied on these external tools to handle globs and recursion. Now, with version 0.3.0, these are possible:

tagalyzer posts/*
tagalyzer posts/*.md
tagalyzer -r posts
tagalyzer -r posts other_directory/file.md

RegEx: Everyone’s least favorite necessity

Regular expressions are powerful, even if they look a bit like an esoteric programming language with coarse language in its name. In fact, I’d argue it’s leaps and bounds better than the previous method, which filtered individual characters. This change has to be one of my favorites: the default output (i.e. when running tagalyzer LICENSE-MIT) is now free of articles and other small, common, words that carry little semantic information. (Was anyone really going to use #the as a tag? If you really want to, I guess you can write your own RegEx pattern that won’t match words like that.)

$ tagalyzer LICENSE-MIT # Case insensitive by default
Sorted wordcount for LICENSE-MIT
software       : 10
without        : 4
including      : 4
copyright      : 4
any            : 4
for            : 3
--- [snip] ---

That list could still be refined, but it quickly devolves into words used very infrequently for such a short file anyway.

So why isn’t it already refined?

I’m still trying to make the tool before I make it better. 0.3.0 is the first release that I feel is even a tool, and there’s some big changes coming before I consider calling the tool 1.0.0 (“feature complete”).

But if that matters to you, or you want to suggest another change, or you just want to help me build the tool – go say something! I host the project on GitLab, where you can open an issue (bugs and suggestions are both welcome), claim an issue and work on it, or just vote on the couple of polls I have open! And since the project is still in major version 0, now’s the time to make your voice heard; I’m way more open to fiddling with default values while the API is marked as unstable.

Get involved!