loc

`loc`

The loc command scans a directory tree, detects built-in languages, and reports:

code lines
comment lines
empty lines
size in bytes

The default output is a terminal table with per-language rows, totals, and elapsed time.

Use --path for local analysis or --url to clone and analyze a remote repository directly.

Output formats

--format selects the rendering, and --output writes it to a file instead of standard output:

sephera loc --path . --format json --output reports/loc.json
sephera loc --path . --format csv
sephera loc --path . --format markdown

format	for
`table`	a terminal; the default
`json`	a dashboard or a script, with elapsed time included
`csv`	a spreadsheet
`markdown`	pasting into a pull request

CSV quotes every language label unconditionally. No built-in language name contains a comma today, but a label chosen outside the tool that silently produces a wrong column count the first time it changes is worse than one that is always right.

Basic usage

sephera loc --path .

Analyze a remote repository directly:

sephera loc --url https://github.com/Reim-developer/Sephera

Analyze a repository tree URL:

sephera loc --url https://github.com/Reim-developer/Sephera/tree/master/crates

Sample Output

Scanning: crates

╭──────────┬──────┬─────────┬───────┬─────────╮
│ Language ┆ Code ┆ Comment ┆ Empty ┆    Size │
│          ┆      ┆         ┆       ┆ (bytes) │
╞══════════╪══════╪═════════╪═══════╪═════════╡
│ Rust     ┆ 9724 ┆     652 ┆  1357 ┆  358209 │
│ TOML     ┆  113 ┆       0 ┆    11 ┆    3493 │
│ Markdown ┆   69 ┆       0 ┆    35 ┆    2980 │
│ Totals   ┆ 9906 ┆     652 ┆  1403 ┆  364682 │
╰──────────┴──────┴─────────┴───────┴─────────╯
Files scanned: 88
Languages detected: 3
Elapsed: 3.909 ms (0.003909 s)

Demo

Defaults from `.sephera.toml`

loc reads [project] and [loc] from .sephera.toml, so the format, the output path and the ignore list do not have to be typed every time:

[project]
ignore = ["vendor"]

[loc]
format = "markdown"
output = "reports/loc.md"

sephera loc --path .            # writes reports/loc.md in markdown
sephera loc --path . --format json    # a typed flag still wins

An unknown key in that file is an error naming the correction, not a setting that quietly does nothing. See .sephera.toml for profiles and aliases.

Ignore patterns

Repeat --ignore to combine multiple patterns:

sephera loc --path . --ignore target --ignore "*.snap"

Patterns containing *, ?, or [ are treated as globs and matched against both the file name and the path relative to the base, so dist/** and **/node_modules/** each exclude a whole tree. Other patterns are compiled as regexes and matched against the relative path, where an unanchored pattern such as target matches anywhere in it.

Every analysis also reads the repository’s own .gitignore and .sepheraignore, applying each one to the directory that holds it. Pass --no-gitignore to analyse the tree as it is on disk instead. Explicit --ignore patterns outrank the repository’s rules, so a pattern you typed is never undone by a ! line in .gitignore.

Progress

loc draws a progress bar while it works, on standard error so that standard output carries the report and nothing else:

Reading files ####>--------------------------- 3,204/8,901 (36%)

The count is real, not an animation. It appears once the walk has finished and the file count is known, which is why the bar is a spinner until then — a percentage of a total nobody has counted yet would be a decoration.

--progress takes three values:

value	behaviour
`auto`	the default: draw only when standard error is a terminal
`always`	draw even without a terminal, for a CI log or a screen recording
`never`	never draw

Nothing needs --progress never in a script. Piping or redirecting standard error already makes auto silent, and every machine-readable format is written to standard output, so a captured report is unaffected either way.

Remote refs

For repo URLs, use --ref to analyze a specific branch, tag, or commit:

sephera loc --url https://github.com/Reim-developer/Sephera --ref v0.5.0

--ref applies to repo URLs only. Tree URLs already encode the ref in the URL itself.

Notes on correctness

Sephera’s scanner is byte-oriented and comment-token aware. It is designed to be fast, stable, and portable across newline styles, rather than to fully parse each language grammar.

In practice, that means:

support for LF, CRLF, and classic CR
support for the built-in comment styles declared in the language registry
stable behavior across all supported language lookups

For the current language metadata source of truth, see config/languages.yml.

loc

loc