Skip to content

Feature: tokens count #1268

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Arichy wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Feature: tokens count #1268

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
84eef56
feat: add tokens count
Arichy Jun 14, 2025
7808da3
feat: add tokens column to CLI output and stats display
Arichy Jun 14, 2025
e7bc4e1
refactor: remove debug print statements from main and fs modules
Arichy Jun 14, 2025
51dac37
feat: implement token counting from byte slices and add tokens module
Arichy Jun 14, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
46 changes: 46 additions & 0 deletions Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

1 change: 1 addition & 0 deletions Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,6 +60,7 @@ regex = "1.10.6"
serde_json = "1.0.125"
etcetera = "0.8.0"
table_formatter = "0.6.1"
tiktoken-rs = "0.7.0"

[dependencies.env_logger]
features = []
Expand Down
7 changes: 4 additions & 3 deletions src/cli.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ use crate::{
cli_utils::{crate_version, parse_or_exit, NumberFormatStyle},
consts::{
BLANKS_COLUMN_WIDTH, CODE_COLUMN_WIDTH, COMMENTS_COLUMN_WIDTH, LANGUAGE_COLUMN_WIDTH,
LINES_COLUMN_WIDTH, PATH_COLUMN_WIDTH,
LINES_COLUMN_WIDTH, PATH_COLUMN_WIDTH, TOKENS_COLUMN_WIDTH,
},
input::Format,
};
Expand Down Expand Up @@ -446,13 +446,14 @@ impl Cli {
}),
Some(Streaming::Simple) => Some(|l: LanguageType, e| {
println!(
"{:>LANGUAGE_COLUMN_WIDTH$} {:<PATH_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
"{:>LANGUAGE_COLUMN_WIDTH$} {:<PATH_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
l.name(),
e.name.to_string_lossy().to_string(),
e.stats.lines(),
e.stats.code,
e.stats.comments,
e.stats.blanks
e.stats.blanks,
e.stats.tokens,
);
}),
_ => None,
Expand Down
26 changes: 16 additions & 10 deletions src/cli_utils.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,12 +15,12 @@ use tokei::{find_char_boundary, CodeStats, Language, LanguageType, Report};

use crate::consts::{
BLANKS_COLUMN_WIDTH, CODE_COLUMN_WIDTH, COMMENTS_COLUMN_WIDTH, FILES_COLUMN_WIDTH,
LINES_COLUMN_WIDTH,
LINES_COLUMN_WIDTH, TOKENS_COLUMN_WIDTH,
};

const NO_LANG_HEADER_ROW_LEN: usize = 69;
const NO_LANG_ROW_LEN: usize = 63;
const NO_LANG_ROW_LEN_NO_SPACES: usize = 56;
const NO_LANG_HEADER_ROW_LEN: usize = 69 + 13;
const NO_LANG_ROW_LEN: usize = 63 + 13;
const NO_LANG_ROW_LEN_NO_SPACES: usize = 56 + 13;
const IDENT_INACCURATE: &str = "(!)";

pub fn crate_version() -> String {
Expand Down Expand Up @@ -155,13 +155,14 @@ impl<W: Write> Printer<W> {
let files_column_width: usize = FILES_COLUMN_WIDTH + 6;
writeln!(
self.writer,
" {:<6$} {:>files_column_width$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
" {:<7$} {:>files_column_width$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
"Language".bold().blue(),
"Files".bold().blue(),
"Lines".bold().blue(),
"Code".bold().blue(),
"Comments".bold().blue(),
"Blanks".bold().blue(),
"Tokens".bold().blue(),
self.columns - NO_LANG_HEADER_ROW_LEN
)?;
self.print_row()
Expand All @@ -183,7 +184,7 @@ impl<W: Write> Printer<W> {
write!(self.writer, " ")?;
writeln!(
self.writer,
"{:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
"{:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
language
.reports
.len()
Expand All @@ -192,6 +193,7 @@ impl<W: Write> Printer<W> {
language.code.to_formatted_string(&self.number_format),
language.comments.to_formatted_string(&self.number_format),
language.blanks.to_formatted_string(&self.number_format),
language.tokens.to_formatted_string(&self.number_format),
)
}

Expand All @@ -203,7 +205,7 @@ impl<W: Write> Printer<W> {
write!(self.writer, " ")?;
writeln!(
self.writer,
"{:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
"{:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
language
.children
.values()
Expand All @@ -227,6 +229,7 @@ impl<W: Write> Printer<W> {
.blanks
.to_formatted_string(&self.number_format)
.blue(),
language.tokens.to_formatted_string(&self.number_format).blue(),
)
}

Expand Down Expand Up @@ -284,12 +287,13 @@ impl<W: Write> Printer<W> {
} else {
writeln!(
self.writer,
" {:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
" {:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
stats.len().to_formatted_string(&self.number_format),
(code + comments + blanks).to_formatted_string(&self.number_format),
code.to_formatted_string(&self.number_format),
comments.to_formatted_string(&self.number_format),
blanks.to_formatted_string(&self.number_format),
stats.iter().map(|s| s.tokens).sum::<usize>().to_formatted_string(&self.number_format),
)
}
}
Expand Down Expand Up @@ -414,12 +418,13 @@ impl<W: Write> Printer<W> {

writeln!(
self.writer,
" {:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
" {:>FILES_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
" ",
stats.lines().to_formatted_string(&self.number_format),
stats.code.to_formatted_string(&self.number_format),
stats.comments.to_formatted_string(&self.number_format),
stats.blanks.to_formatted_string(&self.number_format),
stats.tokens.to_formatted_string(&self.number_format),
)
}

Expand Down Expand Up @@ -467,7 +472,7 @@ impl<W: Write> Printer<W> {
let lines_column_width: usize = FILES_COLUMN_WIDTH + 6;
writeln!(
self.writer,
" {: <max$} {:>lines_column_width$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
" {: <max$} {:>lines_column_width$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
name,
report
.stats
Expand All @@ -479,6 +484,7 @@ impl<W: Write> Printer<W> {
.comments
.to_formatted_string(&self.number_format),
report.stats.blanks.to_formatted_string(&self.number_format),
report.stats.tokens.to_formatted_string(&self.number_format),
max = max_len
)
}
Expand Down
5 changes: 4 additions & 1 deletion src/consts.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
// Set of common pub consts.

/// Fallback row length
pub const FALLBACK_ROW_LEN: usize = 81;
pub const FALLBACK_ROW_LEN: usize = 81 + 13;

// Column widths used for console printing.

Expand All @@ -25,3 +25,6 @@ pub const COMMENTS_COLUMN_WIDTH: usize = 12;

/// Blanks column width
pub const BLANKS_COLUMN_WIDTH: usize = 12;

/// Tokens column width
pub const TOKENS_COLUMN_WIDTH: usize = 12;
15 changes: 10 additions & 5 deletions src/language/language_type.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,24 +108,26 @@ impl LanguageType {
// first character in the column, so removing starting whitespace
// could cause a miscount.
let line = if is_fortran { line } else { line.trim() };
let tokens = crate::tokens::count_tokens_from_bytes(line);
if line.trim().is_empty() {
(1, 0, 0)
(1, 0, 0, tokens)
} else if is_literate
|| comments.iter().any(|c| line.starts_with(c.as_bytes()))
{
(0, 0, 1)
(0, 0, 1, tokens)
} else {
(0, 1, 0)
(0, 1, 0, tokens)
}
})
.reduce(|| (0, 0, 0), |a, b| (a.0 + b.0, a.1 + b.1, a.2 + b.2))
.reduce(|| (0, 0, 0, 0), |a, b| (a.0 + b.0, a.1 + b.1, a.2 + b.2, a.3 + b.3))
};

let (mut stats, (blanks, code, comments)) = rayon::join(parse_lines, simple_parse);
let (mut stats, (blanks, code, comments, tokens)) = rayon::join(parse_lines, simple_parse);

stats.blanks += blanks;
stats.code += code;
stats.comments += comments;
stats.tokens += tokens;
stats
} else {
self.parse_lines(config, text, CodeStats::new(), syntax)
Expand Down Expand Up @@ -211,6 +213,9 @@ impl LanguageType {
}
}

let tokens = crate::tokens::count_tokens_from_bytes(lines);
stats.tokens += tokens;

stats
}

Expand Down
1 change: 1 addition & 0 deletions src/language/languages.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,6 +107,7 @@ impl Languages {
total.comments += language.comments;
total.blanks += language.blanks;
total.code += language.code;
total.tokens += language.tokens;
total.inaccurate |= language.inaccurate;
total.children.insert(*ty, language.reports.clone());
}
Expand Down
6 changes: 6 additions & 0 deletions src/language/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,8 @@ pub struct Language {
pub code: usize,
/// The total number of comments(both single, and multi-line)
pub comments: usize,
/// The total number of tokens.
pub tokens: usize,
/// A collection of statistics of individual files.
pub reports: Vec<Report>,
/// A map of any languages found in the reports.
Expand Down Expand Up @@ -77,6 +79,7 @@ impl Language {
summary.comments += stats.comments;
summary.code += stats.code;
summary.blanks += stats.blanks;
summary.tokens += stats.tokens;
}
}

Expand Down Expand Up @@ -104,16 +107,19 @@ impl Language {
let mut blanks = 0;
let mut code = 0;
let mut comments = 0;
let mut tokens = 0;

for report in &self.reports {
blanks += report.stats.blanks;
code += report.stats.code;
comments += report.stats.comments;
tokens += report.stats.tokens;
}

self.blanks = blanks;
self.code = code;
self.comments = comments;
self.tokens = tokens;
}

/// Checks if the language is empty. Empty meaning it doesn't have any
Expand Down
1 change: 1 addition & 0 deletions src/lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,6 +54,7 @@ mod consts;
mod language;
mod sort;
mod stats;
mod tokens;

pub use self::{
config::Config,
Expand Down
11 changes: 6 additions & 5 deletions src/main.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ use crate::{
cli_utils::Printer,
consts::{
BLANKS_COLUMN_WIDTH, CODE_COLUMN_WIDTH, COMMENTS_COLUMN_WIDTH, FALLBACK_ROW_LEN,
LANGUAGE_COLUMN_WIDTH, LINES_COLUMN_WIDTH, PATH_COLUMN_WIDTH,
LANGUAGE_COLUMN_WIDTH, LINES_COLUMN_WIDTH, PATH_COLUMN_WIDTH, TOKENS_COLUMN_WIDTH,
},
input::add_input,
};
Expand Down Expand Up @@ -61,17 +61,18 @@ fn main() -> Result<(), Box<dyn Error>> {

if cli.streaming == Some(crate::cli::Streaming::Simple) {
println!(
"#{:^LANGUAGE_COLUMN_WIDTH$} {:^PATH_COLUMN_WIDTH$} {:^LINES_COLUMN_WIDTH$} {:^CODE_COLUMN_WIDTH$} {:^COMMENTS_COLUMN_WIDTH$} {:^BLANKS_COLUMN_WIDTH$}",
"language", "path", "lines", "code", "comments", "blanks"
"#{:^LANGUAGE_COLUMN_WIDTH$} {:^PATH_COLUMN_WIDTH$} {:^LINES_COLUMN_WIDTH$} {:^CODE_COLUMN_WIDTH$} {:^COMMENTS_COLUMN_WIDTH$} {:^BLANKS_COLUMN_WIDTH$} {:^TOKENS_COLUMN_WIDTH$}",
"language", "path", "lines", "code", "comments", "blanks", "tokens"
);
println!(
"{:>LANGUAGE_COLUMN_WIDTH$} {:<PATH_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$}",
"{:>LANGUAGE_COLUMN_WIDTH$} {:<PATH_COLUMN_WIDTH$} {:>LINES_COLUMN_WIDTH$} {:>CODE_COLUMN_WIDTH$} {:>COMMENTS_COLUMN_WIDTH$} {:>BLANKS_COLUMN_WIDTH$} {:>TOKENS_COLUMN_WIDTH$}",
(0..10).map(|_| "#").collect::<String>(),
(0..80).map(|_| "#").collect::<String>(),
(0..12).map(|_| "#").collect::<String>(),
(0..12).map(|_| "#").collect::<String>(),
(0..12).map(|_| "#").collect::<String>(),
(0..12).map(|_| "#").collect::<String>()
(0..12).map(|_| "#").collect::<String>(),
(0..12).map(|_| "#").collect::<String>(),
);
}

Expand Down
Loading
Loading