| 1 |
//! **heck** is a case conversion library. |
| 2 |
//! |
| 3 |
//! This library exists to provide case conversion between common cases like |
| 4 |
//! CamelCase and snake_case. It is intended to be unicode aware, internally |
| 5 |
//! consistent, and reasonably well performing. |
| 6 |
//! |
| 7 |
//! ## Definition of a word boundary |
| 8 |
//! |
| 9 |
//! Word boundaries are defined as the "unicode words" defined in the |
| 10 |
//! `unicode_segmentation` library, as well as within those words in this |
| 11 |
//! manner: |
| 12 |
//! |
| 13 |
//! 1. All underscore characters are considered word boundaries. |
| 14 |
//! 2. If an uppercase character is followed by lowercase letters, a word |
| 15 |
//! boundary is considered to be just prior to that uppercase character. |
| 16 |
//! 3. If multiple uppercase characters are consecutive, they are considered to |
| 17 |
//! be within a single word, except that the last will be part of the next word |
| 18 |
//! if it is followed by lowercase characters (see rule 2). |
| 19 |
//! |
| 20 |
//! That is, "HelloWorld" is segmented `Hello|World` whereas "XMLHttpRequest" is |
| 21 |
//! segmented `XML|Http|Request`. |
| 22 |
//! |
| 23 |
//! Characters not within words (such as spaces, punctuations, and underscores) |
| 24 |
//! are not included in the output string except as they are a part of the case |
| 25 |
//! being converted to. Multiple adjacent word boundaries (such as a series of |
| 26 |
//! underscores) are folded into one. ("hello__world" in snake case is therefore |
| 27 |
//! "hello_world", not the exact same string). Leading or trailing word boundary |
| 28 |
//! indicators are dropped, except insofar as CamelCase capitalizes the first |
| 29 |
//! word. |
| 30 |
//! |
| 31 |
//! ### Cases contained in this library: |
| 32 |
//! |
| 33 |
//! 1. UpperCamelCase |
| 34 |
//! 2. lowerCamelCase |
| 35 |
//! 3. snake_case |
| 36 |
//! 4. kebab-case |
| 37 |
//! 5. SHOUTY_SNAKE_CASE |
| 38 |
//! 6. Title Case |
| 39 |
//! 7. SHOUTY-KEBAB-CASE |
| 40 |
//! 8. Train-Case |
| 41 |
#![deny(missing_docs)] |
| 42 |
#![forbid(unsafe_code)] |
| 43 |
|
| 44 |
mod kebab; |
| 45 |
mod lower_camel; |
| 46 |
mod shouty_kebab; |
| 47 |
mod shouty_snake; |
| 48 |
mod snake; |
| 49 |
mod title; |
| 50 |
mod train; |
| 51 |
mod upper_camel; |
| 52 |
|
| 53 |
pub use kebab::{AsKebabCase, ToKebabCase}; |
| 54 |
pub use lower_camel::{AsLowerCamelCase, ToLowerCamelCase}; |
| 55 |
pub use shouty_kebab::{AsShoutyKebabCase, ToShoutyKebabCase}; |
| 56 |
pub use shouty_snake::{ |
| 57 |
AsShoutySnakeCase, AsShoutySnakeCase as AsShoutySnekCase, ToShoutySnakeCase, ToShoutySnekCase, |
| 58 |
}; |
| 59 |
pub use snake::{AsSnakeCase, AsSnakeCase as AsSnekCase, ToSnakeCase, ToSnekCase}; |
| 60 |
pub use title::{AsTitleCase, ToTitleCase}; |
| 61 |
pub use train::{AsTrainCase, ToTrainCase}; |
| 62 |
pub use upper_camel::{ |
| 63 |
AsUpperCamelCase, AsUpperCamelCase as AsPascalCase, ToPascalCase, ToUpperCamelCase, |
| 64 |
}; |
| 65 |
|
| 66 |
use std::fmt; |
| 67 |
|
| 68 |
#[cfg(feature = "unicode")] |
| 69 |
fn get_iterator(s: &str) -> unicode_segmentation::UnicodeWords { |
| 70 |
use unicode_segmentation::UnicodeSegmentation; |
| 71 |
s.unicode_words() |
| 72 |
} |
| 73 |
#[cfg(not(feature = "unicode"))] |
| 74 |
fn get_iterator(s: &str) -> impl Iterator<Item = &str> { |
| 75 |
s.split(|letter: char| !letter.is_ascii_alphanumeric()) |
| 76 |
} |
| 77 |
|
| 78 |
fn transform<F, G>( |
| 79 |
s: &str, |
| 80 |
mut with_word: F, |
| 81 |
mut boundary: G, |
| 82 |
f: &mut fmt::Formatter, |
| 83 |
) -> fmt::Result |
| 84 |
where |
| 85 |
F: FnMut(&str, &mut fmt::Formatter) -> fmt::Result, |
| 86 |
G: FnMut(&mut fmt::Formatter) -> fmt::Result, |
| 87 |
{ |
| 88 |
/// Tracks the current 'mode' of the transformation algorithm as it scans |
| 89 |
/// the input string. |
| 90 |
/// |
| 91 |
/// The mode is a tri-state which tracks the case of the last cased |
| 92 |
/// character of the current word. If there is no cased character |
| 93 |
/// (either lowercase or uppercase) since the previous word boundary, |
| 94 |
/// than the mode is `Boundary`. If the last cased character is lowercase, |
| 95 |
/// then the mode is `Lowercase`. Othertherwise, the mode is |
| 96 |
/// `Uppercase`. |
| 97 |
#[derive(Clone, Copy, PartialEq)] |
| 98 |
enum WordMode { |
| 99 |
/// There have been no lowercase or uppercase characters in the current |
| 100 |
/// word. |
| 101 |
Boundary, |
| 102 |
/// The previous cased character in the current word is lowercase. |
| 103 |
Lowercase, |
| 104 |
/// The previous cased character in the current word is uppercase. |
| 105 |
Uppercase, |
| 106 |
} |
| 107 |
|
| 108 |
let mut first_word = true; |
| 109 |
|
| 110 |
for word in get_iterator(s) { |
| 111 |
let mut char_indices = word.char_indices().peekable(); |
| 112 |
let mut init = 0; |
| 113 |
let mut mode = WordMode::Boundary; |
| 114 |
|
| 115 |
while let Some((i, c)) = char_indices.next() { |
| 116 |
// Skip underscore characters |
| 117 |
if c == '_' { |
| 118 |
if init == i { |
| 119 |
init += 1; |
| 120 |
} |
| 121 |
continue; |
| 122 |
} |
| 123 |
|
| 124 |
if let Some(&(next_i, next)) = char_indices.peek() { |
| 125 |
// The mode including the current character, assuming the |
| 126 |
// current character does not result in a word boundary. |
| 127 |
let next_mode = if c.is_lowercase() { |
| 128 |
WordMode::Lowercase |
| 129 |
} else if c.is_uppercase() { |
| 130 |
WordMode::Uppercase |
| 131 |
} else { |
| 132 |
mode |
| 133 |
}; |
| 134 |
|
| 135 |
// Word boundary after if next is underscore or current is |
| 136 |
// not uppercase and next is uppercase |
| 137 |
if next == '_' || (next_mode == WordMode::Lowercase && next.is_uppercase()) { |
| 138 |
if !first_word { |
| 139 |
boundary(f)?; |
| 140 |
} |
| 141 |
with_word(&word[init..next_i], f)?; |
| 142 |
first_word = false; |
| 143 |
init = next_i; |
| 144 |
mode = WordMode::Boundary; |
| 145 |
|
| 146 |
// Otherwise if current and previous are uppercase and next |
| 147 |
// is lowercase, word boundary before |
| 148 |
} else if mode == WordMode::Uppercase && c.is_uppercase() && next.is_lowercase() { |
| 149 |
if !first_word { |
| 150 |
boundary(f)?; |
| 151 |
} else { |
| 152 |
first_word = false; |
| 153 |
} |
| 154 |
with_word(&word[init..i], f)?; |
| 155 |
init = i; |
| 156 |
mode = WordMode::Boundary; |
| 157 |
|
| 158 |
// Otherwise no word boundary, just update the mode |
| 159 |
} else { |
| 160 |
mode = next_mode; |
| 161 |
} |
| 162 |
} else { |
| 163 |
// Collect trailing characters as a word |
| 164 |
if !first_word { |
| 165 |
boundary(f)?; |
| 166 |
} else { |
| 167 |
first_word = false; |
| 168 |
} |
| 169 |
with_word(&word[init..], f)?; |
| 170 |
break; |
| 171 |
} |
| 172 |
} |
| 173 |
} |
| 174 |
|
| 175 |
Ok(()) |
| 176 |
} |
| 177 |
|
| 178 |
fn lowercase(s: &str, f: &mut fmt::Formatter) -> fmt::Result { |
| 179 |
let mut chars = s.chars().peekable(); |
| 180 |
while let Some(c) = chars.next() { |
| 181 |
if c == 'Σ' && chars.peek().is_none() { |
| 182 |
write!(f, "ς")?; |
| 183 |
} else { |
| 184 |
write!(f, "{}", c.to_lowercase())?; |
| 185 |
} |
| 186 |
} |
| 187 |
|
| 188 |
Ok(()) |
| 189 |
} |
| 190 |
|
| 191 |
fn uppercase(s: &str, f: &mut fmt::Formatter) -> fmt::Result { |
| 192 |
for c in s.chars() { |
| 193 |
write!(f, "{}", c.to_uppercase())?; |
| 194 |
} |
| 195 |
|
| 196 |
Ok(()) |
| 197 |
} |
| 198 |
|
| 199 |
fn capitalize(s: &str, f: &mut fmt::Formatter) -> fmt::Result { |
| 200 |
let mut char_indices = s.char_indices(); |
| 201 |
if let Some((_, c)) = char_indices.next() { |
| 202 |
write!(f, "{}", c.to_uppercase())?; |
| 203 |
if let Some((i, _)) = char_indices.next() { |
| 204 |
lowercase(&s[i..], f)?; |
| 205 |
} |
| 206 |
} |
| 207 |
|
| 208 |
Ok(()) |
| 209 |
} |