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