1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
//! This crate provides methods for finding the overlap between two string slices. //! //! An overlap is here defined as the largest substring contained both at the end of one string //! slice and the beginning of another string slice. //! //! The implementation is provided through the [`Overlap`] trait, which is implemented on [`str`]. //! This allows the user to simply pull the trait into scope and use its methods: //! //! ``` //! use str_overlap::Overlap; //! //! assert_eq!("bcd".overlap_start("abc"), "bc"); //! assert_eq!("abc".overlap_end("bcd"), "bc"); //! ``` //! //! The trait provides two methods: [`overlap_start`] and [`overlap_end`], which find the overlap //! at the beginning and end of the first value respectively. The reason for these two methods is //! to allow the user to specify ownership of the resulting subvalue, regardless of its overlap //! position. //! //! [`overlap_end`]: Overlap::overlap_end //! [`overlap_start`]: Overlap::overlap_start // Since the MSRV is 1.0.0, allowing usage of deprecated items is ok, as the replacements are likely // not available in early versions. #![allow(deprecated)] #![cfg_attr(rustc_1_6, no_std)] #[cfg(not(rustc_1_6))] extern crate std as core; /// Shared logic for finding the index at which two strings overlap. /// /// The `left` and `right` parameters are, conceptually, defined as follows: /// - `left` is the parameter whose suffix will be overlapping /// - `right` is the parameter whose prefix will be overlapping /// /// If no overlap exists, the returned index will be the length of `left`. This allows the result to /// be used to create an empty slice. #[inline] #[must_use] fn string_overlap_index(left: &str, right: &str) -> usize { left.char_indices() .map(|(index, _)| index) .find(|index| { let slice_len = left.len() - index; slice_len <= right.len() && unsafe { // SAFETY: `index` is obtained from `left`'s `CharIndices`, so it will always be // within the bounds of `left`. Additionally, `index` will also always be on // UTF-8 character bounds of `left`. left.slice_unchecked(*index, left.len()) // SAFETY: Since `slice_len - index` is less than or equal to `right.len()`, // `slice_len` will always be within the bounds of `right`. Additionally, since // the string slice is simply cast to bytes when checking equality, we don't // need to worry about whether the slice occurs on a valid UTF-8 character // bound. == right.slice_unchecked(0, slice_len) } }) .unwrap_or_else(|| left.len()) } /// Provides methods for finding overlaps between values. /// /// This trait provides methods for finding overlaps at both the start and end of `self`. This /// allows for returning overlapping values that are owned by `self`, regardless of which side of /// `self` the overlap is occurring. /// /// This trait is made available by pulling it into scope: /// /// ``` /// use str_overlap::Overlap; /// ``` /// /// `Overlap` is implemented on [`str`], which means its methods are usable by `str` and any types /// which implement [`Deref<Target = str>`], such as [`String`]. /// /// [`Deref<Target = str>`]: core::ops::Deref /// [`String`]: https://doc.rust-lang.org/std/string/struct.String.html pub trait Overlap { /// Returns the overlap found at the start of `self` and the end of `other`. /// /// # Example /// This method can be used through its implementation on [`str`], like so: /// /// ``` /// use str_overlap::Overlap; /// /// assert_eq!("bcd".overlap_start("abc"), "bc"); /// ``` fn overlap_start(&self, other: &Self) -> &Self; /// Returns the overlap found at the end of `self` and the start of `other`. /// /// # Example /// This method can be used through its implementation on `str`, like so: /// /// ``` /// use str_overlap::Overlap; /// /// assert_eq!("abc".overlap_end("bcd"), "bc"); /// ``` fn overlap_end(&self, other: &Self) -> &Self; } /// Overlap methods for string slices. /// /// This allows for the returned string slice to be a subset of either string slice from which an /// overlap is obtained. impl Overlap for str { /// Returns the substring which is both the prefix to `self` and the suffix to `other`. /// /// The returned string slice is a reference to the substring contained in `self`. /// /// # Example /// ``` /// use str_overlap::Overlap; /// /// assert_eq!("bcd".overlap_start("abc"), "bc"); /// ``` #[inline] #[must_use] fn overlap_start(&self, other: &Self) -> &Self { unsafe { // SAFETY: The result of `string_overlap_index()` subtracted from `other.len()` will // always be on a character bound of `self`, since it is found by comparing directly the // bytes of the start of `self` and the end of `other`. Therefore, the range will be // within `self`'s bounds and also will uphold `str` invariants. self.slice_unchecked(0, other.len() - string_overlap_index(other, self)) } } /// Returns the substring which is both the suffix to `self` and the prefix to `other`. /// /// The returned string slice is a reference to the substring contained in `self`. /// /// # Example /// ``` /// use str_overlap::Overlap; /// /// assert_eq!("abc".overlap_end("bcd"), "bc"); /// ``` #[inline] #[must_use] fn overlap_end(&self, other: &Self) -> &Self { unsafe { // SAFETY: The result of `string_overlap_index()` will always be on a character bound of // `self`, since it is found from running over the CharIndices of `self`. Therefore, the // range will be within `self`'s bounds and also will uphold `str` invariants. self.slice_unchecked(string_overlap_index(self, other), self.len()) } } } #[cfg(test)] mod tests { use Overlap; #[test] fn partial_overlap_start() { assert_eq!("bcd".overlap_start("abc"), "bc"); } #[test] fn partial_overlap_end() { assert_eq!("abc".overlap_end("bcd"), "bc"); } #[test] fn full_overlap_start() { assert_eq!("abc".overlap_start("abc"), "abc"); } #[test] fn full_overlap_end() { assert_eq!("abc".overlap_end("abc"), "abc"); } #[test] fn no_overlap_start() { assert_eq!("abc".overlap_start("def"), ""); } #[test] fn no_overlap_end() { assert_eq!("abc".overlap_end("def"), ""); } #[test] fn other_substring_of_self_start() { assert_eq!("abcd".overlap_start("abc"), "abc"); } #[test] fn other_substring_of_self_end() { assert_eq!("abcd".overlap_end("bcd"), "bcd"); } #[test] fn self_substring_of_other_start() { assert_eq!("bcd".overlap_start("abcd"), "bcd"); } #[test] fn self_substring_other_end() { assert_eq!("abc".overlap_end("abcd"), "abc"); } #[test] fn only_checks_overlap_one_way_start() { assert_eq!("abc".overlap_start("bcd"), ""); } #[test] fn only_checks_overlap_one_way_end() { assert_eq!("bcd".overlap_end("abc"), ""); } #[test] fn self_empty_start() { assert_eq!("".overlap_start("abc"), ""); } #[test] fn self_empty_end() { assert_eq!("".overlap_end("abc"), ""); } #[test] fn other_empty_start() { assert_eq!("abc".overlap_start(""), ""); } #[test] fn other_empty_end() { assert_eq!("abc".overlap_end(""), ""); } #[test] fn all_empty_start() { assert_eq!("".overlap_end(""), ""); } #[test] fn all_empty_end() { assert_eq!("".overlap_start(""), ""); } #[test] fn multi_byte_start() { assert_eq!("語a日bc本".overlap_start("b日本語a"), "語a"); } #[test] fn multi_byte_end() { assert_eq!("b日本語a".overlap_end("語a日bc本"), "語a"); } }