This library exists to provide case conversion between common cases like CamelCase and snake_case. It is intended to be unicode aware, internally consistent, and reasonably well performing.
Word boundaries are defined by non-alphanumeric characters, as well as within those words in this manner:
- If an uppercase character is followed by lowercase letters, a word boundary is considered to be just prior to that uppercase character.
- If multiple uppercase characters are consecutive, they are considered to be within a single word, except that the last will be part of the next word if it is followed by lowercase characters (see rule 1).
That is, "HelloWorld" is segmented Hello|World whereas "XMLHttpRequest" is
segmented XML|Http|Request.
Characters not within words (such as spaces, punctuations, and underscores) are not included in the output string except as they are a part of the case being converted to. Multiple adjacent word boundaries (such as a series of underscores) are folded into one. ("hello__world" in snake case is therefore "hello_world", not the exact same string). Leading or trailing word boundary indicators are dropped, except insofar as CamelCase capitalizes the first word.
- UpperCamelCase
- lowerCamelCase
- snake_case
- kebab-case
- SHOUTY_SNAKE_CASE
- Title Case
- SHOUTY-KEBAB-CASE
- Train-Case
The minimum supported Rust version for this crate is 1.56.0. This may change in minor or patch releases, but we probably won't ever require a very recent version. If you would like to have a stronger guarantee than that, please open an issue.
heck is distributed under the terms of both the MIT license and the Apache License (Version 2.0).
See LICENSE-APACHE and LICENSE-MIT for details.
