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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352
//! Functionality for unfilling and refilling text.
use crate::core::display_width;
use crate::line_ending::NonEmptyLines;
use crate::{fill, LineEnding, Options};
/// Unpack a paragraph of already-wrapped text.
///
/// This function attempts to recover the original text from a single
/// paragraph of wrapped text, such as what [`fill()`] would produce.
/// This means that it turns
///
/// ```text
/// textwrap: a small
/// library for
/// wrapping text.
/// ```
///
/// back into
///
/// ```text
/// textwrap: a small library for wrapping text.
/// ```
///
/// In addition, it will recognize a common prefix and a common line
/// ending among the lines.
///
/// The prefix of the first line is returned in
/// [`Options::initial_indent`] and the prefix (if any) of the the
/// other lines is returned in [`Options::subsequent_indent`].
///
/// Line ending is returned in [`Options::line_ending`]. If line ending
/// can not be confidently detected (mixed or no line endings in the
/// input), [`LineEnding::LF`] will be returned.
///
/// In addition to `' '`, the prefixes can consist of characters used
/// for unordered lists (`'-'`, `'+'`, and `'*'`) and block quotes
/// (`'>'`) in Markdown as well as characters often used for inline
/// comments (`'#'` and `'/'`).
///
/// The text must come from a single wrapped paragraph. This means
/// that there can be no empty lines (`"\n\n"` or `"\r\n\r\n"`) within
/// the text. It is unspecified what happens if `unfill` is called on
/// more than one paragraph of text.
///
/// # Examples
///
/// ```
/// use textwrap::{LineEnding, unfill};
///
/// let (text, options) = unfill("\
/// * This is an
/// example of
/// a list item.
/// ");
///
/// assert_eq!(text, "This is an example of a list item.\n");
/// assert_eq!(options.initial_indent, "* ");
/// assert_eq!(options.subsequent_indent, " ");
/// assert_eq!(options.line_ending, LineEnding::LF);
/// ```
pub fn unfill(text: &str) -> (String, Options<'_>) {
let prefix_chars: &[_] = &[' ', '-', '+', '*', '>', '#', '/'];
let mut options = Options::new(0);
for (idx, line) in text.lines().enumerate() {
options.width = std::cmp::max(options.width, display_width(line));
let without_prefix = line.trim_start_matches(prefix_chars);
let prefix = &line[..line.len() - without_prefix.len()];
if idx == 0 {
options.initial_indent = prefix;
} else if idx == 1 {
options.subsequent_indent = prefix;
} else if idx > 1 {
for ((idx, x), y) in prefix.char_indices().zip(options.subsequent_indent.chars()) {
if x != y {
options.subsequent_indent = &prefix[..idx];
break;
}
}
if prefix.len() < options.subsequent_indent.len() {
options.subsequent_indent = prefix;
}
}
}
let mut unfilled = String::with_capacity(text.len());
let mut detected_line_ending = None;
for (idx, (line, ending)) in NonEmptyLines(text).enumerate() {
if idx == 0 {
unfilled.push_str(&line[options.initial_indent.len()..]);
} else {
unfilled.push(' ');
unfilled.push_str(&line[options.subsequent_indent.len()..]);
}
match (detected_line_ending, ending) {
(None, Some(_)) => detected_line_ending = ending,
(Some(LineEnding::CRLF), Some(LineEnding::LF)) => detected_line_ending = ending,
_ => (),
}
}
// Add back a line ending if `text` ends with the one we detect.
if let Some(line_ending) = detected_line_ending {
if text.ends_with(line_ending.as_str()) {
unfilled.push_str(line_ending.as_str());
}
}
options.line_ending = detected_line_ending.unwrap_or(LineEnding::LF);
(unfilled, options)
}
/// Refill a paragraph of wrapped text with a new width.
///
/// This function will first use [`unfill()`] to remove newlines from
/// the text. Afterwards the text is filled again using [`fill()`].
///
/// The `new_width_or_options` argument specify the new width and can
/// specify other options as well — except for
/// [`Options::initial_indent`] and [`Options::subsequent_indent`],
/// which are deduced from `filled_text`.
///
/// # Examples
///
/// ```
/// use textwrap::refill;
///
/// // Some loosely wrapped text. The "> " prefix is recognized automatically.
/// let text = "\
/// > Memory
/// > safety without garbage
/// > collection.
/// ";
///
/// assert_eq!(refill(text, 20), "\
/// > Memory safety
/// > without garbage
/// > collection.
/// ");
///
/// assert_eq!(refill(text, 40), "\
/// > Memory safety without garbage
/// > collection.
/// ");
///
/// assert_eq!(refill(text, 60), "\
/// > Memory safety without garbage collection.
/// ");
/// ```
///
/// You can also reshape bullet points:
///
/// ```
/// use textwrap::refill;
///
/// let text = "\
/// - This is my
/// list item.
/// ";
///
/// assert_eq!(refill(text, 20), "\
/// - This is my list
/// item.
/// ");
/// ```
pub fn refill<'a, Opt>(filled_text: &str, new_width_or_options: Opt) -> String
where
Opt: Into<Options<'a>>,
{
let mut new_options = new_width_or_options.into();
let (text, options) = unfill(filled_text);
// The original line ending is kept by `unfill`.
let stripped = text.strip_suffix(options.line_ending.as_str());
let new_line_ending = new_options.line_ending.as_str();
new_options.initial_indent = options.initial_indent;
new_options.subsequent_indent = options.subsequent_indent;
let mut refilled = fill(stripped.unwrap_or(&text), new_options);
// Add back right line ending if we stripped one off above.
if stripped.is_some() {
refilled.push_str(new_line_ending);
}
refilled
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn unfill_simple() {
let (text, options) = unfill("foo\nbar");
assert_eq!(text, "foo bar");
assert_eq!(options.width, 3);
assert_eq!(options.line_ending, LineEnding::LF);
}
#[test]
fn unfill_no_new_line() {
let (text, options) = unfill("foo bar");
assert_eq!(text, "foo bar");
assert_eq!(options.width, 7);
assert_eq!(options.line_ending, LineEnding::LF);
}
#[test]
fn unfill_simple_crlf() {
let (text, options) = unfill("foo\r\nbar");
assert_eq!(text, "foo bar");
assert_eq!(options.width, 3);
assert_eq!(options.line_ending, LineEnding::CRLF);
}
#[test]
fn unfill_mixed_new_lines() {
let (text, options) = unfill("foo\r\nbar\nbaz");
assert_eq!(text, "foo bar baz");
assert_eq!(options.width, 3);
assert_eq!(options.line_ending, LineEnding::LF);
}
#[test]
fn test_unfill_consecutive_different_prefix() {
let (text, options) = unfill("foo\n*\n/");
assert_eq!(text, "foo * /");
assert_eq!(options.width, 3);
assert_eq!(options.line_ending, LineEnding::LF);
}
#[test]
fn unfill_trailing_newlines() {
let (text, options) = unfill("foo\nbar\n\n\n");
assert_eq!(text, "foo bar\n");
assert_eq!(options.width, 3);
}
#[test]
fn unfill_mixed_trailing_newlines() {
let (text, options) = unfill("foo\r\nbar\n\r\n\n");
assert_eq!(text, "foo bar\n");
assert_eq!(options.width, 3);
assert_eq!(options.line_ending, LineEnding::LF);
}
#[test]
fn unfill_trailing_crlf() {
let (text, options) = unfill("foo bar\r\n");
assert_eq!(text, "foo bar\r\n");
assert_eq!(options.width, 7);
assert_eq!(options.line_ending, LineEnding::CRLF);
}
#[test]
fn unfill_initial_indent() {
let (text, options) = unfill(" foo\nbar\nbaz");
assert_eq!(text, "foo bar baz");
assert_eq!(options.width, 5);
assert_eq!(options.initial_indent, " ");
}
#[test]
fn unfill_differing_indents() {
let (text, options) = unfill(" foo\n bar\n baz");
assert_eq!(text, "foo bar baz");
assert_eq!(options.width, 7);
assert_eq!(options.initial_indent, " ");
assert_eq!(options.subsequent_indent, " ");
}
#[test]
fn unfill_list_item() {
let (text, options) = unfill("* foo\n bar\n baz");
assert_eq!(text, "foo bar baz");
assert_eq!(options.width, 5);
assert_eq!(options.initial_indent, "* ");
assert_eq!(options.subsequent_indent, " ");
}
#[test]
fn unfill_multiple_char_prefix() {
let (text, options) = unfill(" // foo bar\n // baz\n // quux");
assert_eq!(text, "foo bar baz quux");
assert_eq!(options.width, 14);
assert_eq!(options.initial_indent, " // ");
assert_eq!(options.subsequent_indent, " // ");
}
#[test]
fn unfill_block_quote() {
let (text, options) = unfill("> foo\n> bar\n> baz");
assert_eq!(text, "foo bar baz");
assert_eq!(options.width, 5);
assert_eq!(options.initial_indent, "> ");
assert_eq!(options.subsequent_indent, "> ");
}
#[test]
fn unfill_only_prefixes_issue_466() {
// Test that we don't crash if the first line has only prefix
// chars *and* the second line is shorter than the first line.
let (text, options) = unfill("######\nfoo");
assert_eq!(text, " foo");
assert_eq!(options.width, 6);
assert_eq!(options.initial_indent, "######");
assert_eq!(options.subsequent_indent, "");
}
#[test]
fn unfill_trailing_newlines_issue_466() {
// Test that we don't crash on a '\r' following a string of
// '\n'. The problem was that we removed both kinds of
// characters in one code path, but not in the other.
let (text, options) = unfill("foo\n##\n\n\r");
// The \n\n changes subsequent_indent to "".
assert_eq!(text, "foo ## \r");
assert_eq!(options.width, 3);
assert_eq!(options.initial_indent, "");
assert_eq!(options.subsequent_indent, "");
}
#[test]
fn unfill_whitespace() {
assert_eq!(unfill("foo bar").0, "foo bar");
}
#[test]
fn refill_convert_lf_to_crlf() {
let options = Options::new(5).line_ending(LineEnding::CRLF);
assert_eq!(refill("foo\nbar\n", options), "foo\r\nbar\r\n",);
}
#[test]
fn refill_convert_crlf_to_lf() {
let options = Options::new(5).line_ending(LineEnding::LF);
assert_eq!(refill("foo\r\nbar\r\n", options), "foo\nbar\n",);
}
#[test]
fn refill_convert_mixed_newlines() {
let options = Options::new(5).line_ending(LineEnding::CRLF);
assert_eq!(refill("foo\r\nbar\n", options), "foo\r\nbar\r\n",);
}
#[test]
fn refill_defaults_to_lf() {
assert_eq!(refill("foo bar baz", 5), "foo\nbar\nbaz");
}
}