mdman/format/text.rs
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626
//! Text formatter.
use crate::util::{header_text, unwrap};
use crate::EventIter;
use anyhow::{bail, Error};
use pulldown_cmark::{Alignment, Event, HeadingLevel, LinkType, Tag, TagEnd};
use std::fmt::Write;
use std::mem;
use url::Url;
pub struct TextFormatter {
url: Option<Url>,
}
impl TextFormatter {
pub fn new(url: Option<Url>) -> TextFormatter {
TextFormatter { url }
}
}
impl super::Formatter for TextFormatter {
fn render(&self, input: &str) -> Result<String, Error> {
TextRenderer::render(input, self.url.clone(), 0)
}
fn render_options_start(&self) -> &'static str {
// Tell pulldown_cmark to ignore this.
// This will be stripped out later.
"<![CDATA[\n"
}
fn render_options_end(&self) -> &'static str {
"]]>\n"
}
fn render_option(
&self,
params: &[&str],
block: &str,
_man_name: &str,
) -> Result<String, Error> {
let rendered_options = params
.iter()
.map(|param| TextRenderer::render(param, self.url.clone(), 0))
.collect::<Result<Vec<_>, Error>>()?;
let trimmed: Vec<_> = rendered_options.iter().map(|o| o.trim()).collect();
// Wrap in HTML tags, they will be stripped out during rendering.
Ok(format!(
"<dt>{}</dt>\n<dd>\n{}</dd>\n<br>\n",
trimmed.join(", "),
block
))
}
fn linkify_man_to_md(&self, name: &str, section: u8) -> Result<String, Error> {
Ok(format!("`{}`({})", name, section))
}
}
struct TextRenderer<'e> {
output: String,
indent: usize,
/// The current line being written. Once a line break is encountered (such
/// as starting a new paragraph), this will be written to `output` via
/// `flush`.
line: String,
/// The current word being written. Once a break is encountered (such as a
/// space) this will be written to `line` via `flush_word`.
word: String,
parser: EventIter<'e>,
/// The base URL used for relative URLs.
url: Option<Url>,
table: Table,
}
impl<'e> TextRenderer<'e> {
fn render(input: &str, url: Option<Url>, indent: usize) -> Result<String, Error> {
let parser = crate::md_parser(input, url.clone());
let output = String::with_capacity(input.len() * 3 / 2);
let mut mr = TextRenderer {
output,
indent,
line: String::new(),
word: String::new(),
parser,
url,
table: Table::new(),
};
mr.push_md()?;
Ok(mr.output)
}
fn push_md(&mut self) -> Result<(), Error> {
// If this is true, this is inside a cdata block used for hiding
// content from pulldown_cmark.
let mut in_cdata = false;
// The current list stack. None if unordered, Some if ordered with the
// given number as the current index.
let mut list: Vec<Option<u64>> = Vec::new();
// Used in some cases where spacing isn't desired.
let mut suppress_paragraph = false;
// Whether or not word-wrapping is enabled.
let mut wrap_text = true;
let mut last_seen_link_data = None;
while let Some((event, range)) = self.parser.next() {
let this_suppress_paragraph = suppress_paragraph;
// Always reset suppression, even if the next event isn't a
// paragraph. This is in essence, a 1-token lookahead where the
// suppression is only enabled if the next event is a paragraph.
suppress_paragraph = false;
match event {
Event::Start(tag) => {
match tag {
Tag::Paragraph => {
if !this_suppress_paragraph {
self.flush();
}
}
Tag::Heading { level, .. } => {
self.flush();
if level == HeadingLevel::H1 {
let text = header_text(&mut self.parser)?;
self.push_to_line(&text.to_uppercase());
self.hard_break();
self.hard_break();
} else if level == HeadingLevel::H2 {
let text = header_text(&mut self.parser)?;
self.push_to_line(&text.to_uppercase());
self.flush();
self.indent = 7;
} else {
let text = header_text(&mut self.parser)?;
self.push_indent((level as usize - 2) * 3);
self.push_to_line(&text);
self.flush();
self.indent = (level as usize - 1) * 3 + 1;
}
}
Tag::BlockQuote(_kind) => {
self.indent += 3;
}
Tag::CodeBlock(_kind) => {
self.flush();
wrap_text = false;
self.indent += 4;
}
Tag::List(start) => list.push(start),
Tag::Item => {
self.flush();
match list.last_mut().expect("item must have list start") {
// Ordered list.
Some(n) => {
self.push_indent(self.indent);
write!(self.line, "{}.", n)?;
*n += 1;
}
// Unordered list.
None => {
self.push_indent(self.indent);
self.push_to_line("o ")
}
}
self.indent += 3;
suppress_paragraph = true;
}
Tag::FootnoteDefinition(_label) => unimplemented!(),
Tag::Table(alignment) => {
assert!(self.table.alignment.is_empty());
self.flush();
self.table.alignment.extend(alignment);
let table = self.table.process(&mut self.parser, self.indent)?;
self.output.push_str(&table);
self.hard_break();
self.table = Table::new();
}
Tag::TableHead | Tag::TableRow | Tag::TableCell => {
bail!("unexpected table element")
}
Tag::Emphasis => {}
Tag::Strong => {}
// Strikethrough isn't usually supported for TTY.
Tag::Strikethrough => self.word.push_str("~~"),
Tag::Link {
link_type,
dest_url,
..
} => {
last_seen_link_data = Some((link_type.clone(), dest_url.to_owned()));
if dest_url.starts_with('#') {
// In a man page, page-relative anchors don't
// have much meaning.
continue;
}
match link_type {
LinkType::Autolink | LinkType::Email => {
// The text is a copy of the URL, which is not needed.
match self.parser.next() {
Some((Event::Text(_), _range)) => {}
_ => bail!("expected text after autolink"),
}
}
LinkType::Inline
| LinkType::Reference
| LinkType::Collapsed
| LinkType::Shortcut => {}
// This is currently unused. This is only
// emitted with a broken link callback, but I
// felt it is too annoying to escape `[` in
// option descriptions.
LinkType::ReferenceUnknown
| LinkType::CollapsedUnknown
| LinkType::ShortcutUnknown => {
bail!(
"link with missing reference `{}` located at offset {}",
dest_url,
range.start
);
}
}
}
Tag::Image { .. } => {
bail!("images are not currently supported")
}
Tag::HtmlBlock { .. }
| Tag::MetadataBlock { .. }
| Tag::DefinitionList
| Tag::DefinitionListTitle
| Tag::DefinitionListDefinition => {}
}
}
Event::End(tag_end) => match &tag_end {
TagEnd::Paragraph => {
self.flush();
self.hard_break();
}
TagEnd::Heading(..) => {}
TagEnd::BlockQuote(..) => {
self.indent -= 3;
}
TagEnd::CodeBlock => {
self.hard_break();
wrap_text = true;
self.indent -= 4;
}
TagEnd::List(..) => {
list.pop();
}
TagEnd::Item => {
self.flush();
self.indent -= 3;
self.hard_break();
}
TagEnd::FootnoteDefinition => {}
TagEnd::Table => {}
TagEnd::TableHead => {}
TagEnd::TableRow => {}
TagEnd::TableCell => {}
TagEnd::Emphasis => {}
TagEnd::Strong => {}
TagEnd::Strikethrough => self.word.push_str("~~"),
TagEnd::Link => {
if let Some((link_type, ref dest_url)) = last_seen_link_data {
if dest_url.starts_with('#') {
continue;
}
match link_type {
LinkType::Autolink | LinkType::Email => {}
LinkType::Inline
| LinkType::Reference
| LinkType::Collapsed
| LinkType::Shortcut => self.flush_word(),
_ => {
panic!("unexpected tag {:?}", tag_end);
}
}
self.flush_word();
write!(self.word, "<{}>", dest_url)?;
}
}
TagEnd::HtmlBlock { .. }
| TagEnd::MetadataBlock { .. }
| TagEnd::DefinitionList
| TagEnd::DefinitionListTitle
| TagEnd::Image
| TagEnd::DefinitionListDefinition => {}
},
Event::Text(t) | Event::Code(t) => {
if wrap_text {
let chunks = split_chunks(&t);
for chunk in chunks {
if chunk == " " {
self.flush_word();
} else {
self.word.push_str(chunk);
}
}
} else {
for line in t.lines() {
self.push_indent(self.indent);
self.push_to_line(line);
self.flush();
}
}
}
Event::Html(t) => {
if t.starts_with("<![CDATA[") {
// CDATA is a special marker used for handling options.
in_cdata = true;
self.flush();
} else if in_cdata {
if t.trim().ends_with("]]>") {
in_cdata = false;
} else {
let trimmed = t.trim();
if trimmed.is_empty() {
continue;
}
if trimmed == "<br>" {
self.hard_break();
} else if trimmed.starts_with("<dt>") {
let opts = unwrap(trimmed, "<dt>", "</dt>");
self.push_indent(self.indent);
self.push_to_line(opts);
self.flush();
} else if trimmed.starts_with("<dd>") {
let mut def = String::new();
while let Some((Event::Html(t), _range)) = self.parser.next() {
if t.starts_with("</dd>") {
break;
}
def.push_str(&t);
}
let rendered =
TextRenderer::render(&def, self.url.clone(), self.indent + 4)?;
self.push_to_line(rendered.trim_end());
self.flush();
} else {
self.push_to_line(&t);
self.flush();
}
}
} else {
self.push_to_line(&t);
self.flush();
}
}
Event::FootnoteReference(_t) => {}
Event::SoftBreak => self.flush_word(),
Event::HardBreak => self.flush(),
Event::Rule => {
self.flush();
self.push_indent(self.indent);
self.push_to_line(&"_".repeat(79 - self.indent * 2));
self.flush();
}
Event::TaskListMarker(_b) => unimplemented!(),
Event::InlineHtml(..) => unimplemented!(),
Event::InlineMath(..) => unimplemented!(),
Event::DisplayMath(..) => unimplemented!(),
}
}
Ok(())
}
fn flush(&mut self) {
self.flush_word();
if !self.line.is_empty() {
self.output.push_str(&self.line);
self.output.push('\n');
self.line.clear();
}
}
fn hard_break(&mut self) {
self.flush();
if !self.output.ends_with("\n\n") {
self.output.push('\n');
}
}
fn flush_word(&mut self) {
if self.word.is_empty() {
return;
}
if self.line.len() + self.word.len() >= 79 {
self.output.push_str(&self.line);
self.output.push('\n');
self.line.clear();
}
if self.line.is_empty() {
self.push_indent(self.indent);
self.line.push_str(&self.word);
} else {
self.line.push(' ');
self.line.push_str(&self.word);
}
self.word.clear();
}
fn push_indent(&mut self, indent: usize) {
for _ in 0..indent {
self.line.push(' ');
}
}
fn push_to_line(&mut self, text: &str) {
self.flush_word();
self.line.push_str(text);
}
}
/// Splits the text on whitespace.
///
/// Consecutive whitespace is collapsed to a single ' ', and is included as a
/// separate element in the result.
fn split_chunks(text: &str) -> Vec<&str> {
let mut result = Vec::new();
let mut start = 0;
while start < text.len() {
match text[start..].find(' ') {
Some(i) => {
if i != 0 {
result.push(&text[start..start + i]);
}
result.push(" ");
// Skip past whitespace.
match text[start + i..].find(|c| c != ' ') {
Some(n) => {
start = start + i + n;
}
None => {
break;
}
}
}
None => {
result.push(&text[start..]);
break;
}
}
}
result
}
struct Table {
alignment: Vec<Alignment>,
rows: Vec<Vec<String>>,
row: Vec<String>,
cell: String,
}
impl Table {
fn new() -> Table {
Table {
alignment: Vec::new(),
rows: Vec::new(),
row: Vec::new(),
cell: String::new(),
}
}
/// Processes table events and generates a text table.
fn process(&mut self, parser: &mut EventIter<'_>, indent: usize) -> Result<String, Error> {
while let Some((event, _range)) = parser.next() {
match event {
Event::Start(tag) => match tag {
Tag::TableHead
| Tag::TableRow
| Tag::TableCell
| Tag::Emphasis
| Tag::Strong => {}
Tag::Strikethrough => self.cell.push_str("~~"),
// Links not yet supported, they usually won't fit.
Tag::Link { .. } => {}
_ => bail!("unexpected tag in table: {:?}", tag),
},
Event::End(tag_end) => match tag_end {
TagEnd::Table => return self.render(indent),
TagEnd::TableCell => {
let cell = mem::replace(&mut self.cell, String::new());
self.row.push(cell);
}
TagEnd::TableHead | TagEnd::TableRow => {
let row = mem::replace(&mut self.row, Vec::new());
self.rows.push(row);
}
TagEnd::Strikethrough => self.cell.push_str("~~"),
_ => {}
},
Event::Text(t) | Event::Code(t) => {
self.cell.push_str(&t);
}
Event::Html(t) => bail!("html unsupported in tables: {:?}", t),
_ => bail!("unexpected event in table: {:?}", event),
}
}
bail!("table end not reached");
}
fn render(&self, indent: usize) -> Result<String, Error> {
// This is an extremely primitive layout routine.
// First compute the potential maximum width of each cell.
// 2 for 1 space margin on left and right.
let width_acc = vec![2; self.alignment.len()];
let mut col_widths = self
.rows
.iter()
.map(|row| row.iter().map(|cell| cell.len()))
.fold(width_acc, |mut acc, row| {
acc.iter_mut()
.zip(row)
// +3 for left/right margin and | symbol
.for_each(|(a, b)| *a = (*a).max(b + 3));
acc
});
// Shrink each column until it fits the total width, proportional to
// the columns total percent width.
let max_width = 78 - indent;
// Include total len for | characters, and +1 for final |.
let total_width = col_widths.iter().sum::<usize>() + col_widths.len() + 1;
if total_width > max_width {
let to_shrink = total_width - max_width;
// Compute percentage widths, and shrink each column based on its
// total percentage.
for width in &mut col_widths {
let percent = *width as f64 / total_width as f64;
*width -= (to_shrink as f64 * percent).ceil() as usize;
}
}
// Start rendering.
let mut result = String::new();
// Draw the horizontal line separating each row.
let mut row_line = String::new();
row_line.push_str(&" ".repeat(indent));
row_line.push('+');
let lines = col_widths
.iter()
.map(|width| "-".repeat(*width))
.collect::<Vec<_>>();
row_line.push_str(&lines.join("+"));
row_line.push('+');
row_line.push('\n');
// Draw top of the table.
result.push_str(&row_line);
// Draw each row.
for row in &self.rows {
// Word-wrap and fill each column as needed.
let filled = fill_row(row, &col_widths, &self.alignment);
// Need to transpose the cells across rows for cells that span
// multiple rows.
let height = filled.iter().map(|c| c.len()).max().unwrap();
for row_i in 0..height {
result.push_str(&" ".repeat(indent));
result.push('|');
for filled_row in &filled {
let cell = &filled_row[row_i];
result.push_str(cell);
result.push('|');
}
result.push('\n');
}
result.push_str(&row_line);
}
Ok(result)
}
}
/// Formats a row, filling cells with spaces and word-wrapping text.
///
/// Returns a vec of cells, where each cell is split into multiple lines.
fn fill_row(row: &[String], col_widths: &[usize], alignment: &[Alignment]) -> Vec<Vec<String>> {
let mut cell_lines = row
.iter()
.zip(col_widths)
.zip(alignment)
.map(|((cell, width), alignment)| fill_cell(cell, *width - 2, *alignment))
.collect::<Vec<_>>();
// Fill each cell to match the maximum vertical height of the tallest cell.
let max_lines = cell_lines.iter().map(|cell| cell.len()).max().unwrap();
for (cell, width) in cell_lines.iter_mut().zip(col_widths) {
if cell.len() < max_lines {
cell.extend(std::iter::repeat(" ".repeat(*width)).take(max_lines - cell.len()));
}
}
cell_lines
}
/// Formats a cell. Word-wraps based on width, and adjusts based on alignment.
///
/// Returns a vec of lines for the cell.
fn fill_cell(text: &str, width: usize, alignment: Alignment) -> Vec<String> {
let fill_width = |text: &str| match alignment {
Alignment::None | Alignment::Left => format!(" {:<width$} ", text, width = width),
Alignment::Center => format!(" {:^width$} ", text, width = width),
Alignment::Right => format!(" {:>width$} ", text, width = width),
};
if text.len() < width {
// No wrapping necessary, just format.
vec![fill_width(text)]
} else {
// Word-wrap the cell.
let mut result = Vec::new();
let mut line = String::new();
for word in text.split_whitespace() {
if line.len() + word.len() >= width {
// todo: word.len() > width
result.push(fill_width(&line));
line.clear();
}
if line.is_empty() {
line.push_str(word);
} else {
line.push(' ');
line.push_str(&word);
}
}
if !line.is_empty() {
result.push(fill_width(&line));
}
result
}
}