resb/text/reader.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
// This file is part of ICU4X. For terms of use, please see the file
// called LICENSE at the top level of the ICU4X source tree
// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ).
mod parse_state;
use std::{borrow::Cow, fmt};
use nom::{
branch::alt,
bytes::complete::{tag, take, take_until, take_while1},
character::complete::{hex_digit1, i32, multispace1, not_line_ending, u32},
combinator::{eof, map, map_parser, map_res, opt, peek, value},
error::{context, convert_error, make_error, ContextError, ParseError, VerboseError},
multi::{many0, separated_list0},
sequence::{delimited, pair, preceded, terminated},
Finish, IResult, Parser,
};
use crate::bundle::{Int28, Key, Resource, ResourceBundle, Table};
use self::parse_state::ParseState;
/// Generates appropriate parsers for type ID strings.
///
/// A type may have one or more aliases. Note that if one type alias appears at
/// the beginning of another, the longer alias must appear first.
macro_rules! type_id {
($(#[$meta:meta])* $name:ident, $tag:literal) => {
$(#[$meta])*
fn $name<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>
{
context(stringify!($name), preceded(tag(":"),
tag($tag)
))(input)
}
};
($(#[$meta:meta])* $name:ident, $( $tag:literal ),+) => {
$(#[$meta])*
fn $name<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>
{
context(stringify!($name), preceded(tag(":"), alt((
$( tag($tag) ),+
))))(input)
}
};
}
type_id!(
/// Reads a root table resource type ID. Note that `"(nofallback)"` may only
/// appear on the root resource.
root_table_type,
"table(nofallback)",
"table"
);
type_id!(
/// Reads a string resource type ID.
string_type,
"string"
);
type_id!(
/// Reads an array resource type ID.
array_type,
"array"
);
type_id!(
/// Reads a table resource type ID.
table_type,
"table"
);
type_id!(
/// Reads a binary resource type ID.
binary_type,
"binary",
"bin"
);
type_id!(
/// Reads an integer resource type ID.
integer_type,
"integer",
"int"
);
type_id!(
/// Reads an integer vector resource type ID.
int_vector_type,
"intvector"
);
type_id!(
/// Reads an import resource type ID.
import_type,
"import"
);
type_id!(
/// Reads an include resource type ID.
include_type,
"include"
);
type_id!(
/// Reads a resource alias type ID.
alias_type,
"alias"
);
/// Reads any type ID which may appear as a leaf in the resource tree.
fn type_id_no_root<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"type_id",
alt((
string_type,
array_type,
table_type,
binary_type,
integer_type,
int_vector_type,
import_type,
include_type,
alias_type,
)),
)(input)
}
/// Reads one or more of the core set of ASCII characters used in keys.
///
/// It is presently unclear which characters appear in keys in practice, so in
/// the interest of a simplified parse, only characters which have been
/// encountered in real files are supported.
fn invariant_chars<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"invariant_chars",
take_while1(|c: char| c.is_alphanumeric() || c == '-'),
)(input)
}
/// Reads a single-line comment up to but not including the final newline.
fn eol_comment<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context("eol_comment", preceded(tag("//"), not_line_ending))(input)
}
/// Reads a multi-line comment.
///
/// Does not support nested comments.
fn delimited_comment<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"delimited_comment",
delimited(tag("/*"), take_until("*/"), tag("*/")),
)(input)
}
/// Reads one single-line or multi-line comment.
fn comment<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context("comment", alt((eol_comment, delimited_comment)))(input)
}
/// Reads a string.
///
/// Strings may be quote-delimited or not. While the partial specification
/// indicates that adjacent strings are combined, this is not supported in the
/// current version of the parser. No support for character escapes is
/// indicated.
///
/// The specification is unclear on how string encoding is to be handled, so
/// this parser assumes that strings are to be well-formed in the same encoding
/// as the rest of the file and representable in Unicode.
///
/// The specification also allows for the use of `\u` and `\U` literals. Support
/// for these has also been omitted for the time being.
///
/// See [`Reader`] for more details on the specification.
fn string<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, &'a str, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"string",
map(
token(alt((
delimited(tag("\""), take_until("\""), tag("\"")),
invariant_chars,
))),
|value| value.input(),
),
)(input)
}
/// Reads a signed or unsigned 32-bit integer.
fn integer<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, u32, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context("integer", token(alt((u32, map(i32, |value| value as u32)))))(input)
}
/// Reads and discards any number of comments and any amount of whitespace.
fn discardable<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, (), E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context("discardable", value((), many0(alt((comment, multispace1)))))(input)
}
/// Reads a single token as specified by the provided parser, surrounded by any
/// number of comments and any amount of whitespace.
fn token<'a, F, O, E>(mut parser: F) -> impl FnMut(ParseState<'a>) -> IResult<ParseState<'a>, O, E>
where
F: Parser<ParseState<'a>, O, E>,
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context("token", move |input: ParseState<'a>| {
let (rest, _) = discardable.parse(input)?;
let (rest, token) = parser.parse(rest)?;
let (rest, _) = discardable.parse(rest)?;
Ok((rest, token))
})
}
/// Generates appropriate matchers for simple string tokens.
macro_rules! simple_token {
($name:ident, $str:expr) => {
fn $name<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ParseState<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(stringify!($name), token(tag($str)))(input)
}
};
}
simple_token!(left_brace, "{");
simple_token!(right_brace, "}");
simple_token!(comma, ",");
/// Reads a table key.
fn key<'a, E>(state: ParseState<'a>) -> IResult<ParseState<'a>, Key<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
let input = state.clone();
context(
"key",
map(
terminated(
string,
// If the string is not followed by a resource, as indicated by
// the presence of a type ID or resource delimiter, it is not a
// key and we shouldn't record it.
peek(alt((type_id_no_root, left_brace))),
),
|key| {
let key = Key::from(key);
state.encounter_key(key.clone());
key
},
),
)(input)
}
/// Reads a single table entry, consisting of a key and an associated resource.
fn table_entry<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, (Key<'a>, Resource<'a>), E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context("table_entry", pair(key, resource))(input)
}
/// Generates a parser for a resource body, consisting of a resource
/// type-specific body layout delimited by curly braces.
macro_rules! resource_body {
($body_parser:expr) => {
delimited(left_brace, $body_parser, right_brace)
};
}
/// Generates a parser for a resource with a mandatory type ID.
macro_rules! resource {
($type:ident, $body_parser:expr) => {
preceded($type, resource_body!($body_parser))
};
}
/// Generates a parser for a resource with an optional type ID.
macro_rules! resource_opt_tag {
($type:ident, $body_parser:expr) => {
preceded(opt($type), resource_body!($body_parser))
};
}
/// Reads the body of a table resource, consisting of any number of key-resource
/// pair entries.
fn table_body<'a, E>(state: ParseState<'a>) -> IResult<ParseState<'a>, Table<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
let (state, entries) = context("table_body", many0(table_entry)).parse(state)?;
// Build the table struct itself.
let mut table = Table::new();
for (k, v) in entries {
table.insert(k, v);
}
Ok((state, table))
}
/// Reads a table resource, consisting of key-resource pairs.
///
/// Though it is not required in the resource bundle text, the final in-memory
/// representation sorts these pairs lexically by key.
fn table_resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"table_resource",
map(resource_opt_tag!(table_type, table_body), Resource::Table),
)(input)
}
/// Reads the root table resource, noting whether locale fallback is disabled.
fn root_table_resource<'a, E>(
input: ParseState<'a>,
) -> IResult<ParseState<'a>, (bool, Resource<'a>), E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
// Determine whether locale fallback is enabled by the type ID of the table,
// if present. The type ID is optional for tables, so it's okay if we don't
// find one, at which point we default to enabling locale fallback.
let (input, is_fallback_enabled) = opt(root_table_type).parse(input)?;
let is_fallback_enabled =
is_fallback_enabled.map_or(true, |type_id| type_id.input() != "table(nofallback)");
// Read the body of the root resource itself.
let (input, resource) = map(resource_body!(table_body), Resource::Table).parse(input)?;
Ok((input, (is_fallback_enabled, resource)))
}
/// Reads an array resource consisting of any number of hetereogeneously-typed
/// resources, optionally separated by commas.
fn array_resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
// Note that `nom`'s `separated_list0` parser requires that the separator
// always consume, so we can't use it for optional commas. However, the
// partial specification allows for trailing commas, so `terminated` works
// fine here.
context(
"array_resource",
map(
resource_opt_tag!(array_type, many0(terminated(resource, opt(comma)))),
Resource::Array,
),
)(input)
}
/// Reads an integer resource, consisting of a single 28-bit integer.
///
/// These integers have no inherent signedness; consumers specify whether a
/// signed or unsigned integer is desired. Because of this, note that 28-bit
/// integers require special handling in-memory. See [`Int28`] for more details.
fn int_resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"int_resource",
map(resource!(integer_type, integer), |value| {
Resource::Integer(Int28::from(value))
}),
)(input)
}
/// Reads an integer vector resource, consisting of any number of 32-bit
/// integers separated by commas. A trailing comma is optional.
fn int_vector_resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"int_vector_resource",
map(
resource!(
int_vector_type,
terminated(separated_list0(comma, integer), opt(comma))
),
Resource::IntVector,
),
)(input)
}
/// Reads a single byte represented as a pair of hex digits, each representing a
/// nibble of the byte. The high nibble is listed first.
fn binary_byte(input: &str) -> IResult<&str, u8> {
context(
"binary_byte",
map_res(map_parser(take(2usize), hex_digit1), |word| {
u8::from_str_radix(word, 16)
}),
)(input)
}
/// Reads a binary resource, consisting of a string of any number of paired hex
/// digits representing byte values.
///
/// See [`binary_byte`].
fn binary_resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
let (rest, string) = resource!(binary_type, string).parse(input.clone())?;
let (_, elements) = match many0(binary_byte).parse(string) {
Ok(elements) => elements,
Err(err) => {
println!("{err}");
return Err(nom::Err::Error(make_error(
input,
nom::error::ErrorKind::HexDigit,
)));
}
};
Ok((rest, Resource::Binary(Cow::from(elements))))
}
/// Reads a string resource.
///
/// Unlike other resources, a string may appear bare, without delimiting
/// braces.
///
/// See [`string`] for more details on the representation of strings.
fn string_resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"string_resource",
map(
alt((resource_opt_tag!(string_type, string), string)),
|value| Resource::String(Cow::from(value)),
),
)(input)
}
/// Reads a single resource.
fn resource<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, Resource<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"resource",
alt((
int_resource,
int_vector_resource,
string_resource,
binary_resource,
table_resource,
array_resource,
)),
)(input)
}
/// Reads a complete resource bundle.
fn bundle<'a, E>(input: ParseState<'a>) -> IResult<ParseState<'a>, ResourceBundle<'a>, E>
where
E: ParseError<ParseState<'a>> + ContextError<ParseState<'a>>,
{
context(
"bundle",
map(
terminated(
pair(
// Note that the name of the bundle is a string and not a key,
// and it is not included in the list of keys in the bundle.
string,
root_table_resource,
),
eof,
),
|(name, (is_locale_fallback_enabled, root))| {
ResourceBundle::new(Cow::from(name), root, is_locale_fallback_enabled)
},
),
)(input)
}
/// The `Reader` struct provides a means of parsing a [`ResourceBundle`] from
/// a string representing the contents of a text-based resource bundle file.
///
/// Note that the partial [specification] of the text-based format allows for
/// files to be in encodings other than UTF-8 and indicates that autodetection
/// of encoding should occur. At present, `Reader` only supports input as
/// UTF-8.
///
/// [specification]: https://github.com/unicode-org/icu-docs/blob/main/design/bnf_rb.txt
#[derive(Debug)]
#[non_exhaustive]
pub struct Reader;
impl Reader {
/// Parses the given string into a resource bundle.
///
/// Returns the parsed resource bundle and a list of the keys encountered in
/// the resource bundle in the order they were encountered.
pub fn read(input: &str) -> Result<(ResourceBundle, Vec<Key>), TextParserError> {
let input = ParseState::new(input);
let (final_state, bundle) = bundle::<VerboseError<ParseState>>(input.clone())
.finish()
.map_err(|err| TextParserError {
trace: convert_error(input, err),
})?;
let keys_in_discovery_order = final_state.take_keys().into_iter().collect();
Ok((bundle, keys_in_discovery_order))
}
}
#[derive(Debug)]
pub struct TextParserError {
trace: String,
}
impl fmt::Display for TextParserError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.write_fmt(format_args!(
"Parse error while reading text bundle:\n{}",
self.trace
))
}
}