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
// Copyright 2018 Fredrik Portström <https://portstrom.com>
// Copyright (c) 2023 Olivier ROLAND
// This is free software distributed under the terms specified in
// the file LICENSE at the top-level directory of this distribution.
//! Parse XML dumps exported from Mediawiki.
//!
//! This module parses [XML dumps](https://www.mediawiki.org/wiki/Help:Export) exported from Mediawiki, providing each page from the dump through an iterator. This is useful for parsing the [dumps from Wikipedia and other Wikimedia projects](https://dumps.wikimedia.org).
//!
//! # Caution
//!
//! If you need to parse any wiki text extracted from a dump, please use the crate Parse Wiki Text ([crates.io](https://crates.io/crates/parse_wiki_text), [Github](https://github.com/portstrom/parse_wiki_text)). Correctly parsing wiki text requires dealing with an astonishing amount of difficult and counterintuitive cases. Parse Wiki Text automatically deals with all these cases, giving you an unambiguous tree of parsed elements that is easy to work with.
//!
//! # Limitations
//!
//! This module only parses dumps containing only one revision of each page. This is what you get from the page `Special:Export` when enabling the option “Include only the current revision, not the full history”, as well as what you get from the Wikimedia dumps with file names ending with `-pages-articles.xml.bz2`.
//!
//! This module ignores the `siteinfo` element, every child element of the `page` element except `ns`, `revision` and `title`, and every element inside the `revision` element except `format`, `model` and `text`.
//!
//! Until there is a real use case that justifies going beyond these limitations, they will remain in order to avoid premature design driven by imagined requirements.
//!
//! # Examples
//!
//! Parse a bzip2 compressed file and distinguish ordinary articles from other pages. A running example with complete error handling is available in the `examples` folder.
//!
//! ```rust,no_run
//! extern crate bzip2;
//! extern crate parse_mediawiki_dump_reboot;
//! use parse_mediawiki_dump_reboot::schema::Namespace;
//!
//! fn main() {
//! let file = std::fs::File::open("example.xml.bz2").unwrap();
//! let file = std::io::BufReader::new(file);
//! let file = bzip2::bufread::MultiBzDecoder::new(file);
//! let file = std::io::BufReader::new(file);
//! for result in parse_mediawiki_dump_reboot::parse(file) {
//! match result {
//! Err(error) => {
//! eprintln!("Error: {}", error);
//! break;
//! }
//! Ok(page) => if page.namespace == Namespace::Main && match &page.format {
//! None => false,
//! Some(format) => format == "text/x-wiki"
//! } && match &page.model {
//! None => false,
//! Some(model) => model == "wikitext"
//! } {
//! println!(
//! "The page {title:?} is an ordinary article with byte length {length}.",
//! title = page.title,
//! length = page.text.len()
//! );
//! } else {
//! println!("The page {:?} has something special to it.", page.title);
//! }
//! }
//! }
//! }
//! ```
#![forbid(unsafe_code)]
extern crate quick_xml;
pub mod schema;
use quick_xml::{events::Event, name::ResolveResult, NsReader};
use schema::Namespace;
use std::io::BufRead;
enum PageChildElement {
Ns,
Revision,
Title,
Unknown,
}
enum RevisionChildElement {
Format,
Model,
Text,
Unknown,
}
#[derive(Debug)]
/// The error type for `Parser`.
pub enum Error {
/// Format not matching expectations.
///
/// Indicates the position in the stream.
Format(usize),
/// The source contains a feature not supported by the parser.
///
/// In particular, this means a `page` element contains more than one `revision` element.
NotSupported(usize),
/// Error from the XML reader.
XmlReader(quick_xml::Error),
}
/// Parsed page.
///
/// Parsed from the `page` element.
///
/// Although the `format` and `model` elements are defined as mandatory in the [schema](https://www.mediawiki.org/xml/export-0.10.xsd), previous versions of the schema don't contain them. Therefore the corresponding fields can be `None`.
#[derive(Debug)]
pub struct Page {
/// The format of the revision if any.
///
/// Parsed from the text content of the `format` element in the `revision` element. `None` if the element is not present.
///
/// For ordinary articles the format is `text/x-wiki`.
pub format: Option<String>,
/// The model of the revision if any.
///
/// Parsed from the text content of the `model` element in the `revision` element. `None` if the element is not present.
///
/// For ordinary articles the model is `wikitext`.
pub model: Option<String>,
/// The namespace of the page.
///
/// Parsed from the text content of the `ns` element in the `page` element.
///
/// For ordinary articles the namespace is 0.
pub namespace: Namespace,
/// The text of the revision.
///
/// Parsed from the text content of the `text` element in the `revision` element.
pub text: String,
/// The title of the page.
///
/// Parsed from the text content of the `title` element in the `page` element.
pub title: String,
}
/// Parser working as an iterator over pages.
pub struct Parser<R: BufRead> {
buffer: Vec<u8>,
reader: NsReader<R>,
started: bool,
}
impl std::fmt::Display for Error {
fn fmt(&self, formatter: &mut std::fmt::Formatter) -> std::fmt::Result {
match self {
Error::Format(position) => write!(formatter, "Invalid format at position {}", position),
Error::NotSupported(position) => write!(
formatter,
"The element at position {} is not supported",
position
),
Error::XmlReader(error) => error.fmt(formatter),
}
}
}
impl From<quick_xml::Error> for Error {
fn from(value: quick_xml::Error) -> Self {
Error::XmlReader(value)
}
}
impl<R: BufRead> Iterator for Parser<R> {
type Item = Result<Page, Error>;
fn next(&mut self) -> Option<Self::Item> {
Some(match next(self) {
Err(error) => Err(error),
Ok(item) => Ok(item?),
})
}
}
fn match_namespace(namespace: ResolveResult<'_>) -> bool {
matches!(
namespace,
ResolveResult::Bound(quick_xml::name::Namespace(
b"http://www.mediawiki.org/xml/export-0.10/",
))
)
}
fn next(parser: &mut Parser<impl BufRead>) -> Result<Option<Page>, Error> {
if !parser.started {
loop {
parser.buffer.clear();
if let (namespace, Event::Start(event)) =
parser.reader.read_resolved_event_into(&mut parser.buffer)?
{
if match_namespace(namespace) && event.local_name().as_ref() == b"mediawiki" {
break;
}
return Err(Error::Format(parser.reader.buffer_position()));
}
}
parser.started = true;
}
loop {
parser.buffer.clear();
if !match parser
.reader
//.read_namespaced_event(&mut parser.buffer, &mut parser.namespace_buffer)?
.read_resolved_event_into(&mut parser.buffer)?
{
(_, Event::End(_)) => return Ok(None),
(namespace, Event::Start(event)) => {
match_namespace(namespace) && event.local_name().as_ref() == b"page"
}
_ => continue,
} {
skip_element(parser)?;
continue;
}
let mut format = None;
let mut model = None;
let mut namespace = None;
let mut text = None;
let mut title = None;
loop {
parser.buffer.clear();
match match parser.reader.read_resolved_event_into(&mut parser.buffer)? {
(_, Event::End(_)) => {
return match (namespace, text, title) {
(Some(namespace), Some(text), Some(title)) => Ok(Some(Page {
format,
model,
namespace,
text,
title,
})),
_ => Err(Error::Format(parser.reader.buffer_position())),
}
}
(namespace, Event::Start(event)) => {
if match_namespace(namespace) {
match event.local_name().as_ref() {
b"ns" => PageChildElement::Ns,
b"revision" => PageChildElement::Revision,
b"title" => PageChildElement::Title,
_ => PageChildElement::Unknown,
}
} else {
PageChildElement::Unknown
}
}
_ => continue,
} {
PageChildElement::Ns => match parse_text(parser, &namespace)?.parse::<i32>() {
Err(_) => return Err(Error::Format(parser.reader.buffer_position())),
Ok(value) => {
namespace = Some(Namespace::from(value));
continue;
}
},
PageChildElement::Revision => {
if text.is_some() {
return Err(Error::NotSupported(parser.reader.buffer_position()));
}
loop {
parser.buffer.clear();
match match parser.reader.read_resolved_event_into(&mut parser.buffer)? {
(_, Event::End(_)) => match text {
None => return Err(Error::Format(parser.reader.buffer_position())),
Some(_) => break,
},
(namespace, Event::Start(event)) => {
if match_namespace(namespace) {
match event.local_name().as_ref() {
b"format" => RevisionChildElement::Format,
b"model" => RevisionChildElement::Model,
b"text" => RevisionChildElement::Text,
_ => RevisionChildElement::Unknown,
}
} else {
RevisionChildElement::Unknown
}
}
_ => continue,
} {
RevisionChildElement::Format => {
format = Some(parse_text(parser, &format)?)
}
RevisionChildElement::Model => {
model = Some(parse_text(parser, &model)?)
}
RevisionChildElement::Text => text = Some(parse_text(parser, &text)?),
RevisionChildElement::Unknown => skip_element(parser)?,
}
}
continue;
}
PageChildElement::Title => {
title = Some(parse_text(parser, &title)?);
continue;
}
PageChildElement::Unknown => skip_element(parser)?,
}
}
}
}
/// Creates a parser for a stream.
///
/// The stream is parsed as an XML dump exported from Mediawiki. The parser is an iterator over the pages in the dump.
pub fn parse<R: BufRead>(source: R) -> Parser<R> {
let mut reader = NsReader::from_reader(source);
reader.expand_empty_elements(true);
Parser {
buffer: vec![],
reader,
started: false,
}
}
fn parse_text(
parser: &mut Parser<impl BufRead>,
output: &Option<impl Sized>,
) -> Result<String, Error> {
if output.is_some() {
return Err(Error::Format(parser.reader.buffer_position()));
}
parser.buffer.clear();
let text = match parser
.reader
.read_resolved_event_into(&mut parser.buffer)?
.1
{
Event::Text(text) => text.unescape().unwrap().to_string(),
Event::End { .. } => return Ok(String::new()),
_ => return Err(Error::Format(parser.reader.buffer_position())),
};
parser.buffer.clear();
if let Event::End(_) = parser
.reader
.read_resolved_event_into(&mut parser.buffer)?
.1
{
Ok(text)
} else {
Err(Error::Format(parser.reader.buffer_position()))
}
}
fn skip_element(parser: &mut Parser<impl BufRead>) -> Result<(), quick_xml::Error> {
let mut level = 0;
loop {
parser.buffer.clear();
match parser
.reader
.read_resolved_event_into(&mut parser.buffer)?
.1
{
Event::End(_) => {
if level == 0 {
return Ok(());
}
level -= 1;
}
Event::Start(_) => level += 1,
_ => {}
}
}
}