organicmaps/wikiparser
13
Fork 0
Code Issues 13 Pull requests 2 Projects Releases Wiki Activity Actions

Add additional checks for langs/titles

Browse source 
Signed-off-by: Evan Lloyd New-Schmidt <evan@new-schmidt.com>
This commit is contained in:
Evan Lloyd New-Schmidt 2023-08-25 11:23:44 -04:00 • committed by Evan Lloyd New-Schmidt
parent 292eeac081
commit a584498c65
1 changed files with 16 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

18
src/wm/title.rs
View file

@ -106,15 +106,25 @@ impl Title {
pub fn from_title(title: &str, lang: &str) -> Result<Self, ParseTitleError> {
let title = title.trim();
let lang = lang.trim();
if title.is_empty() {
return Err(ParseTitleError::NoTitle);
}
// Wikipedia titles must be less than 256 bytes of UTF-8.
// See: https://en.wikipedia.org/wiki/Wikipedia:Naming_conventions_(technical_restrictions)#Title_length
if !title.len() < 256 {
return Err(ParseTitleError::TitleLong);
}
let lang = lang.trim();
if lang.is_empty() {
return Err(ParseTitleError::NoLang);
}
if lang.contains(|c: char| !(c.is_ascii_alphabetic() || c == '-')) {
return Err(ParseTitleError::LangBadChar);
}
let lang = lang.to_ascii_lowercase();
let name = Self::normalize_title(title);
let lang = lang.to_owned();
Ok(Self { name, lang })
}
@ -135,8 +145,12 @@ pub enum ParseTitleError {
Empty,
#[error("title is empty or whitespace")]
NoTitle,
#[error("title is too long")]
TitleLong,
#[error("lang is empty or whitespace")]
NoLang,
#[error("lang contains character that is not alphabetic or '-'")]
LangBadChar,
#[error("no ':' separating lang and title")]
MissingColon,