Migrate book_creates_reasonable_search_index to BookTest
This commit is contained in:
Eric Huss
parent
3fce1151dd
commit
cad8988f8d
12 changed files with 167 additions and 61 deletions
|
|
@ -393,67 +393,6 @@ mod search {
|
||||||
serde_json::from_str(&index.replace("\\'", "'").replace("\\\\", "\\")).unwrap()
|
serde_json::from_str(&index.replace("\\'", "'").replace("\\\\", "\\")).unwrap()
|
||||||
}
|
}
|
||||||
|
|
||||||
#[test]
|
|
||||||
fn book_creates_reasonable_search_index() {
|
|
||||||
let temp = DummyBook::new().build().unwrap();
|
|
||||||
let md = MDBook::load(temp.path()).unwrap();
|
|
||||||
md.build().unwrap();
|
|
||||||
|
|
||||||
let index = read_book_index(temp.path());
|
|
||||||
|
|
||||||
let doc_urls = index["doc_urls"].as_array().unwrap();
|
|
||||||
eprintln!("doc_urls={doc_urls:#?}",);
|
|
||||||
let get_doc_ref =
|
|
||||||
|url: &str| -> String { doc_urls.iter().position(|s| s == url).unwrap().to_string() };
|
|
||||||
|
|
||||||
let first_chapter = get_doc_ref("first/index.html#first-chapter");
|
|
||||||
let introduction = get_doc_ref("intro.html#introduction");
|
|
||||||
let some_section = get_doc_ref("first/index.html#some-section");
|
|
||||||
let summary = get_doc_ref("first/includes.html#summary");
|
|
||||||
let no_headers = get_doc_ref("first/no-headers.html");
|
|
||||||
let duplicate_headers_1 = get_doc_ref("first/duplicate-headers.html#header-text-1");
|
|
||||||
let conclusion = get_doc_ref("conclusion.html#conclusion");
|
|
||||||
let heading_attrs = get_doc_ref("first/heading-attributes.html#both");
|
|
||||||
|
|
||||||
let bodyidx = &index["index"]["index"]["body"]["root"];
|
|
||||||
let textidx = &bodyidx["t"]["e"]["x"]["t"];
|
|
||||||
assert_eq!(textidx["df"], 5);
|
|
||||||
assert_eq!(textidx["docs"][&first_chapter]["tf"], 1.0);
|
|
||||||
assert_eq!(textidx["docs"][&introduction]["tf"], 1.0);
|
|
||||||
|
|
||||||
let docs = &index["index"]["documentStore"]["docs"];
|
|
||||||
assert_eq!(docs[&first_chapter]["body"], "more text.");
|
|
||||||
assert_eq!(docs[&some_section]["body"], "");
|
|
||||||
assert_eq!(
|
|
||||||
docs[&summary]["body"],
|
|
||||||
"Dummy Book Introduction First Chapter Nested Chapter Includes Recursive Markdown Unicode No Headers Duplicate Headers Heading Attributes Second Chapter Nested Chapter Conclusion"
|
|
||||||
);
|
|
||||||
assert_eq!(
|
|
||||||
docs[&summary]["breadcrumbs"],
|
|
||||||
"First Chapter » Includes » Summary"
|
|
||||||
);
|
|
||||||
// See note about InlineHtml in search.rs. Ideally the `alert()` part
|
|
||||||
// should not be in the index, but we don't have a way to scrub inline
|
|
||||||
// html.
|
|
||||||
assert_eq!(docs[&conclusion]["body"], "I put <HTML> in here! Sneaky inline event alert(\"inline\");. But regular inline is indexed.");
|
|
||||||
assert_eq!(
|
|
||||||
docs[&no_headers]["breadcrumbs"],
|
|
||||||
"First Chapter » No Headers"
|
|
||||||
);
|
|
||||||
assert_eq!(
|
|
||||||
docs[&duplicate_headers_1]["breadcrumbs"],
|
|
||||||
"First Chapter » Duplicate Headers » Header Text"
|
|
||||||
);
|
|
||||||
assert_eq!(
|
|
||||||
docs[&no_headers]["body"],
|
|
||||||
"Capybara capybara capybara. Capybara capybara capybara. ThisLongWordIsIncludedSoWeCanCheckThatSufficientlyLongWordsAreOmittedFromTheSearchIndex."
|
|
||||||
);
|
|
||||||
assert_eq!(
|
|
||||||
docs[&heading_attrs]["breadcrumbs"],
|
|
||||||
"First Chapter » Heading Attributes » Heading with id and classes"
|
|
||||||
);
|
|
||||||
}
|
|
||||||
|
|
||||||
#[test]
|
#[test]
|
||||||
fn can_disable_individual_chapters() {
|
fn can_disable_individual_chapters() {
|
||||||
let temp = DummyBook::new().build().unwrap();
|
let temp = DummyBook::new().build().unwrap();
|
||||||
|
|
|
||||||
|
|
@ -15,6 +15,8 @@ mod print;
|
||||||
mod redirects;
|
mod redirects;
|
||||||
mod renderer;
|
mod renderer;
|
||||||
mod rendering;
|
mod rendering;
|
||||||
|
#[cfg(feature = "search")]
|
||||||
|
mod search;
|
||||||
|
|
||||||
mod prelude {
|
mod prelude {
|
||||||
pub use crate::book_test::BookTest;
|
pub use crate::book_test::BookTest;
|
||||||
|
|
|
||||||
78
tests/testsuite/search.rs
Normal file
78
tests/testsuite/search.rs
Normal file
|
|
@ -0,0 +1,78 @@
|
||||||
|
//! Tests for search support.
|
||||||
|
|
||||||
|
use crate::prelude::*;
|
||||||
|
use std::path::Path;
|
||||||
|
|
||||||
|
fn read_book_index(root: &Path) -> serde_json::Value {
|
||||||
|
let index = root.join("book/searchindex.js");
|
||||||
|
let index = std::fs::read_to_string(index).unwrap();
|
||||||
|
let index = index.trim_start_matches("window.search = JSON.parse('");
|
||||||
|
let index = index.trim_end_matches("');");
|
||||||
|
// We need unescape the string as it's supposed to be an escaped JS string.
|
||||||
|
serde_json::from_str(&index.replace("\\'", "'").replace("\\\\", "\\")).unwrap()
|
||||||
|
}
|
||||||
|
|
||||||
|
// Some spot checks for the generation of the search index.
|
||||||
|
#[test]
|
||||||
|
fn reasonable_search_index() {
|
||||||
|
let mut test = BookTest::from_dir("search/reasonable_search_index");
|
||||||
|
test.build();
|
||||||
|
let index = read_book_index(&test.dir);
|
||||||
|
|
||||||
|
let doc_urls = index["doc_urls"].as_array().unwrap();
|
||||||
|
eprintln!("doc_urls={doc_urls:#?}",);
|
||||||
|
let get_doc_ref = |url: &str| -> String {
|
||||||
|
doc_urls
|
||||||
|
.iter()
|
||||||
|
.position(|s| s == url)
|
||||||
|
.unwrap_or_else(|| panic!("failed to find {url}"))
|
||||||
|
.to_string()
|
||||||
|
};
|
||||||
|
|
||||||
|
let first_chapter = get_doc_ref("first/index.html#first-chapter");
|
||||||
|
let introduction = get_doc_ref("intro.html#introduction");
|
||||||
|
let some_section = get_doc_ref("first/index.html#some-section");
|
||||||
|
let summary = get_doc_ref("first/includes.html#summary");
|
||||||
|
let no_headers = get_doc_ref("first/no-headers.html");
|
||||||
|
let duplicate_headers_1 = get_doc_ref("first/duplicate-headers.html#header-text-1");
|
||||||
|
let heading_attrs = get_doc_ref("first/heading-attributes.html#both");
|
||||||
|
let sneaky = get_doc_ref("intro.html#sneaky");
|
||||||
|
|
||||||
|
let bodyidx = &index["index"]["index"]["body"]["root"];
|
||||||
|
let textidx = &bodyidx["t"]["e"]["x"]["t"];
|
||||||
|
assert_eq!(textidx["df"], 5);
|
||||||
|
assert_eq!(textidx["docs"][&first_chapter]["tf"], 1.0);
|
||||||
|
assert_eq!(textidx["docs"][&introduction]["tf"], 1.0);
|
||||||
|
|
||||||
|
let docs = &index["index"]["documentStore"]["docs"];
|
||||||
|
assert_eq!(docs[&first_chapter]["body"], "more text.");
|
||||||
|
assert_eq!(docs[&some_section]["body"], "");
|
||||||
|
assert_eq!(
|
||||||
|
docs[&summary]["body"],
|
||||||
|
"Introduction First Chapter Includes Unicode No Headers Duplicate Headers Heading Attributes"
|
||||||
|
);
|
||||||
|
assert_eq!(
|
||||||
|
docs[&summary]["breadcrumbs"],
|
||||||
|
"First Chapter » Includes » Summary"
|
||||||
|
);
|
||||||
|
// See note about InlineHtml in search.rs. Ideally the `alert()` part
|
||||||
|
// should not be in the index, but we don't have a way to scrub inline
|
||||||
|
// html.
|
||||||
|
assert_eq!(docs[&sneaky]["body"], "I put <HTML> in here! Sneaky inline event alert(\"inline\");. But regular inline is indexed.");
|
||||||
|
assert_eq!(
|
||||||
|
docs[&no_headers]["breadcrumbs"],
|
||||||
|
"First Chapter » No Headers"
|
||||||
|
);
|
||||||
|
assert_eq!(
|
||||||
|
docs[&duplicate_headers_1]["breadcrumbs"],
|
||||||
|
"First Chapter » Duplicate Headers » Header Text"
|
||||||
|
);
|
||||||
|
assert_eq!(
|
||||||
|
docs[&no_headers]["body"],
|
||||||
|
"Capybara capybara capybara. Capybara capybara capybara. ThisLongWordIsIncludedSoWeCanCheckThatSufficientlyLongWordsAreOmittedFromTheSearchIndex."
|
||||||
|
);
|
||||||
|
assert_eq!(
|
||||||
|
docs[&heading_attrs]["breadcrumbs"],
|
||||||
|
"First Chapter » Heading Attributes » Heading with id and classes"
|
||||||
|
);
|
||||||
|
}
|
||||||
File diff suppressed because one or more lines are too long
|
|
@ -0,0 +1,10 @@
|
||||||
|
# Summary
|
||||||
|
|
||||||
|
[Introduction](intro.md)
|
||||||
|
|
||||||
|
- [First Chapter](first/index.md)
|
||||||
|
- [Includes](first/includes.md)
|
||||||
|
- [Unicode](first/unicode.md)
|
||||||
|
- [No Headers](first/no-headers.md)
|
||||||
|
- [Duplicate Headers](first/duplicate-headers.md)
|
||||||
|
- [Heading Attributes](first/heading-attributes.md)
|
||||||
|
|
@ -0,0 +1,9 @@
|
||||||
|
# Duplicate headers
|
||||||
|
|
||||||
|
This page validates behaviour of duplicate headers.
|
||||||
|
|
||||||
|
# Header Text
|
||||||
|
|
||||||
|
# Header Text
|
||||||
|
|
||||||
|
# header-text
|
||||||
|
|
@ -0,0 +1,5 @@
|
||||||
|
# Heading Attributes {#attrs}
|
||||||
|
|
||||||
|
## Heading with classes {.class1 .class2}
|
||||||
|
|
||||||
|
## Heading with id and classes {#both .class1 .class2}
|
||||||
|
|
@ -0,0 +1,3 @@
|
||||||
|
# Includes
|
||||||
|
|
||||||
|
{{#include ../SUMMARY.md::}}
|
||||||
|
|
@ -0,0 +1,5 @@
|
||||||
|
# First Chapter
|
||||||
|
|
||||||
|
more text.
|
||||||
|
|
||||||
|
## Some Section
|
||||||
|
|
@ -0,0 +1,5 @@
|
||||||
|
Capybara capybara capybara.
|
||||||
|
|
||||||
|
Capybara capybara capybara.
|
||||||
|
|
||||||
|
ThisLongWordIsIncludedSoWeCanCheckThatSufficientlyLongWordsAreOmittedFromTheSearchIndex.
|
||||||
|
|
@ -0,0 +1,21 @@
|
||||||
|
# Unicode stress tests
|
||||||
|
|
||||||
|
Please be careful editing, this contains carefully crafted characters.
|
||||||
|
|
||||||
|
Two byte character: spatiëring
|
||||||
|
|
||||||
|
Combining character: spatiëring
|
||||||
|
|
||||||
|
Three byte character: 书こんにちは
|
||||||
|
|
||||||
|
Four byte character: 𐌀𐌁𐌂𐌃𐌄𐌅𐌆𐌇𐌈
|
||||||
|
|
||||||
|
Right-to-left: مرحبا
|
||||||
|
|
||||||
|
Emoticons: 🔊 😍 💜 1️⃣
|
||||||
|
|
||||||
|
right-to-left mark: hello באמת!
|
||||||
|
|
||||||
|
|
||||||
|
Zalgo: ǫ̛̖̱̗̝͈̋͒͋̏ͥͫ̒̆ͩ̏͌̾͊͐ͪ̾̚
|
||||||
|
|
||||||
28
tests/testsuite/search/reasonable_search_index/src/intro.md
Normal file
28
tests/testsuite/search/reasonable_search_index/src/intro.md
Normal file
|
|
@ -0,0 +1,28 @@
|
||||||
|
# Introduction
|
||||||
|
|
||||||
|
Here's some interesting text...
|
||||||
|
|
||||||
|
## Sneaky
|
||||||
|
|
||||||
|
<p>
|
||||||
|
<!--secret secret-->
|
||||||
|
I put <HTML> in here!<br/>
|
||||||
|
</p>
|
||||||
|
<script type="text/javascript" >
|
||||||
|
// I probably shouldn't do this
|
||||||
|
if (3 < 5 > 10)
|
||||||
|
{
|
||||||
|
alert("The sky is falling!");
|
||||||
|
}
|
||||||
|
</script >
|
||||||
|
<style >
|
||||||
|
/*
|
||||||
|
css looks, like this {
|
||||||
|
foo: < 3 <bar >
|
||||||
|
}
|
||||||
|
*/
|
||||||
|
</style>
|
||||||
|
|
||||||
|
Sneaky inline event <script>alert("inline");</script>.
|
||||||
|
|
||||||
|
But regular <b>inline</b> is indexed.
|
||||||
Loading…
Add table
Reference in a new issue