| 1 |
use proptest::prelude::*; |
| 2 |
|
| 3 |
|
| 4 |
fn html_fragment() -> impl Strategy<Value = String> { |
| 5 |
let tags = prop::sample::select(vec![ |
| 6 |
"p", |
| 7 |
"div", |
| 8 |
"span", |
| 9 |
"strong", |
| 10 |
"em", |
| 11 |
"a", |
| 12 |
"h1", |
| 13 |
"h2", |
| 14 |
"h3", |
| 15 |
"ul", |
| 16 |
"ol", |
| 17 |
"li", |
| 18 |
"blockquote", |
| 19 |
"pre", |
| 20 |
"code", |
| 21 |
"br", |
| 22 |
"hr", |
| 23 |
"img", |
| 24 |
"table", |
| 25 |
"tr", |
| 26 |
"td", |
| 27 |
"th", |
| 28 |
"b", |
| 29 |
"i", |
| 30 |
"del", |
| 31 |
"sup", |
| 32 |
"sub", |
| 33 |
]); |
| 34 |
|
| 35 |
let text = "[a-zA-Z0-9 .,!?]{0,100}"; |
| 36 |
|
| 37 |
prop::collection::vec( |
| 38 |
prop_oneof![ |
| 39 |
|
| 40 |
text.prop_map(|s| s), |
| 41 |
|
| 42 |
(tags.clone(), text).prop_map(|(tag, content)| { format!("<{tag}>{content}</{tag}>") }), |
| 43 |
|
| 44 |
tags.clone().prop_map(|tag| format!("<{tag}/>")), |
| 45 |
|
| 46 |
(tags.clone(), tags.clone(), text).prop_map(|(outer, inner, content)| { |
| 47 |
format!("<{outer}><{inner}>{content}</{inner}></{outer}>") |
| 48 |
}), |
| 49 |
], |
| 50 |
1..10, |
| 51 |
) |
| 52 |
.prop_map(|parts| parts.join("")) |
| 53 |
} |
| 54 |
|
| 55 |
proptest! { |
| 56 |
#[test] |
| 57 |
fn never_panics(html in html_fragment()) { |
| 58 |
let _ = pter::convert(&html); |
| 59 |
} |
| 60 |
|
| 61 |
#[test] |
| 62 |
fn never_panics_on_arbitrary_bytes(s in "\\PC{0,500}") { |
| 63 |
let _ = pter::convert(&s); |
| 64 |
} |
| 65 |
|
| 66 |
#[test] |
| 67 |
fn output_contains_no_html_tags(html in html_fragment()) { |
| 68 |
let md = pter::convert(&html); |
| 69 |
|
| 70 |
|
| 71 |
let without_code_blocks: String = md |
| 72 |
.split("```") |
| 73 |
.enumerate() |
| 74 |
.filter(|(i, _)| i % 2 == 0) |
| 75 |
.map(|(_, s)| s) |
| 76 |
.collect(); |
| 77 |
|
| 78 |
|
| 79 |
assert!(!without_code_blocks.contains("<script"), "leaked <script> in: {md}"); |
| 80 |
assert!(!without_code_blocks.contains("<style"), "leaked <style> in: {md}"); |
| 81 |
assert!(!without_code_blocks.contains("<head"), "leaked <head> in: {md}"); |
| 82 |
} |
| 83 |
|
| 84 |
#[test] |
| 85 |
fn output_is_valid_utf8(html in html_fragment()) { |
| 86 |
let md = pter::convert(&html); |
| 87 |
|
| 88 |
|
| 89 |
assert!(!md.contains('\u{FFFD}'), "replacement char in: {md}"); |
| 90 |
} |
| 91 |
|
| 92 |
#[test] |
| 93 |
fn no_trailing_whitespace_on_lines(html in html_fragment()) { |
| 94 |
let md = pter::convert(&html); |
| 95 |
for (i, line) in md.lines().enumerate() { |
| 96 |
assert!( |
| 97 |
line == line.trim_end(), |
| 98 |
"trailing whitespace on line {i}: '{line}'" |
| 99 |
); |
| 100 |
} |
| 101 |
} |
| 102 |
|
| 103 |
#[test] |
| 104 |
fn empty_input_returns_empty(s in "\\s{0,20}") { |
| 105 |
let html = format!("<html><body>{s}</body></html>"); |
| 106 |
let md = pter::convert(&html); |
| 107 |
|
| 108 |
assert!(md.trim().is_empty() || !s.trim().is_empty()); |
| 109 |
} |
| 110 |
} |
| 111 |
|
| 112 |
|
| 113 |
|
| 114 |
|
| 115 |
|
| 116 |
proptest! { |
| 117 |
#![proptest_config(ProptestConfig::with_cases(100_000))] |
| 118 |
|
| 119 |
#[test] |
| 120 |
fn no_excessive_blank_lines(html in html_fragment()) { |
| 121 |
let md = pter::convert(&html); |
| 122 |
assert!(!md.contains("\n\n\n"), "triple newline in output: {md}"); |
| 123 |
} |
| 124 |
} |
| 125 |
|