LLM4APR commited on
Commit
5c9687b
·
verified ·
1 Parent(s): b305580

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +74 -1
README.md CHANGED
@@ -4,4 +4,77 @@ tags:
4
  - web
5
  - accessibility
6
  - repair
7
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  - web
5
  - accessibility
6
  - repair
7
+ ---
8
+
9
+ # A11YBench
10
+
11
+ ## 😃Dataset Summary
12
+ A11YBench consists of 60 real-world web projects, encompassing 145 web pages and 8,874 accessibility violations detected by the IBM Accessibility Checker (4.0.9).
13
+ The projects vary substantially in size, from 123 to 43,198 source files and from 3,610 to 1,555,532 lines of code, covering both lightweight documentation sites and large production-grade applications.
14
+ This scale ensures that A11YBench reflects the structural and technical diversity of modern web ecosystems.
15
+
16
+ ## 🚀Dataset Features
17
+
18
+ Real-world accessibility violations extracted from popular open-source web repositories.
19
+ Multi-domain coverage including UI libraries, developer tools, cloud platforms, documentation sites, and more.
20
+
21
+ Two configurations:
22
+ A11YBench-Lite – small, quick-to-run subset (10 repos).
23
+ A11YBench-Full – complete benchmark with all collected repositories.
24
+
25
+ ## 📃How to use it?
26
+
27
+ ### 📥 Downloading the Dataset
28
+
29
+ You can download via HuggingFace Hub:
30
+ ```bash
31
+ git lfs install
32
+ git clone https://huggingface.co/datasets/LLM4APR/A11YBench
33
+ ```
34
+ Or load directly in Python:
35
+ ```bash
36
+ from datasets import load_dataset
37
+ ds = load_dataset("LLM4APR/A11YBench")
38
+ ```
39
+ ### 🧩 Installing Project Instances
40
+ Before running any repair task, you must locally checkout and install the dependencies for each project included in the benchmark.
41
+ This setup requires a one-time manual installation and is typically fast and straightforward.
42
+
43
+ Make sure you have the following installed:
44
+
45
+ · Node.js
46
+ · npm
47
+ · pnpm
48
+ · yarn
49
+
50
+ Below is an example setup process using the repository ***microsoft/TypeScript-Website***.
51
+
52
+ 1. Clone the repository
53
+ '''bash
54
+ git clone https://github.com/microsoft/TypeScript-Website.git
55
+ '''
56
+ 2. Checkout the version used in A11YBench
57
+ '''bash
58
+ git checkout 1019cfc8f96ded4729e8ccd342742534a9826a9d
59
+ '''
60
+ 3. Install dependencies and start the project
61
+ '''bash
62
+ pnpm install && pnpm bootstrap && pnpm start
63
+ '''
64
+ Once the installation finishes, the project will be served locally.
65
+
66
+ ### 🚀 Launching the Repair Workflow
67
+
68
+ With the project successfully running locally, you can open the webpage in your browser:
69
+
70
+ 👉 http://localhost:8000/
71
+
72
+ You should see a page similar to this:
73
+ ![image](https://cdn-uploads.huggingface.co/production/uploads/64648f5802f8858f2128d179/V2nYM--5mnLgSUvVo1u28.png)
74
+
75
+ ### 🧪 Generating Accessibility Violations
76
+
77
+ Next, use any Web Accessibility Checker to detect A11Y violations on the active page.
78
+ We recommend: ***IBM Accessibility Checker***
79
+
80
+ These tools will generate a Violation Report, which can be passed to your repair system to perform automated A11Y fixing.