Marvin Zhang commited on
Commit ·
67a3e42
1
Parent(s): ec5df57
initial commit
Browse files
README.md
CHANGED
|
@@ -1,3 +1,54 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
---
|
| 4 |
+
# CrawlEval
|
| 5 |
+
|
| 6 |
+
Resources and tools for evaluating the performance and behavior of web crawling systems.
|
| 7 |
+
|
| 8 |
+
## Overview
|
| 9 |
+
|
| 10 |
+
CrawlEval provides a comprehensive suite of tools and datasets for evaluating web crawling systems, with a particular focus on HTML pattern extraction and content analysis. The project includes:
|
| 11 |
+
|
| 12 |
+
1. A curated dataset of web pages with ground truth patterns
|
| 13 |
+
2. Tools for fetching and analyzing web content
|
| 14 |
+
3. Evaluation metrics and benchmarking capabilities
|
| 15 |
+
|
| 16 |
+
## Dataset
|
| 17 |
+
|
| 18 |
+
The dataset is designed to test and benchmark web crawling systems' ability to extract structured data from HTML. It includes:
|
| 19 |
+
|
| 20 |
+
- Raw HTML files with various structures and complexities
|
| 21 |
+
- Ground truth PagePattern JSON files
|
| 22 |
+
- Metadata about each example (query, complexity, etc.)
|
| 23 |
+
|
| 24 |
+
See the [dataset documentation](crawleval/README.md) for detailed information about the dataset structure and usage.
|
| 25 |
+
|
| 26 |
+
## Tools
|
| 27 |
+
|
| 28 |
+
### Web Page Fetcher (`fetch_webpage.py`)
|
| 29 |
+
|
| 30 |
+
A powerful tool for collecting and analyzing web pages for evaluation purposes.
|
| 31 |
+
|
| 32 |
+
Key features:
|
| 33 |
+
- Fetches web pages with proper JavaScript rendering using Selenium
|
| 34 |
+
- Extracts and analyzes metadata (DOM structure, nesting levels, etc.)
|
| 35 |
+
- Content deduplication using SHA-256 hashing
|
| 36 |
+
- URL deduplication with normalization
|
| 37 |
+
- Parallel processing of multiple URLs
|
| 38 |
+
- Progress tracking and detailed logging
|
| 39 |
+
|
| 40 |
+
Usage:
|
| 41 |
+
```bash
|
| 42 |
+
python -m crawleval.fetch_webpage --batch urls.txt [options]
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
Options:
|
| 46 |
+
- `--dir DIR`: Base directory for storing data
|
| 47 |
+
- `--list-hashes`: Display the content hash index
|
| 48 |
+
- `--list-urls`: Display the URL index
|
| 49 |
+
- `--save-results FILE`: Save batch processing results to a JSON file
|
| 50 |
+
- `--workers N`: Number of parallel workers (default: 4)
|
| 51 |
+
|
| 52 |
+
## Contributing
|
| 53 |
+
|
| 54 |
+
We welcome contributions to improve the dataset and tools. Please see the [dataset documentation](crawleval/README.md) for guidelines on adding new examples.
|