Task task8_regex_extraction

You'll find a text file called `mixed_content.txt` in the `./input` directory containing various types of data mixed
together.

Extract ALL instances of the following patterns using regular expressions:

- Email addresses
- Phone numbers (various formats: (123) 456-7890, 123-456-7890, 123.456.7890)
- URLs (http and https)
- Dates (formats: YYYY-MM-DD, MM/DD/YYYY, DD-MM-YYYY)

Generate a JSON file called `extracted_data.json` with the following structure:

```json
{
"emails": [
"[email protected]",
"[email protected]",
...
],
"phone_numbers": [
"(123) 456-7890",
"123-456-7890",
...
],
"urls": [
"https://example.com",
"http://example.org",
...
],
"dates": [
"2024-01-15",
"01/15/2024",
...
]
}
```

Each array should contain the extracted values in the order they appear in the file. Duplicates should be included if they appear multiple times.

PS: You are currently working in an automated system and cannot ask any question or have back and forth with an user.

Results

23
Models Tested
69.6%
Success Rate
2m 33s
Avg Duration
1m 20s - 7m 5s
Duration Range

Details

Score Model Duration Session (KB) test_1_file_exists.sh test_2_valid_json.sh test_3_email_extraction.sh test_4_phone_extraction.sh test_5_url_extraction.sh test_6_date_extraction.sh
100.0% openrouter/openai/gpt-5 3m 39s 361.1
100.0% openrouter/openai/gpt-5-nano 2m 35s 262.3
100.0% openrouter/anthropic/claude-3.5-sonnet 2m 1s 79.3
100.0% openrouter/google/gemini-2.5-pro 1m 55s 20.3
100.0% openrouter/openai/gpt-4o-mini 2m 24s 150.9
100.0% openrouter/google/gemini-2.5-flash-lite-preview-09-2025 1m 27s 22.3
100.0% openrouter/openai/gpt-oss-20b 7m 5s 2819.5
100.0% openrouter/anthropic/claude-3.7-sonnet 3m 21s 228.9
100.0% openrouter/anthropic/claude-haiku-4.5 2m 4s 57.5
100.0% openrouter/deepseek/deepseek-v3.1-terminus 3m 7s 100.4
100.0% litellm/GLM-4.5-Air-FP8-dev 2m 44s 83.1
100.0% openrouter/anthropic/claude-sonnet-4.5 2m 6s 63.8
100.0% openrouter/openai/gpt-5-mini 3m 7s 206.9
100.0% openrouter/anthropic/claude-3.5-haiku 2m 14s 79.0
100.0% openrouter/anthropic/claude-sonnet-4 2m 29s 124.0
100.0% openrouter/openai/gpt-4.1-mini 2m 15s 222.0
83.3% openrouter/anthropic/claude-3-haiku 1m 54s 83.6
83.3% openrouter/qwen/qwen3-coder 2m 21s 31.5
66.7% openrouter/x-ai/grok-3-mini 3m 7s 946.6
66.7% openrouter/openai/gpt-4.1-nano 1m 33s 63.6
0.0% openrouter/google/gemini-2.5-flash-preview-09-2025 1m 34s 15.4
0.0% openrouter/openai/gpt-oss-120b 2m 5s 91.7
0.0% openrouter/deepseek/deepseek-chat-v3-0324 1m 20s 20.9