String Extraction: Parse and Extract Structured Fields from a Raw Log String

Question

## Problem

You receive raw log lines from a server. Each line follows this pattern:

```
[LEVEL] TIMESTAMP key1=val1 key2=val2 ... message="free text here"
```

Write a function that parses a log line and returns a dict with keys: `level`, `timestamp`, `fields` (dict of key-value pairs), and `message`.

```python
def parse_log_line(line: str) -> dict:
    # Returns: {"level": str, "timestamp": str, "fields": dict, "message": str}
    pass
```

**Example:**
```
Input:
  '[ERROR] 2024-03-15T10:22:01Z user_id=42 retries=3 message="connection timed out"'

Output:
  {
    "level":     "ERROR",
    "timestamp": "2024-03-15T10:22:01Z",
    "fields":    {"user_id": "42", "retries": "3"},
    "message":   "connection timed out"
  }
```

## Follow-ups

1. What happens if `message` contains an `=` sign or escaped quotes? How do you handle it robustly?
2. How would you extend this parser to support multi-line log entries (stack traces)?
3. If log volume is 500k lines/second, what would you change to make parsing throughput-optimal in Python?
4. How would you write unit tests to cover malformed lines without crashing the parser?

String Extraction: Parse and Extract Structured Fields from a Raw Log String

Question Details

Full Details

Topics

More from Netflix