Custom Parsers & Extension

tickertape-client uses a modular parser registry design. You can register custom scrapers and response parsers without modifying the core client codebase.

---

The BaseParser Interface

All parsers inherit from BaseParser[T] and implement the parse(content) method:

from typing import Any
from tickertape.parsers import BaseParser
from tickertape.errors import TickerTapeParseError

class MyCustomParser(BaseParser[dict]):
    """Extract custom information from HTML or JSON."""

    def parse(self, content: Any) -> dict:
        if not isinstance(content, dict):
            raise TickerTapeParseError(f"Expected dict, got {type(content).__name__}")

        return {
            "processed": True,
            "data": content,
        }

---

Registering and Retrieving Parsers

Use register_parser() to add your parser to the global registry, and get_parser() to instantiate it by name:

from tickertape.parsers import register_parser, get_parser

# 1. Register under a unique identifier
register_parser("my_custom_type", MyCustomParser)

# 2. Instantiate from registry
parser = get_parser("my_custom_type")

# 3. Execute
result = parser.parse({"key": "value"})
print(result)

---

Built-In Default Parsers

The registry includes the following built-in parsers:

Registry Key	Parser Class	Status
"mf", "mutualfunds"	MFParser	Full implementation (Next.js SSR)
"sitemap"	SitemapParser	Full implementation (XML urlset/sitemapindex)
"stocks"	StockParser	Stub (extensible)
"etf"	ETFParser	Stub (extensible)
"screens"	ScreenParser	Stub (extensible)