Markdown Tables

Basic Usage

from pydantic import BaseModel
from md2pydantic import MDConverter

class User(BaseModel):
    name: str
    age: int
    active: bool

markdown = """
| name  | age | active |
|-------|-----|--------|
| Alice | 30  | Yes    |
| Bob   | 25  | No     |
"""

users = MDConverter(User).parse_tables(markdown)
# [User(name='Alice', age=30, active=True),
#  User(name='Bob', age=25, active=False)]

Note

parse_tables always returns a list of model instances, even if the table has only one row.

Header-to-Field Mapping

Table headers are mapped to model fields by exact name match. The mapping is case-sensitive: a header Name will not match a field named name.

| name | age |    <-- must match field names exactly
|------|-----|
| ...  | ... |

Bool Coercion

md2pydantic maps common boolean words to True/False before passing data to Pydantic. The mapping is case-insensitive:

Input values	Result
Yes, Y, true, on, 1	True
No, N, false, off, 0	False

This only applies to fields typed as bool.

Whitespace Handling

Leading and trailing whitespace in cell values is stripped automatically. This means | Alice | and |Alice| both produce "Alice".

Surrounding Prose

md2pydantic ignores prose before and after the table. LLM output like this works fine:

Here are the results I found:

| name  | age |
|-------|-----|
| Alice | 30  |

Let me know if you need anything else!