Reader Examples

from definable.agent import Agent
from definable.media import File
from definable.model import OpenAIChat

agent = Agent(
  model=OpenAIChat(id="gpt-4o-mini"),
  instructions="You are a helpful assistant. Analyze any files the user provides.",
  readers=True,
)

file = File(
  content=b"Q3 Revenue: $2.5M\nQ3 Expenses: $1.8M\nQ3 Net Income: $700K\nGrowth: 15% YoY",
  filename="financials.txt",
  mime_type="text/plain",
)

output = agent.run("Summarize the key financial metrics from this file.", files=[file])
print(output.content)

from typing import List, Set

from definable.media import File
from definable.reader.base import BaseReader
from definable.reader.models import ContentBlock, ReaderConfig
from definable.reader.parsers.base_parser import BaseParser
from definable.reader.registry import ParserRegistry

class MarkdownParser(BaseParser):
  """Custom parser that adds markdown-specific metadata."""

  def supported_mime_types(self) -> List[str]:
    return ["text/markdown"]

  def supported_extensions(self) -> Set[str]:
    return {".md"}

  def parse(self, data: bytes, *, mime_type: str | None = None, config: ReaderConfig | None = None) -> List[ContentBlock]:
    encoding = config.encoding if config else "utf-8"
    text = data.decode(encoding, errors="replace")
    headings = [line for line in text.split("\n") if line.startswith("#")]
    return [
      ContentBlock(
        content_type="text",
        content=text,
        mime_type="text/markdown",
        metadata={"heading_count": len(headings)},
      )
    ]

# Register — higher priority wins over defaults
registry = ParserRegistry()
registry.register(MarkdownParser(), priority=200)
reader = BaseReader(registry=registry)

md_file = File(
  content=b"# Title\n\nSome content.\n\n## Section\n\nMore content.",
  filename="readme.md",
  mime_type="text/markdown",
)
result = reader.read(md_file)
print(f"Content: {result.content[:100]}")

from definable.media import File
from definable.reader import BaseReader

reader = BaseReader()

files = [
  File(content=b"Hello, world!", filename="greeting.txt", mime_type="text/plain"),
  File(content=b'{"name": "Alice", "age": 30}', filename="user.json", mime_type="application/json"),
  File(
    content=b"name,score\nAlice,95\nBob,87\nCharlie,92",
    filename="scores.csv",
    mime_type="text/csv",
  ),
]

for file in files:
  result = reader.read(file)
  print(f"{result.filename}: {result.content[:100]}")

# Check which parser handles a file
parser = reader.get_parser(files[0])
print(f"Parser for .txt: {type(parser).__name__}")

from typing import List, Set

from definable.agent import Agent
from definable.agent.testing import MockModel
from definable.media import File
from definable.reader.base import BaseReader
from definable.reader.models import ContentBlock, ReaderConfig
from definable.reader.parsers.base_parser import BaseParser
from definable.reader.registry import ParserRegistry

class CloudPDFParser(BaseParser):
  """Simulated cloud-based PDF parser."""

  def supported_mime_types(self) -> List[str]:
    return ["application/pdf"]

  def supported_extensions(self) -> Set[str]:
    return {".pdf"}

  def parse(self, data: bytes, *, mime_type: str | None = None, config: ReaderConfig | None = None) -> List[ContentBlock]:
    return [ContentBlock(content_type="text", content="[CloudPDF] Extracted text with high-quality OCR")]

# Default reader
default_reader = BaseReader()
pdf_file = File(content=b"%PDF", filename="report.pdf", mime_type="application/pdf")
parser = default_reader.get_parser(pdf_file)
print(f"Default PDF parser: {type(parser).__name__}")

# Override with cloud provider
registry = ParserRegistry()
registry.register(CloudPDFParser(), priority=200)
custom_reader = BaseReader(registry=registry)
parser = custom_reader.get_parser(pdf_file)
print(f"Custom PDF parser: {type(parser).__name__}")

# Use with Agent
agent = Agent(model=MockModel(responses=["Analyzed."]), readers=custom_reader)
output = agent.run("Analyze this PDF.", files=[pdf_file])