docwire::content_type Namespace Reference

Provides a multi-stage pipeline for content type detection. More...

Namespaces
	by_file_extension
	by_signature
	Provides content type detection based on file signatures (magic bytes).

Classes
class	detector
	Content type detection chain element. More...

Functions
DOCWIRE_CONTENT_TYPE_EXPORT void	detect (data_source &data, const by_signature::database &signatures_db_to_use=by_signature::database{})

Detailed Description

Provides a multi-stage pipeline for content type detection.

Architecture

The content type detection in DocWire uses a multi-stage pipeline to ensure accuracy and performance:

1. By File Extension: Fast lookup using a comprehensive dictionary. Note that we intentionally keep multiple MIME type aliases for a single extension (e.g., .xml maps to both text/xml and application/xml) so users can query any valid historical variant.
2. By Signature (libmagic): Reads magic bytes.
3. Heuristic Fallbacks: Custom detectors (e.g., HTML, OOXML, Images) that correct limitations in signature detection.

Stream Processing & Performance

When processing non-seekable streams (like network sockets), signature detection cannot seek to the end of the file. This causes ZIP-based formats (DOCX, ODT) to be detected generically as application/zip. To fix this, heuristic detectors are used. Rule: Heuristic detectors must prioritize performance by reading only a small initial buffer (e.g., 4KB) to check for local file headers before falling back to deep inspection (like ZIP parsing), preventing massive files from being downloaded into memory.

Function Documentation

detect()

DOCWIRE_CONTENT_TYPE_EXPORT void docwire::content_type::detect	(	data_source &	data,
		const by_signature::database &	signatures_db_to_use = by_signature::database{}
	)

Detects and assigns content types to the provided data source using various detection strategies.

This function attempts to identify the content type of the data by using the following detection methods:

• By file extension
• By file signature
• Image content detection
• ODF and OOXML format detection
• ASP content detection
• HTML content detection
• iWork content detection
• ODF Flat format detection
• Outlook format detection
• XLSB format detection

Parameters

data	The data source to be analyzed for content type detection.
signatures_db_to_use	The loaded database of signatures used for signature-based content detection. It will be created (and loaded) if not provided.

See also

performing file type detection example

content_type::detector

content_type::by_signature::database

content_type::by_file_extension::detect

content_type::by_signature::detect

content_type::image::detect

content_type::odf_ooxml::detect

content_type::asp::detect

content_type::html::detect

content_type::iwork::detect

content_type::odf_flat::detect

content_type::outlook::detect

content_type::xlsb::detect

Examples

file_type_determination.cpp.