001. Python API Specification¶
Overview¶
This document specifies the Python API implementing context-aware text detection with pluggable backend support, confidence-based detection, and optional dependency architecture.
The design follows established project practices for interface contracts, module organization, naming conventions, and provides both simple string-based APIs and confidence-aware APIs with structured result types.
Public Interface Specification¶
Core Type Definitions¶
Confidence-Based Result Types
class CharsetResult( __.immut.DataclassObject ):
''' Character set encoding with detection confidence. '''
charset: __.typx.Annotated[
__.typx.Optional[ str ],
__.ddoc.Doc( ''' Detected character set encoding. May be None. ''' ),
]
confidence: __.typx.Annotated[
float, __.ddoc.Doc( ''' Detection confidence from 0.0 to 1.0. ''' )
]
class MimetypeResult( __.immut.DataclassObject ):
''' MIME type with detection confidence. '''
mimetype: __.typx.Annotated[
str, __.ddoc.Doc( ''' Detected MIME type. ''' )
]
confidence: __.typx.Annotated[
float, __.ddoc.Doc( ''' Detection confidence from 0.0 to 1.0. ''' )
]
Configuration Types
class BehaviorTristate( __.enum.Enum ):
''' When to apply behavior. '''
Never = __.enum.auto( )
AsNeeded = __.enum.auto( )
Always = __.enum.auto( )
class DetectFailureActions( __.enum.Enum ):
''' Possible responses to detection failure. '''
Default = __.enum.auto( )
Error = __.enum.auto( )
class CodecSpecifiers( __.enum.Enum ):
''' Specifiers for dynamic codecs. '''
FromInference = __.enum.auto( )
OsDefault = __.enum.auto( )
PythonDefault = __.enum.auto( )
UserSupplement = __.enum.auto( )
class Behaviors( __.immut.DataclassObject ):
''' How functions behave. '''
charset_detectors_order: __.typx.Annotated[
__.cabc.Sequence[ str ],
__.ddoc.Doc( ''' Order in which charset detectors are applied. ''' ),
] = ( 'chardet', 'charset-normalizer' )
charset_on_detect_failure: __.typx.Annotated[
DetectFailureActions,
__.ddoc.Doc( ''' Action to take on charset detection failure. ''' ),
] = DetectFailureActions.Default
mimetype_detectors_order: __.typx.Annotated[
__.cabc.Sequence[ str ],
__.ddoc.Doc( ''' Order in which MIME type detectors are applied. ''' ),
] = ( 'magic', 'puremagic' )
mimetype_on_detect_failure: __.typx.Annotated[
DetectFailureActions,
__.ddoc.Doc( ''' Action to take on MIME type detection failure. ''' ),
] = DetectFailureActions.Default
charset_detect: __.typx.Annotated[
BehaviorTristate,
__.ddoc.Doc( ''' When to detect charset from content. ''' ),
] = BehaviorTristate.AsNeeded
mimetype_detect: __.typx.Annotated[
BehaviorTristate,
__.ddoc.Doc( ''' When to detect MIME type from content. ''' ),
] = BehaviorTristate.AsNeeded
Simple String-Based Detection Functions¶
Character Encoding Detection
def detect_charset(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
default: str = CHARSET_DEFAULT,
supplement: __.Absential[ str ] = __.absent,
mimetype: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
) -> __.typx.Optional[ str ]:
''' Detects character encoding.
Returns the most likely character encoding. When configured for
default return behavior, returns the default value on detection
failure rather than raising an exception.
'''
def detect_mimetype(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
default: str = MIMETYPE_DEFAULT,
charset: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
) -> str:
''' Detects MIME type.
Returns the most likely MIME type. When configured for default
return behavior, returns the default value on detection failure
rather than raising an exception.
'''
Inference Functions with Context Support
def infer_charset(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
charset_default: str = CHARSET_DEFAULT,
http_content_type: __.Absential[ str ] = __.absent,
charset_supplement: __.Absential[ str ] = __.absent,
mimetype_supplement: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
) -> __.typx.Optional[ str ]:
''' Infers charset through various means.
Utilizes HTTP Content-Type headers, location hints, and content
analysis for contextual charset inference. Supports configurable
default return behavior on inference failure.
'''
def infer_mimetype_charset(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
charset_default: str = CHARSET_DEFAULT,
mimetype_default: str = MIMETYPE_DEFAULT,
http_content_type: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
charset_supplement: __.Absential[ str ] = __.absent,
mimetype_supplement: __.Absential[ str ] = __.absent,
) -> tuple[ str, __.typx.Optional[ str ] ]:
''' Detects MIME type and charset with context support.
Returns tuple of (mimetype, charset). Provides comprehensive
detection utilizing all available context with configurable
default behavior on detection failure.
'''
Confidence-Based Detection Functions¶
Core Confidence Functions
def detect_charset_confidence(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
default: str = CHARSET_DEFAULT,
supplement: __.Absential[ str ] = __.absent,
mimetype: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
) -> CharsetResult:
''' Detects character encoding with confidence scoring.
Returns CharsetResult with charset and confidence level. When
configured for default return behavior, returns default value
with zero confidence on detection failure.
'''
def detect_mimetype_confidence(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
default: str = MIMETYPE_DEFAULT,
charset: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
) -> MimetypeResult:
''' Detects MIME type with confidence scoring.
Returns MimetypeResult with mimetype and confidence level. When
configured for default return behavior, returns default value
with zero confidence on detection failure.
'''
Advanced Confidence Inference
def infer_charset_confidence(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
charset_default: str = CHARSET_DEFAULT,
http_content_type: __.Absential[ str ] = __.absent,
charset_supplement: __.Absential[ str ] = __.absent,
mimetype_supplement: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
) -> CharsetResult:
''' Infers charset with confidence through various means.
Utilizes contextual information for enhanced detection quality.
Supports configurable default return behavior on inference failure.
'''
def infer_mimetype_charset_confidence(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
charset_default: str = CHARSET_DEFAULT,
mimetype_default: str = MIMETYPE_DEFAULT,
http_content_type: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
charset_supplement: __.Absential[ str ] = __.absent,
mimetype_supplement: __.Absential[ str ] = __.absent,
) -> tuple[ MimetypeResult, CharsetResult ]:
''' Detects MIME type and charset with confidence scoring.
Returns tuple of (MimetypeResult, CharsetResult) with full
confidence information for both detection results. Supports
configurable default behavior on detection failure.
'''
Confidence Utility Functions
def confidence_from_bytes_quantity(
content: Content,
behaviors: Behaviors = BEHAVIORS_DEFAULT
) -> float:
''' Calculates confidence score based on content length.
Returns confidence value from 0.0 to 1.0 based on the amount
of content available for analysis.
'''
High-Level Decoding and Validation¶
Content Decoding
def decode(
content: Content, /, *,
behaviors: Behaviors = BEHAVIORS_DEFAULT,
profile: TextValidationProfile = PROFILE_TEXTUAL,
charset_default: str = CHARSET_DEFAULT,
mimetype_default: str = MIMETYPE_DEFAULT,
http_content_type: __.Absential[ str ] = __.absent,
location: __.Absential[ Location ] = __.absent,
charset_supplement: __.Absential[ str ] = __.absent,
mimetype_supplement: __.Absential[ str ] = __.absent,
) -> str:
''' High-level bytes-to-text decoding with validation.
Performs comprehensive detection, decoding, and validation
for robust text extraction from byte content. Supports
configurable default values for graceful degradation.
'''
Textual Content Validation
def is_textual_mimetype( mimetype: str ) -> bool:
''' Validates if MIME type represents textual content.
Returns True for MIME types representing textual content.
'''
def is_valid_text(
text: str,
profile: TextValidationProfile = PROFILE_TEXTUAL
) -> bool:
''' Unicode-aware text validation with configurable profiles.
Returns True for text meeting the specified validation profile.
'''
Line Separator Processing¶
LineSeparators Enum (unchanged from v1.x specification)
class LineSeparators( __.enum.Enum ):
''' Line separators for cross-platform text processing. '''
CR = '\r' # Classic MacOS (0xD)
CRLF = '\r\n' # DOS/Windows (0xD 0xA)
LF = '\n' # Unix/Linux (0xA)
@classmethod
def detect_bytes(
selfclass,
content: __.cabc.Sequence[ int ] | bytes,
limit: int = 1024
) -> __.typx.Optional[ 'LineSeparators' ]:
''' Detects line separator from byte content sample. '''
@classmethod
def normalize_universal( selfclass, content: str ) -> str:
''' Normalizes all line separators to Unix LF format. '''
def normalize( self, content: str ) -> str:
''' Normalizes specific line separator to Unix LF format. '''
def nativize( self, content: str ) -> str:
''' Converts Unix LF to this platform's line separator. '''
Type Annotation Patterns¶
Module Constants:
CHARSET_DEFAULT: str = 'utf-8'
MIMETYPE_DEFAULT: str = 'application/octet-stream'
Common Type Aliases:
Content: __.typx.TypeAlias = __.typx.Annotated[
bytes,
__.ddoc.Doc( "Raw byte content for analysis." )
]
Location: __.typx.TypeAlias = __.typx.Annotated[
str | __.pathlib.Path,
__.ddoc.Doc( "File path or URL for detection context." )
]
Absential Pattern for Context Parameters: - Distinguish “not provided” (absent) from “explicitly None” - Enable three-state parameters: absent | None | value - Support complex context handling for HTTP headers and supplements
Return Type Patterns: - Simple APIs return str or __.typx.Optional[ str ] - Confidence APIs return structured types: CharsetResult, MimetypeResult - Combined APIs return immutable tuples: tuple[ MimetypeResult, CharsetResult ] - Default return behavior: confidence = 0.0 indicates detection failure with fallback value
Default Return Behavior Pattern: - DetectFailureActions.Default: Return default value with zero confidence - DetectFailureActions.Error: Raise appropriate exception (legacy behavior) - All detection functions accept default parameters for graceful degradation
Exception Hierarchy Design¶
Following Omnierror Pattern¶
class Omniexception(
__.immut.Object, BaseException,
instances_visibles = (
'__cause__', '__context__', __.is_public_identifier ),
):
''' Base for all exceptions raised by package API. '''
class Omnierror( Omniexception, Exception ):
''' Base for error exceptions raised by package API. '''
# Detection-specific exceptions
class CharsetDetectFailure( Omnierror, TypeError, ValueError ):
''' Raised when character encoding detection fails. '''
class CharsetInferFailure( Omnierror, TypeError, ValueError ):
''' Raised when character encoding inference fails. '''
class MimetypeDetectFailure( Omnierror, TypeError, ValueError ):
''' Raised when MIME type detection fails. '''
class ContentDecodeFailure( Omnierror, UnicodeError ):
''' Raised when content cannot be decoded with detected charset. '''
Exception Design Principles: - Follow nomenclature patterns: <Noun><Verb>Failure - Inherit from appropriate built-in exception types - Support location context in error messages - Enable package-wide exception catching via Omnierror
Implementation Considerations¶
Context-Aware Detection Strategy¶
Detection Priority Order: 1. HTTP Content-Type headers (when available) 2. Location/filename extension analysis 3. Magic bytes content analysis 4. Fallback to defaults based on available information
Registry-Based Backend Selection: - Configurable detector precedence via Behaviors - Dynamic fallback when detectors return NotImplemented - Support for multiple optional dependencies per detection type
Confidence Integration: - Length-based confidence calculation - Backend-specific confidence scoring - AsNeeded behavior triggering based on confidence thresholds
Performance Characteristics: - Lazy evaluation of detection operations - Sample-based analysis for large content - Minimal abstraction preserving detector performance