tenets.core.summarizer Package

Content summarization system for Tenets.

This package provides intelligent text and code summarization capabilities using multiple strategies from simple extraction to advanced ML approaches. The summarization system helps compress large codebases to fit within token limits while preserving the most important information.

Main components: - Summarizer: Main orchestrator for summarization operations - Strategies: Different summarization approaches (extractive, compressive, etc.) - LLMSummarizer: Integration with Large Language Models (costs $)

Example usage

from tenets.core.summarizer import Summarizer, create_summarizer

Create summarizer

summarizer = create_summarizer(mode="extractive")

Summarize text

result = summarizer.summarize( ... long_text, ... target_ratio=0.3 # Compress to 30% of original ... )

print(f"Reduced by {result.reduction_percent:.1f}%")

Attributes

ML_AVAILABLEmodule-attribute

Python

ML_AVAILABLE = True

Classes

LLMConfigdataclass

Python

LLMConfig(provider: LLMProvider = LLMProvider.OPENAI, model: str = 'gpt-4o-mini', api_key: Optional[str] = None, base_url: Optional[str] = None, temperature: float = 0.3, max_tokens: int = 500, system_prompt: str = 'You are an expert at summarizing code and technical documentation. \nYour summaries are concise, accurate, and preserve critical technical details.', user_prompt: str = 'Summarize the following text to approximately {target_percent}% of its original length. \nFocus on the most important information and maintain technical accuracy.\n\nText to summarize:\n{text}\n\nSummary:', retry_attempts: int = 3, retry_delay: float = 1.0, timeout: float = 30.0)

Configuration for LLM summarization.

ATTRIBUTE	DESCRIPTION
provider	LLM provider to use TYPE:LLMProvider
model	Model name/ID TYPE:str
api_key	API key (if not in environment) TYPE:Optional[str]
base_url	Base URL for API (for custom endpoints) TYPE:Optional[str]
temperature	Sampling temperature (0-1) TYPE:float
max_tokens	Maximum tokens in response TYPE:int
system_prompt	System prompt template TYPE:str
user_prompt	User prompt template TYPE:str
retry_attempts	Number of retry attempts TYPE:int
retry_delay	Delay between retries in seconds TYPE:float
timeout	Request timeout in seconds TYPE:float

Attributes

providerclass-attributeinstance-attribute

Python

provider: LLMProvider = OPENAI

modelclass-attributeinstance-attribute

Python

model: str = 'gpt-4o-mini'

api_keyclass-attributeinstance-attribute

Python

api_key: Optional[str] = None

base_urlclass-attributeinstance-attribute

Python

base_url: Optional[str] = None

temperatureclass-attributeinstance-attribute

Python

temperature: float = 0.3

max_tokensclass-attributeinstance-attribute

Python

max_tokens: int = 500

system_promptclass-attributeinstance-attribute

Python

system_prompt: str = 'You are an expert at summarizing code and technical documentation. \nYour summaries are concise, accurate, and preserve critical technical details.'

user_promptclass-attributeinstance-attribute

Python

user_prompt: str = 'Summarize the following text to approximately {target_percent}% of its original length. \nFocus on the most important information and maintain technical accuracy.\n\nText to summarize:\n{text}\n\nSummary:'

retry_attemptsclass-attributeinstance-attribute

Python

retry_attempts: int = 3

retry_delayclass-attributeinstance-attribute

Python

retry_delay: float = 1.0

timeoutclass-attributeinstance-attribute

Python

timeout: float = 30.0

Functions

get_api_key

Python

get_api_key() -> Optional[str]

Get API key from config or environment.

RETURNS	DESCRIPTION
Optional[str]	API key or None

LLMProvider

Bases: Enum

Supported LLM providers.

Attributes

OPENAIclass-attributeinstance-attribute

Python

OPENAI = 'openai'

ANTHROPICclass-attributeinstance-attribute

Python

ANTHROPIC = 'anthropic'

OPENROUTERclass-attributeinstance-attribute

Python

OPENROUTER = 'openrouter'

LOCALclass-attributeinstance-attribute

Python

LOCAL = 'local'

LLMSummarizer

Python

LLMSummarizer(config: Optional[LLMConfig] = None)

Base class for LLM-based summarization.

Provides common functionality for different LLM providers. Handles API calls, retries, and error handling.

Initialize LLM summarizer.

PARAMETER	DESCRIPTION
config	LLM configuration TYPE:Optional[LLMConfig]DEFAULT:None

Attributes

configinstance-attribute

Python

config = config or LLMConfig()

loggerinstance-attribute

Python

logger = get_logger(__name__)

clientinstance-attribute

Python

client = None

Functions

summarize

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None, custom_prompt: Optional[str] = None) -> str

Summarize text using LLM.

PARAMETER	DESCRIPTION
text	Text to summarize TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None
custom_prompt	Custom prompt override TYPE:Optional[str]DEFAULT:None

RETURNS	DESCRIPTION
str	Summarized text

RAISES	DESCRIPTION
RuntimeError	If API call fails after retries

estimate_cost

Python

estimate_cost(text: str) -> Dict[str, float]

Estimate cost of summarization.

PARAMETER	DESCRIPTION
text	Text to summarize TYPE:str

RETURNS	DESCRIPTION
Dict[str, float]	Dictionary with cost estimates

LLMSummaryStrategy

Python

LLMSummaryStrategy(provider: Union[str, LLMProvider] = LLMProvider.OPENAI, model: str = 'gpt-4o-mini', api_key: Optional[str] = None)

LLM-based summarization strategy for use with Summarizer.

Wraps LLMSummarizer to match the SummarizationStrategy interface.

WARNING: This strategy incurs API costs. Always estimate costs before use.

Initialize LLM strategy.

PARAMETER	DESCRIPTION
provider	LLM provider name or enum TYPE:Union[str, LLMProvider]DEFAULT:OPENAI
model	Model to use TYPE:strDEFAULT:'gpt-4o-mini'
api_key	API key (if not in environment) TYPE:Optional[str]DEFAULT:None

Attributes

nameclass-attributeinstance-attribute

Python

name = 'llm'

descriptionclass-attributeinstance-attribute

Python

description = 'High-quality summarization using Large Language Models (costs $)'

requires_mlclass-attributeinstance-attribute

Python

requires_ml = False

loggerinstance-attribute

Python

logger = get_logger(__name__)

summarizerinstance-attribute

Python

summarizer = LLMSummarizer(config)

Functions

summarize

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None) -> str

Summarize text using LLM.

PARAMETER	DESCRIPTION
text	Input text TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None

RETURNS	DESCRIPTION
str	LLM-generated summary

estimate_cost

Python

estimate_cost(text: str) -> Dict[str, float]

Estimate cost for summarizing text.

PARAMETER	DESCRIPTION
text	Text to summarize TYPE:str

RETURNS	DESCRIPTION
Dict[str, float]	Cost estimate dictionary

CompressiveStrategy

Python

CompressiveStrategy(use_nlp: bool = True)

Bases: SummarizationStrategy

Compressive summarization using NLP tokenization.

Removes redundant words and phrases while maintaining meaning. Uses NLP tokenizer for better word processing.

Initialize compressive strategy.

PARAMETER	DESCRIPTION
use_nlp	Whether to use NLP components TYPE:boolDEFAULT:True

Attributes

nameclass-attributeinstance-attribute

Python

name = 'compressive'

descriptionclass-attributeinstance-attribute

Python

description = 'Remove redundancy using NLP tokenization'

requires_mlclass-attributeinstance-attribute

Python

requires_ml = False

loggerinstance-attribute

Python

logger = get_logger(__name__)

use_nlpinstance-attribute

Python

use_nlp = use_nlp and NLP_AVAILABLE

tokenizerinstance-attribute

Python

tokenizer = TextTokenizer(use_stopwords=True)

stopword_managerinstance-attribute

Python

stopword_manager = StopwordManager()

stopwordsinstance-attribute

Python

stopwords = get_set('prompt')

Functions

summarize

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None) -> str

Compress text by removing redundancy.

PARAMETER	DESCRIPTION
text	Input text TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None

RETURNS	DESCRIPTION
str	Compressed text

ExtractiveStrategy

Python

ExtractiveStrategy(use_nlp: bool = True)

Bases: SummarizationStrategy

Extractive summarization using NLP components.

Selects the most important sentences based on keyword density, position, and optionally semantic similarity. Uses centralized NLP components for improved sentence scoring.

Initialize extractive strategy.

PARAMETER	DESCRIPTION
use_nlp	Whether to use NLP components for enhanced extraction TYPE:boolDEFAULT:True

Attributes

nameclass-attributeinstance-attribute

Python

name = 'extractive'

descriptionclass-attributeinstance-attribute

Python

description = 'Extract important sentences using NLP analysis'

requires_mlclass-attributeinstance-attribute

Python

requires_ml = False

loggerinstance-attribute

Python

logger = get_logger(__name__)

use_nlpinstance-attribute

Python

use_nlp = use_nlp and NLP_AVAILABLE

keyword_extractorinstance-attribute

Python

keyword_extractor = KeywordExtractor(use_stopwords=True, stopword_set='prompt')

tokenizerinstance-attribute

Python

tokenizer = TextTokenizer(use_stopwords=True)

Functions

summarize

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None) -> str

Extract important sentences to create summary.

PARAMETER	DESCRIPTION
text	Input text TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None

RETURNS	DESCRIPTION
str	Extractive summary

SummarizationStrategy

Bases: ABC

Abstract base class for summarization strategies.

Attributes

nameclass-attributeinstance-attribute

Python

name: str = 'base'

descriptionclass-attributeinstance-attribute

Python

description: str = 'Base summarization strategy'

requires_mlclass-attributeinstance-attribute

Python

requires_ml: bool = False

Functions

summarizeabstractmethod

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None) -> str

Summarize text.

PARAMETER	DESCRIPTION
text	Input text TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None

RETURNS	DESCRIPTION
str	Summarized text

TextRankStrategy

Python

TextRankStrategy(use_nlp: bool = True)

Bases: SummarizationStrategy

TextRank summarization with NLP preprocessing.

Graph-based ranking algorithm that uses NLP components for better text preprocessing and similarity computation.

Initialize TextRank strategy.

PARAMETER	DESCRIPTION
use_nlp	Whether to use NLP components TYPE:boolDEFAULT:True

Attributes

nameclass-attributeinstance-attribute

Python

name = 'textrank'

descriptionclass-attributeinstance-attribute

Python

description = 'Graph-based summarization with NLP preprocessing'

requires_mlclass-attributeinstance-attribute

Python

requires_ml = True

loggerinstance-attribute

Python

logger = get_logger(__name__)

use_nlpinstance-attribute

Python

use_nlp = use_nlp and NLP_AVAILABLE and SKLEARN_AVAILABLE

tfidf_calcinstance-attribute

Python

tfidf_calc = TFIDFCalculator(use_stopwords=True)

Functions

summarize

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None) -> str

Summarize using TextRank algorithm.

PARAMETER	DESCRIPTION
text	Input text TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None

RETURNS	DESCRIPTION
str	TextRank summary

TransformerStrategy

Python

TransformerStrategy(model_name: str = 'facebook/bart-large-cnn')

Bases: SummarizationStrategy

Transformer-based neural summarization.

Uses pre-trained transformer models for high-quality abstractive summarization.

Initialize transformer strategy.

PARAMETER	DESCRIPTION
model_name	HuggingFace model name TYPE:strDEFAULT:'facebook/bart-large-cnn'

Attributes

nameclass-attributeinstance-attribute

Python

name = 'transformer'

descriptionclass-attributeinstance-attribute

Python

description = 'Neural summarization using transformers'

requires_mlclass-attributeinstance-attribute

Python

requires_ml = True

loggerinstance-attribute

Python

logger = get_logger(__name__)

model_nameinstance-attribute

Python

model_name = model_name

summarizerinstance-attribute

Python

summarizer = None

Functions

summarize

Python

summarize(text: str, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None) -> str

Summarize using transformer model.

PARAMETER	DESCRIPTION
text	Input text TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
max_length	Maximum summary length TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length TYPE:Optional[int]DEFAULT:None

RETURNS	DESCRIPTION
str	Neural summary

BatchSummarizationResultdataclass

Python

BatchSummarizationResult(results: List[SummarizationResult], total_original_length: int, total_summary_length: int, overall_compression_ratio: float, total_time_elapsed: float, files_processed: int, files_failed: int)

Result from batch summarization.

Attributes

resultsinstance-attribute

Python

results: List[SummarizationResult]

total_original_lengthinstance-attribute

Python

total_original_length: int

total_summary_lengthinstance-attribute

Python

total_summary_length: int

overall_compression_ratioinstance-attribute

Python

overall_compression_ratio: float

total_time_elapsedinstance-attribute

Python

total_time_elapsed: float

files_processedinstance-attribute

Python

files_processed: int

files_failedinstance-attribute

Python

files_failed: int

Functions

to_dict

Python

to_dict() -> Dict[str, Any]

Convert to dictionary.

SummarizationMode

Bases: Enum

Available summarization modes.

Attributes

EXTRACTIVEclass-attributeinstance-attribute

Python

EXTRACTIVE = 'extractive'

COMPRESSIVEclass-attributeinstance-attribute

Python

COMPRESSIVE = 'compressive'

TEXTRANKclass-attributeinstance-attribute

Python

TEXTRANK = 'textrank'

TRANSFORMERclass-attributeinstance-attribute

Python

TRANSFORMER = 'transformer'

LLMclass-attributeinstance-attribute

Python

LLM = 'llm'

AUTOclass-attributeinstance-attribute

Python

AUTO = 'auto'

SummarizationResultdataclass

Python

SummarizationResult(original_text: str, summary: str, original_length: int, summary_length: int, compression_ratio: float, strategy_used: str, time_elapsed: float, metadata: Dict[str, Any] = None)

Result from summarization operation.

ATTRIBUTE	DESCRIPTION
original_text	Original text TYPE:str
summary	Summarized text TYPE:str
original_length	Original text length TYPE:int
summary_length	Summary length TYPE:int
compression_ratio	Actual compression ratio achieved TYPE:float
strategy_used	Which strategy was used TYPE:str
time_elapsed	Time taken to summarize TYPE:float
metadata	Additional metadata TYPE:Dict[str, Any]

Attributes

original_textinstance-attribute

Python

original_text: str

summaryinstance-attribute

Python

summary: str

original_lengthinstance-attribute

Python

original_length: int

summary_lengthinstance-attribute

Python

summary_length: int

compression_ratioinstance-attribute

Python

compression_ratio: float

strategy_usedinstance-attribute

Python

strategy_used: str

time_elapsedinstance-attribute

Python

time_elapsed: float

metadataclass-attributeinstance-attribute

Python

metadata: Dict[str, Any] = None

reduction_percentproperty

Python

reduction_percent: float

Get reduction percentage.

Functions

to_dict

Python

to_dict() -> Dict[str, Any]

Convert to dictionary.

Summarizer

Python

Summarizer(config: Optional[TenetsConfig] = None, default_mode: Optional[str] = None, enable_cache: bool = True)

Main summarization orchestrator.

Coordinates different summarization strategies and provides a unified interface for content compression. Supports single and batch processing, strategy selection, and caching.

ATTRIBUTE	DESCRIPTION
config	TenetsConfig instance
logger	Logger instance
strategies	Available summarization strategies TYPE:Dict[SummarizationMode, SummarizationStrategy]
cache	Summary cache for repeated content TYPE:Dict[str, SummarizationResult]
stats	Summarization statistics

Initialize summarizer.

PARAMETER	DESCRIPTION
config	Tenets configuration TYPE:Optional[TenetsConfig]DEFAULT:None
default_mode	Default summarization mode TYPE:Optional[str]DEFAULT:None
enable_cache	Whether to enable caching TYPE:boolDEFAULT:True

Attributes

configinstance-attribute

Python

config = config or TenetsConfig()

loggerinstance-attribute

Python

logger = get_logger(__name__)

default_modeinstance-attribute

Python

default_mode = SummarizationMode(default_mode)

strategiesinstance-attribute

Python

strategies: Dict[SummarizationMode, SummarizationStrategy] = {EXTRACTIVE: ExtractiveStrategy(), COMPRESSIVE: CompressiveStrategy(), TEXTRANK: TextRankStrategy()}

enable_cacheinstance-attribute

Python

enable_cache = enable_cache

cacheinstance-attribute

Python

cache: Dict[str, SummarizationResult] = {}

statsinstance-attribute

Python

stats = {'total_summarized': 0, 'total_time': 0.0, 'cache_hits': 0, 'cache_misses': 0, 'strategies_used': {}}

Functions

summarize

Python

summarize(text: str, mode: Optional[Union[str, SummarizationMode]] = None, target_ratio: float = 0.3, max_length: Optional[int] = None, min_length: Optional[int] = None, force_strategy: Optional[SummarizationStrategy] = None) -> SummarizationResult

Summarize text content.

PARAMETER	DESCRIPTION
text	Text to summarize TYPE:str
mode	Summarization mode (uses default if None) TYPE:Optional[Union[str, SummarizationMode]]DEFAULT:None
target_ratio	Target compression ratio (0.3 = 30% of original) TYPE:floatDEFAULT:0.3
max_length	Maximum summary length in characters TYPE:Optional[int]DEFAULT:None
min_length	Minimum summary length in characters TYPE:Optional[int]DEFAULT:None
force_strategy	Force specific strategy instance TYPE:Optional[SummarizationStrategy]DEFAULT:None

RETURNS	DESCRIPTION
SummarizationResult	SummarizationResult with summary and metadata

Example

summarizer = Summarizer() result = summarizer.summarize( ... long_text, ... mode="extractive", ... target_ratio=0.25 ... ) print(f"Reduced by {result.reduction_percent:.1f}%")

summarize_file

Python

summarize_file(file: FileAnalysis, mode: Optional[Union[str, SummarizationMode]] = None, target_ratio: float = 0.3, preserve_structure: bool = True, prompt_keywords: Optional[List[str]] = None) -> SummarizationResult

Summarize a code file intelligently.

Handles code files specially by preserving important elements like class/function signatures while summarizing implementations. Enhanced with context-aware documentation summarization that preserves relevant sections based on prompt keywords.

PARAMETER	DESCRIPTION
file	FileAnalysis object TYPE:FileAnalysis
mode	Summarization mode TYPE:Optional[Union[str, SummarizationMode]]DEFAULT:None
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
preserve_structure	Whether to preserve code structure TYPE:boolDEFAULT:True
prompt_keywords	Keywords from user prompt for context-aware summarization TYPE:Optional[List[str]]DEFAULT:None

RETURNS	DESCRIPTION
SummarizationResult	SummarizationResult

batch_summarize

Python

batch_summarize(texts: List[Union[str, FileAnalysis]], mode: Optional[Union[str, SummarizationMode]] = None, target_ratio: float = 0.3, parallel: bool = True, prompt_keywords: Optional[List[str]] = None) -> BatchSummarizationResult

Summarize multiple texts in batch.

PARAMETER	DESCRIPTION
texts	List of texts or FileAnalysis objects TYPE:List[Union[str, FileAnalysis]]
mode	Summarization mode TYPE:Optional[Union[str, SummarizationMode]]DEFAULT:None
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
parallel	Whether to process in parallel TYPE:boolDEFAULT:True
prompt_keywords	Keywords from user prompt for context-aware documentation summarization TYPE:Optional[List[str]]DEFAULT:None

RETURNS	DESCRIPTION
BatchSummarizationResult	BatchSummarizationResult

clear_cache

Python

clear_cache()

Clear the summary cache.

get_stats

Python

get_stats() -> Dict[str, Any]

Get summarization statistics.

RETURNS	DESCRIPTION
Dict[str, Any]	Dictionary of statistics

Functions

create_llm_summarizer

Python

create_llm_summarizer(provider: str = 'openai', model: Optional[str] = None, api_key: Optional[str] = None) -> LLMSummaryStrategy

Create an LLM summarizer with defaults.

PARAMETER	DESCRIPTION
provider	Provider name (openai, anthropic, openrouter) TYPE:strDEFAULT:'openai'
model	Model name (uses provider default if None) TYPE:Optional[str]DEFAULT:None
api_key	API key (uses environment if None) TYPE:Optional[str]DEFAULT:None

RETURNS	DESCRIPTION
LLMSummaryStrategy	Configured LLMSummaryStrategy

summarizer = create_llm_summarizer("openai", "gpt-4o-mini") >>> summary = summarizer.summarize(long_text, target_ratio=0.2)

create_summarizer

Python

create_summarizer(config: Optional[TenetsConfig] = None, mode: str = 'auto', enable_cache: bool = True) -> Summarizer

Create a configured summarizer.

Convenience function to quickly create a summarizer with sensible defaults.

PARAMETER	DESCRIPTION
config	Optional configuration TYPE:Optional[TenetsConfig]DEFAULT:None
mode	Default summarization mode TYPE:strDEFAULT:'auto'
enable_cache	Whether to enable caching TYPE:boolDEFAULT:True

RETURNS	DESCRIPTION
Summarizer	Configured Summarizer instance

Example

summarizer = create_summarizer(mode="extractive") result = summarizer.summarize(text, target_ratio=0.25)

estimate_compression

Python

estimate_compression(text: str, target_ratio: float = 0.3, mode: str = 'extractive') -> dict

Estimate compression results without actually summarizing.

Useful for planning and understanding how much compression is possible for given text.

PARAMETER	DESCRIPTION
text	Text to analyze TYPE:str
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
mode	Summarization mode TYPE:strDEFAULT:'extractive'

RETURNS	DESCRIPTION
dict	Dictionary with estimates

Example

estimate = estimate_compression(long_text, 0.25) print(f"Expected output: ~{estimate['expected_length']} chars")

summarize_files

Python

summarize_files(files: list, target_ratio: float = 0.3, mode: str = 'auto', config: Optional[TenetsConfig] = None) -> BatchSummarizationResult

Summarize multiple files in batch.

Convenience function for batch processing.

PARAMETER	DESCRIPTION
files	List of FileAnalysis objects or text strings TYPE:list
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3
mode	Summarization mode TYPE:strDEFAULT:'auto'
config	Optional configuration TYPE:Optional[TenetsConfig]DEFAULT:None

RETURNS	DESCRIPTION
BatchSummarizationResult	BatchSummarizationResult

Example

from tenets.core.summarizer import summarize_files results = summarize_files(file_list, target_ratio=0.25) print(f"Compressed {results.files_processed} files")

quick_summary

Python

quick_summary(text: str, max_length: int = 500) -> str

Quick summary with simple length constraint.

Convenience function for quick summarization without needing to manage summarizer instances.

PARAMETER	DESCRIPTION
text	Text to summarize TYPE:str
max_length	Maximum length in characters TYPE:intDEFAULT:500

RETURNS	DESCRIPTION
str	Summarized text

Example

from tenets.core.summarizer import quick_summary summary = quick_summary(long_text, max_length=200)

summarize_code

Python

summarize_code(code: str, language: str = 'python', preserve_structure: bool = True, target_ratio: float = 0.3) -> str

Summarize code while preserving structure.

Specialized function for code summarization that maintains imports, signatures, and key structural elements.

PARAMETER	DESCRIPTION
code	Source code TYPE:str
language	Programming language TYPE:strDEFAULT:'python'
preserve_structure	Keep imports and signatures TYPE:boolDEFAULT:True
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3

RETURNS	DESCRIPTION
str	Summarized code

Example

from tenets.core.summarizer import summarize_code summary = summarize_code( ... long_module, ... language="python", ... target_ratio=0.25 ... )

estimate_llm_cost

Python

estimate_llm_cost(text: str, provider: str = 'openai', model: str = 'gpt-3.5-turbo', target_ratio: float = 0.3) -> dict

Estimate cost of LLM summarization.

Calculate expected API costs before summarizing.

PARAMETER	DESCRIPTION
text	Text to summarize TYPE:str
provider	LLM provider TYPE:strDEFAULT:'openai'
model	Model name TYPE:strDEFAULT:'gpt-3.5-turbo'
target_ratio	Target compression ratio TYPE:floatDEFAULT:0.3

RETURNS	DESCRIPTION
dict	Cost estimate dictionary

Example

from tenets.core.summarizer import estimate_llm_cost cost = estimate_llm_cost(text, "openai", "gpt-4") print(f"Estimated cost: ${cost['total_cost']:.4f}")

select_best_strategy

Python

select_best_strategy(text: str, target_ratio: float, constraints: Optional[dict] = None) -> str

Select best summarization strategy for given text.

Analyzes text characteristics and constraints to recommend the optimal summarization approach.

PARAMETER	DESCRIPTION
text	Text to analyze TYPE:str
target_ratio	Target compression ratio TYPE:float
constraints	Optional constraints (time, quality, cost) TYPE:Optional[dict]DEFAULT:None

RETURNS	DESCRIPTION
str	Recommended strategy name

Example

from tenets.core.summarizer import select_best_strategy strategy = select_best_strategy( ... text, ... 0.25, ... {'max_time': 1.0, 'quality': 'high'} ... ) print(f"Recommended: {strategy}")

Modules

• llm - Llm module
• strategies - Strategies module
• summarizer - Summarizer module