Versioned File Indexer

A memory-efficient C++17 command-line tool that builds word-frequency indexes over large text files using a fixed-size buffer. Memory usage is always independent of file size.

Build

make

Usage

Word Count

./analyzer --file dataset.txt --version v1 --buffer 512 --query word --word error

Top-K

./analyzer --file dataset.txt --version v1 --buffer 512 --query top --top 10

Difference between two versions

./analyzer --file1 dataset_v1.txt --version1 v1 \
           --file2 dataset_v2.txt --version2 v2 \
           --buffer 512 --query diff --word error

Design

Class	Role
BufferedFileReader	RAII binary file reader, fixed-size chunks
Tokenizer	Byte stream → lowercase words, handles cross-buffer tokens
VersionedIndexer	word→frequency map for one version
VersionManager	Owns all indexer instances
Query (abstract)	Base class, pure virtual execute()
WordQuery	Word count query
DiffQuery	Frequency difference across two versions
TopKQuery	Top-K via min-heap, O(N log K)

C++ Features

• Two function templates: measureSeconds<F>, clamp<T>
• Inheritance + runtime polymorphism via unique_ptr<Query>
• Function overloading: two buildIndex signatures
• Exception handling throughout
• RAII resource management