TextSearch

0.0.24 · active · verified Wed Apr 15

TextSearch is a Python library designed for efficient and convenient searching and replacing of multiple strings within text. It leverages C-speed through an Aho-Corasick implementation, making it significantly faster than equivalent regex operations for specific tasks. The library focuses on providing convenience for Natural Language Processing (NLP) and text search tasks, often defaulting to full word matches rather than sub-matches. The current version is 0.0.24, with releases appearing on an as-needed basis.

Warnings

gotcha By default, TextSearch focuses on full word matches (tokens). Users accustomed to standard regex might expect sub-string matches. This behavior can be configured but is a key distinction from typical regex patterns.
Fix: Understand the `TextSearch` constructor parameters, especially how it processes and matches tokens. If sub-string matching is required, ensure your search terms and configuration align with that expectation, or consider alternative libraries if the core 'full word' matching is not desired.
gotcha The library relies on a C-module (`pyahocorasick`) for its performance benefits. This means installation might require a C compiler and development headers on some systems, potentially leading to build errors if not available.
Fix: Ensure your system has the necessary build tools (e.g., `build-essential` on Debian/Ubuntu, Xcode command-line tools on macOS, or Visual C++ build tools on Windows) installed before attempting `pip install textsearch`.
gotcha While TextSearch is significantly faster than regex for *multiple* string searches, for searching a *single* simple string, Python's built-in `str.find()` or `in` operator might be sufficient and have less overhead.
Fix: Evaluate your use case. If you are searching for one or very few fixed strings, simpler built-in Python methods might be more appropriate. TextSearch's benefits shine when dealing with a large dictionary of terms to find or replace.

Install

pip install textsearch Install with pip

Imports

TextSearch
```
from textsearch import TextSearch
```

Quickstart

This quickstart demonstrates how to initialize TextSearch, add multiple keywords, and then use it to find all occurrences or replace them within a given text. It highlights the `case` and `returns` parameters for flexible search behavior.

from textsearch import TextSearch

ts = TextSearch(case="ignore", returns="match")
words_to_find = ["hi", "bye", "hello"]
ts.add(words_to_find)

text = "Hello, hi Pascal, bye, how are you?"
found_matches = ts.findall(text)

print(f"Original text: {text}")
print(f"Words added for search: {words_to_find}")
print(f"Found matches: {found_matches}")

# Example of replacement
ts_replace = TextSearch(case="ignore", returns="replace")
ts_replace.add("hi", "GREETING")
ts_replace.add("bye", "FAREWELL")

replaced_text = ts_replace.replace(text)
print(f"Text after replacement: {replaced_text}")

view raw JSON →