A Lightweight and Versatile NLP Toolkit R package textpress version 1.1.1 A toolkit for web scraping, modular NLP pipelines, and text preparation for large language models. Organized around four core actions: fetching, reading, processing, and searching. Covers the full pipeline from raw web data acquisition to structural text processing and BM25 indexing. Supports multiple retrieval strategies including regex, dictionary matching, and ranked keyword search. Pipe-friendly with no heavy dependencies; all outputs are plain data frames or data.tables. Software Depends: R (>= 3.5) Imports: data.table, httr, Matrix, rvest, stringi, stringr, xml2, pbapply, jsonlite, lubridate Suggests: SnowballC (>= 0.7.0), DT, dplyr Jason Timm <JaTimm@salud.unm.edu> Comprehensive R Archive Network (CRAN) Jason Timm [aut, cre] MIT + file LICENSE (https://CRAN.R-project.org/package=textpress/LICENSE) 2026-03-17 application/tgz https://CRAN.R-project.org/package=textpress doi:10.32614/CRAN.package.textpress