π SimpleTextExtract
SimpleTextExtract attempts extract text from various file types and is recommended as a guard before resorting to something more extreme like Apache Tika. It is built specifically with ActiveStorage in mind and originally built for the purpose of extracting text from attachments in order to index the text in ElasticSearch using SearchKick.
SimpleTextExtract handles parsing text from:
.pdf.docx.doc.xlsx.xls.csv.txtπ
If no text is parsed (for pdf), or a file format is not supported (like images), then nil is returned and you can move on to the heavy-duty tools like Henkei πͺ.
Installation
Add this line to your application's Gemfile:
gem "simple_text_extract"
And then execute:
$ bundle
Or install it yourself as:
$ gem install simple_text_extract
Usage
Text can be parsed from raw file content or files in the filesystem t by calling SimpleTextExtract.extract:
# using ActiveStorage >= 6
extract = .open { |tmp| SimpleTextExtract.extract(tempfile: tmp) }
# raw file content or when ActiveStorage < 6
extract = SimpleTextExtract.extract(filename: .blob.filename, raw: .download)
# filesystem
extract = SimpleTextExtract.extract(filepath: "path_to_file.pdf")
Usage Dependencies
You can choose to use SimpleTextExtract without the following dependencies, but it won't work for specific file types:
pdf parsing requires poppler-utils
brew install poppler
doc parsing requires antiword and unzip
brew install antiword
xlsx and xls parsing requires ssconvert which is part of gnumeric
brew install gnumeric
Usage on Heroku
To use on Heroku you'll have to add some custom buildpacks.
heroku-buildpack-activestorage-preview
If you're using ActiveStorage, you might already have the heroku-buildpack-activestorage-preview added, which means you already have poppler-utils installed π
If not, you can either add that buildpack, or add poppler-utils to your Aptfile (see below).
heroku-buildpack-apt
To add antiword and/or gnumeric* as a dependency on Heroku, install the heroku-buildpack-apt buildpack and follow the install instructions.
In your Aptfile, add:
antiword
gnumeric
unzip
- There is currently an issue with the heroku-18 stack that requires additional dependencies added to the Aptfile to get
gnumericto work properly. You can reference the linked issue above to figure out those dependencies, or downgrade to heroku-16 until it is fixed.
Benchmarks
Benchmarks test extracting text from the same file 50 times (Macbook pro)
| File format | SimpleTextExtract | Henkei (i.e. Yomu/Apache Tika) |
|---|---|---|
| .doc | 1.40s | 74.27s |
| .docx | 0.78s | 71.44s |
| .pdf* | 1.73s | 82.86s |
| .xlsx | 1.16s | 51.89s |
| .xls | 0.80s | 67.88s |
| .txt | 0.04s | 39.25s |
- SimpleTextExtract is limited in its text extraction from pdfs, as Tika can also perform OCR on pdfs with Tesseract
Development
After checking out the repo, run bin/setup to install dependencies. Then, run rake test to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.
To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and tags, and push the .gem file to rubygems.org.
Contributing
Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/simple_text_extract. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.
License
The gem is available as open source under the terms of the MIT License.
Code of Conduct
Everyone interacting in the SimpleTextExtract projectβs codebases, issue trackers, chat rooms and mailing lists is expected to follow the code of conduct.