Hi, Indie hackers!
RawText is a selfhosted solution for extracting text from any document (*.doc, *.xls, *.pdf, images, etc). It's ultra easy to setup and use, just upload any file into it and get the text from it. RawText handles almost every file format, it also performs OCR on images and PDFs if needed.
Existing Apache Tika, PDFBox and others are too complicated and require much time to get them working properly. So RawText is a ready-to-go solution that is intalled by just one command and does text extraction right out of the box.
It has a simple REST API and supports almost every popular programming language.
We're collecting early requests right now, if we receieve over 1000 requests we will go live.
Cool idea. Minor note about your API. I'm a Python and C# developer and the API is unexpected to me because it's rare to have to pass a callback function as a parameter just to get text as a return value in either language.
It looks idiomatic to me as JavaScript, but not Python or C#. I think it's unusual in Java as well, but I don't do any Java development.
For Python, I recommend following PEP8 style, which would look like this:
Also, I can't figure out how to follow the installation instructions:
The domain name is not complete and https://rawtext.ambar.cloud/rawtext.py does not exist.
Thank you for feedback!
I think you are onto something here. I might know some people that would be very interested in this but a 1MB limit is very hamstrung for their use case. What is the best way to reach out to you to tell you a use case?
Thank you for feedback! The 1 MB limit is only on the web version hosted on our servers, if you install library on your machine the limit will be 250MB which is pretty good I think
Ah snap the file limit is 1MB, I suggest that a file limit is mentioned on the page before user upload the file.
Thank you for feedback, will fix it ASAP