How to extract text and keep their semantics?

[Jun711]

I found an example on how to extract text on a StackOverflow thread

This is the example code linked on the thread

But, it just grabs all the text without keeping the semantics. I wonder if there is a API method that is provided by pdf.js to extract text semantically?

Thanks

[timvandermeij]

The getTextContent API (refer to https://github.com/mozilla/pdf.js/blob/master/examples/node/getinfo.js#L45 for a usage example) can only give you the text content of a single page, but there are no more semantics. This is mainly because in the PDF format text is just a series of glyphs and positions and in general no more information is included. Exceptions are tagged PDFs, which we don't support yet but we do track the support in #6269.