Pymupdf: Question / Comment: EPUB: extracting blocks without experiencing page breaks

[andrei-volkau]

Let me consider the following code.

import fitz


if __name__ == '__main__':
    filename = "Test.epub"
    doc = fitz.open(filename)
    for page in doc:
        for b in page.getText("blocks"):
            print("Block", b[-2])
            print(b[4])
            print("-" * 50)

So I am able to extract blocks per each page. I suppose that the actual page breaks are virtual since EPUB is a reflowable document format. I would like to extract all blocks from Test.epub document without encountering page breaks.

Comment: The page breaks are bad for me since they can split a block (actually paragraph inside initial text) into two pieces. So I will need to do postprocessing while trying to merge such blocks again. It would be much simpler to get rid of page breaks at the very beginning.

[JorjMcKie]

actual page breaks are virtual since EPUB is a reflowable document

That's correct.

I would like to extract all blocks from Test.epub document without encountering page breaks.

This is not achievable - except with your own code. I don't know if something is possible by re-layouting the document with a really giant page size, so big that the document then only has one page ... virtually.
Basically something along this idea:

• get first page dimension
• get number of pages
• doc.layout() with a rectangle having a height of doc[0].rect.height * doc.pageCount.
• then do your extract .

[andrei-volkau]

@JorjMcKie, your approach is working! I tested it. Thank you very much! I am closing this question.