How can I get the same layout as the original PDF?

[mrobe]

I tried converting a simple PDF article using this page: https://pdf.oomol.com

It mostly works but some text is lost and the original layout is not preserved.

I mean: same page dimensions, same font size, same margins, line breaks, etc.

I tried opening the EPUB using Calibre e-book viewer on macOS, but the text layout is completely weird.

How can I do this?

[Moskize91]

There are three issues here. The first is the loss of some text. This depends on which text is missing. If it is only the content of the header and footer, it is as expected. However, if the text of the main body is missing, it is a bug (in particular, footer comments need to be specifically required to be retained during conversion to avoid loss).

Secondly, the original paragraph breaks and line breaks are inevitably lost when converting from PDF to EPUB. This is because EPUB paragraphs and PDF paragraphs are not the same concept. In EPUB readers, due to differences in screen size, the text will be reformatted, and the line break positions cannot be fixed at all.

Thirdly, preserving font size and line spacing is a feature that pdf-craft currently lacks and has no plans to add in the future. A possible plan is to allow users to manually set the font and style.

In other words, you might need to determine what effect you want most. If you want to preserve the original style of the PDF to a high degree, you might need to enhance the Scanned PDF feature, which pdf-craft doesn't actually do.

[mrobe]

Thanks for clarifying everything. Regarding the loss of text, it was the very first line — a title —, so I guess pdf-craft believes it is a header and ignored it for that reason.

Regarding the other two issues, I see now these are limitations with EPUB, so they are not bugs or problems with pdf-craft.

Normally, I work with PDF, and aside from being very large files, scanned PDFs typically have a bunch of issues, like the text is slightly skewed, or there is keystone distortion, or lots of scan noise around the edges of the page, or OCR errors are 'invisible' because there is scanned text on the page, etc.

Ideally, I would like an app that can clean up a PDF by doing OCR and then "reconstructing" the correct layout in a new PDF, e.g., following the original page dimensions, line breaks, font size, etc. but all cleaned up. I assume a bunch of hints would need to be provided, for the app to do this well.

Would you have any suggestions about how to achieve this?

[Moskize91]

I don't recall knowing which app could fulfill your PDF cleanup function 😂