How to get information of superscript and subscript text in the pdf

[sarveshkrg]

Hi, anyone please help to get superscript subscript text identification from normal text in the pdf so that I can save the chemical formula and other constant values present in the pdf as it is into the excel. Thanks in advance.
Input

Output

[jsvine]

Hi @sarveshkrg, and thanks for your interest in this library. Subscript and superscript do not have their own concepts in PDFs; rather, they're just text that use smaller font sizes and positioning. I recommend examining the size and bottom/top attributes of the objects in Page.chars. From there, you may have to write some custom code to extract the characters in superscript in a way that suits your particular goals.

[gulfamhussain80]

@jsvine I am trying to follow this approach, but the results are inconsistent—sometimes accurate, but other times superscripts are extracted as subscripts and vice versa.
@sarveshkrg Were you able to find a solution, I am also facing a similar problm.

[petermr]

The relative positions of superscripts (and subscripts) are dependent on the font and the publishing tool that the authors use. Note how the -23 is differently placed in the san-serif and serif fonts. So your code has to use both the sizes and relative positions.
The superscript/subscript problem can occur when the lines are close together and it's not clear to which line the sub/superscript applies. If it's a single line then it looks like an error of logic.