extract_words() behaving poorly while extract_text() works fine with the same parameters.

[PrimoJefe]

When I use extract_text() it prints out fine as you can see here :

With every other pdfs for now I did not have any problems, but in this pdf when using extract_words() on the same part it includes white spaces :

At first I thought I needed to change the tolerance but while checking the values of x0 and x1 those white spaces are not supposed to be detected, unless there is something I did not understand. Here are the values :

Anyone has any idea what is going on?
Thanks in advance

[jsvine]

Hello, and thanks for your interest in this library. Apologies for not responding sooner. Are you able to provide the PDF and specific page number? It's a bit difficult to debug without it. But a couple of observations that may or may not be helpful:

• You're passing a y_tolerance in your first example, but not in your second.

• Passing extra_attrs changes how pdfplumber groups characters into words. In your example, if two characters do not share the same size and fontname, they will not be grouped into the same word.

[PrimoJefe]

Hi, thank you for your answer.

Indeed, the problem was that some characters inside the same words did not have the same fontname for some reason.
I did not expect that behavior, also they were pretty similar (BHKDKN+stoneSans vs BHJKJD+stoneSans).

Thank you again for your time, this library is great.