"keep_blank_chars" not supported in "extract_table" for version 0.8.1

[malek-ewa]

Hello there 👋

I had a code that's been working fine so far, it's extracting tables from a PDF, and it looks like this:

page.extract_table(table_settings={
    "keep_blank_chars" : True,
    "horizontal_strategy": "text",
    "snap_y_tolerance": 4,
    "vertical_strategy": "lines",
    "intersection_x_tolerance": 100,
    "explicit_horizontal_lines": page.curves+page.edges
})

The code was working fine with version 0.7.4 of pdfplumber, until I upgraded to version 0.8.1, and then it started throwing a TypeError: __init__() got an unexpected keyword argument 'keep_blank_chars' error for the above code.

Is this an expected behaviour ? Was keep_blank_chars removed in the most recent versions ?

Thanks in advance.
Malek

[jsvine]

Hi @malek-ewa, and my apologies for the inconvenience. In v0.8.0, keep_blank_chars was changed to text_keep_blank_chars to be more consistent with the rest of the text-based options, and to make it easier for pdfplumber to relay those options to the text extraction methods.

[feelosophy13]

@jsvine FYI, the README still references the keep_blank_chars instead of text_keep_blank_chars.

[jsvine]

Apologies for the confusion. This was fixed in 401b82c and will be in the stable-branch README in the next release, hopefully soon.