Try to use table from landscape pdf?

[Rapid1898-code]

Hello - i try to read the table-information from the attached pdf using the following code:

test.pdf

import pdfplumber
import os
import sys
import xlwings as xw
import time
import mysql.connector

if __name__ == '__main__':
  # read table data
  path = os.path.abspath(os.path.dirname(sys.argv[0])) 
  fn = os.path.join(path, "test.pdf")
  table_settings = {
      "vertical_strategy": "lines",
      "horizontal_strategy": "lines"
  }
  pdf = pdfplumber.open(fn)    
    
  workTable= []
  for idxPage, page in enumerate(pdf.pages, start=1):                         
    print(f"Reading tables for page {idxPage}")    
    erg=page.extract_tables(table_settings)
    for idxTable, table in enumerate(erg, start=1):
      for row in table:
        if row[0] in ["", None]:
          continue
        wRow = [idxPage, row[0]]
        worker = row[1:]
        worker = [x for x in worker if x not in ["",None]]
        worker = " ".join(worker)
        wRow.append(worker)
        workTable.append(wRow)        
    
  for e in workTable:
    print(e)    

But i only get the following as output:

[2, 't\ns\ni\nL\n \nm\ne\nt\nI', 's\nk\nr\na\nm\ne\nR']
[3, '1', '1 1 1 2']
[3, 'G\nN\nI\nR\n-\nO', 'G\nN\nI\nR\n-\nO G\nN\nI\nR\n-\nO G\nN\nI\nR\n-\nO p\nu\nk\nc\na\nb\n \n,\nG\nN\nI\nR']
[3, '.\n.\n.\n.', '.\n.\n.\n. .\n.\n.\n. .\n.\n.\n. .\n.\n.\n.']
[3, '0\n4\n1\n0\n1\n-\n0\n0\n2\n4\n1', '0\n8\n1\n0\n1\n-\n0\n0\n2\n4\n1 0\n2\n2\n0\n1\n-\n0\n0\n2\n4\n1 0\n8\n2\n0\n1\n-\n0\n0\n2\n4\n1 0\n4\n1\n0\n2\n-\n0\n1\n2\n4\n1']
[3, '3\n2', '4\n2 5\n2 6\n2 7\n2']

Is it somehow possible to extract the data from this tables?

[jsvine]

Hi, and thanks for sharing this example!

Using the settings dict(text_vertical_ttb=False, vertical_strategy="lines_strict", horizontal_strategy="lines_strict") should work (text_vertical_ttb tells pdfplumber to read the vertical text bottom-to-top instead of top-to-bottom)

... but it appears that there may be a bug in how pdfplumber handles it, because I'm seeing results like this:

  'slew\nASSY,\nMOTOR',
  'SALE\nFOR\nNOT',
  'KIT\nPINION',
  'SEAL',
  'O-RING',
  'KIT\nCARRIER',
  'cap\nSCREW,',
  'SALE\nFOR\nNOT',
  'seal\nKIT,',

... where it seems spaces are being treated as newlines, and the order of those space-separated words are being reversed. I'll look into fixing that.