how too extract the table name with the specified table

[Pranavagrl1]

For Example:
I have shared a image from the pdf i want to extract the table as well as the heading of that table only how can i do it.Is there any possible answer related to that.

[cmdlineluser]

Hi @Pranavagrl1

If you can share the actual page in PDF form it makes it easier to help.

Perhaps you could search for the last line of text that appears before the table:

table = page.find_tables()[0]

for line in page.extract_text_lines():
   if line['top'] > table.bbox[1]:
      break
   previous_line = line
   
print(previous_line['text'])

[Pranavagrl1]

Yeah Let me share you the pdf I want that result in the following output like it will show me the table name then its data.
PDF - Dummy.pdf

[cmdlineluser]

Ah okay, I think you need to alter the snap_tolerance to find those particular tables successfully.

But the same principle applies, e.g.

for page in pdf.pages:
   lines = page.extract_text_lines()
   for table in page.find_tables(dict(snap_tolerance=5)):
      for line in lines:
         if line['top'] > table.bbox[1]:
            break
         previous_line = line
      name = previous_line['text']
      name, table

('Document Table:', <pdfplumber.table.Table at 0x1244d8be0>)
('History:', <pdfplumber.table.Table at 0x131670940>)
('Approval list:', <pdfplumber.table.Table at 0x1320e6350>)
('Client Approval list:', <pdfplumber.table.Table at 0x1321ee7a0>)
('List:', <pdfplumber.table.Table at 0x130c22e00>)

[Pranavagrl1]

Hey it is possible to extract the tables using crop methods because while i am extracting tables using the extract table method it is not giving the exact result. can you give me an example of that using the above pdf.

[cmdlineluser]

Well, this is the output from the previous code example but with name, table.extract() instead:

('Document Table:',
 [[None, '', '', 'Description'],
  ['Product Name:', 'Iphone', None, None],
  ['Sector Name:', 'Technology', None, None],
  ['Department /Function Name:', 'Mobile', None, None],
  ['Version Number:', '5.0', None, None],
  ['Process Owner:', 'Apple', None, None],
  ['Reviewed by (Business Process Management):', 'Me', None, None],
  ['Version Date:', '24/5/2013', None, None],
  ['Next Revision Date:', '25/5/2014', None, None]])
('History:',
 [['Version', 'Date', 'Prepared by', 'Reviewed by', 'Brief Explanation'],
  ['5.0', '24/5/2013', 'Pranav01', 'Karan', 'New document.']])
('Approval list:',
 [['Name', 'Title', 'Signature'],
  ['Rahul', 'Solftware developer', '28 February 2013'],
  ['pranjal', 'Architecture', '22 March 2013'],
  ['Abhimannu', 'QA Testing', '24 May 2013']])
('Client Approval list:',
 [['Name', '', '', 'Title', 'Signature'],
  ['Shrivastav',
   'Machine Learning\nEngineer',
   None,
   None,
   '9 March 2013 (Shared by email)'],
  ['Chandak', 'IT Manager', None, None, '9 March 2013 (Shared by email)'],
  ['Abhinav', 'Director', None, None, '9 March 2013 (Shared by email)']])
('List:',
 [['Name', 'Title', 'Comments for Approver'],
  ['Ahmed', 'Access Delivery Expert', '']])

How does this differ from the result you expect?