Thanks for the note! Glad you find these helpful!

Hello!!

@@PythonicAccountant hello

Thank you!

Thanks!!

Thank you!

Thanks, my pleasure!

Thanks!

Thank you!

Thank you!

Great!

Awesome!

Glad you enjoyed it!

Great to hear!

Sure that’s easy! If the files are the same format, you can create a function that takes a file name as input, and in the function run all the steps needed to read the file, parse, and output. Then you can create a list of filenames and iterate through them, calling the function on each one. You could either manually create the file name list or use pathlib or os.path

Not if you don't tell them ;-)

Haha wow what a compliment!

Sure that would be easily doable if too have the data. Would just need to add a field for report date and use that form the x axis

Each document has to be taken case by case. In that scenario it would depend where that column fell. If there was a clear pattern before or after that column (e.g. a specific length of digits before and a $ after) I could use regex to identify what’s before and after, with everything in the middle belonging to that country column

Any specific use cases or examples you’re looking at?

Sounds like you really like it!

So - if you have an answer - I would love to hear. But I did solve the problem by using PyPDF2 to extract and rotate the pages I needed to analyze and then ran them through PDFPlumber - and while i haven't had a chance to parse the text lines yet - I do have a series of lines that looks appropriate. Thanks Bill

I’d try Bill’s suggestion, basically you want to try and rotate the page using a method that permanently rotates it to the correct position, rather than just rotating the view.

Please check the directory that you are working now and the file saved

Try asking chat gpt :)

How is it formatted? By character location? If so you can just specify the start and end positions of each column in pandas I believe

Don’t recall exactly but I think I found pdfplumber to be either more pythonic or have more functionality

The first pattern of .* is used in the “re” or regular expression context, which is used to do pattern matching. The “.” means any single character, and the “*” means zero or more of the previous pattern. So “.*” literally means to match everything, and it’s usually used to catch everything between other patterns defined before and after. For more info on regular expressions I suggest checking out Al Sweigert’s fantastic content automatetheboringstuff.com/chapter7/ For your second question about *items, in this context I am using a python 3 pattern (believe it started in 3.6) that allows you to unpack an iterable. If I didn’t use the “*”, then it would have added a list as one item rather than each item individually, which would have thrown an error because Line would not have had enough items input into it. Trey Hunner has an awesome article on the use of asterisks in python treyhunner.com/2018/10/asterisks-in-python-what-they-are-and-how-to-use-them/

Thank you so much for your detailed reply!

Yeah that can cause some challenges. Basically if you don’t need the full text, you can just ignore those rows. But if you want the full text, you’ll need to use some way to tell if you have reached the next row or not, then append a string for that cell each row with the new row’s content, and finally add the full record to your list of records once you’ve reached the last row of additional cel text. I’ll usually use a Boolean flag for that, like new_row=True, then flip it to false when you reach the first row of a new row, and check to see if you are at a new row. If you are not, then keep appending, otherwise flip it to True and add to your list of records.

@@PythonicAccountant Hi , if we have multi tables , how we can extract, supposed we have 3k tables in 20 pdf files.

No, but I’m sure you can find lots of freelancers on fiver or other similar sites

@@PythonicAccountant Thanks for responding and recommending Fiver. Keep up the good work.

No but try pdfsplit

This method will not work for scanned PDFs as is, but there are a few other python options that can work decently well depending on the quality of the scan

I love it, think it can but would need to experiment with it!

you’ll want to make sure that the case matches. You could just make everything lowercase. Iterate through each page and look for the string in each page, and if it’s in the page, print the whole page

@@PythonicAccountant Understood... good tip!! Thanks!!

Try with a completely different PDF file. Perhaps it’s an issue with the format of that PDF

@@PythonicAccountant No, it's really a issue: github.com/euske/pdfminer/issues/122

Probably need to import pdfplumber, and if it’s not installed then pip install it

​@@PythonicAccountant ok, got it, the file had been renamed with .pdf.pdf

In general you can tell what packages are needed by looking at the import statements of code. You can also tell by the error message you get in the traceback. In this specific case you would need to install pdfplumber, and the rest should already be included in the anaconda distro.

That’s easy to do but the code would be a little different. You’d want to create separate data frames for each file, then concat the data frames together once you standardize the columns if necessary

im looking for the same. please let me know if you got it

@@hari-codes i got the answer bro , i used tabula to convert PDF to CSV and then read that CSV data ...data will come in for of 2D list like [['1.1',chapter1],['1.2',chapter1]] like this , now iterate to access data using for loop,

@@nilekarmayur If that file is scanned pdf in that case what i can do?

@@srikantpadhy9476 So you'd have to text recognize it. The text recognition in PDF isn't great on scanned PDFs. Just my experience though.

That means that the pdfplumber module hasn’t been installed on the same environment you are running your code in. Make sure to pip install pdfplumber then try it again.

@@PythonicAccountant thanks for your reply.. I have done pip install pdfplumber several times .. but again same error is coming . I'm using python 3.8. please advise .as this is an important program helpful for all accountants in analysis

Vivek Karanath type pip freeze in the environment you are using, and see if pdfplumber is included in that list

I typed pip freeze in command prompt it's not showing anything

Vivek Karanath it sounds like you might not have pip installed. Are you using miniconda or anaconda?