Extract text, links, images, tables from Pdf with Python | PyMuPDF, PyPdf, PdfPlumber tutorial

[@yp4577]

Thank you so much for this! I've been looking for a clear video on how to get information out of pdf's, and you provided a very good start

[@SreesFun]

Great Video! I have a challenge on getting large table which is spanned across pages. The table starts from one page and extends to the next page. I want to read this as a single table. Please can you advice me on this?

[@jonolavabeland8042]

In the last part of the video it is said that a table of content can be extracted with pymupdf, but I dont see anything like that in the code you are showing?

[@gadomix3989]

Thank you 🙏 so easy to understand and helpful I hope you explain desktop applications

[@generic-youtube-user]

hello @Pythonology good stuff! Do you know what can be the case if PDFPlumber is not detecting a table, even tho all that page has is a table? it reads everything under normal text for some reason. Also, do you know how multi column PDFs are parsed?

[@basicelifeexperions8536]

thanks for video and the proper documentation, appreciate your work keep-it-up bro..

[@ideationtosuccess5439]

Awesome. I am also interested in knowing how to extract text and import into EXCEL file which is my ultimate requirement.

[@shashankshekhar7659]

Can you try for merged cells, rows and columns? Those are tricky and I really do not know which library is best while extracting data from merged cells.

[@Applepievava]

really appreciate your effort simple and clear !

[@Pythonology]

Find the source code here: pythonology.eu/what-is-the-best-python-pdf-library/

[@kalisrani6243]

Someone please tell me where is the file.pdf used on this video?

[@mohmmedaloustah9075]

Thank you very much , did you to use it with arabic pdfs ? since im facing issue with string is correbted .

[@ahmedebenhassine2828]

is ther a way to combine tables and text extraction, I men the result should be "text1, then a table [name, etc], another text"

[@Jimbooos]

not an easy way. You have to do it "by hand" which may become tideous

[@adhy612000151]

great!!! thanks for your explanation! God bless!

[@douglas_techbot]

Very Nice my friend!!! Thank you

[@asheeshmathur]

Good Tutorial, how do I read a PDF in Bulgaria, it has a different Charset and have text in table etc. Thansk

[@nicolassuarez2933]

Outstanding! how to extract table of contents? Thanks

[@eliaszeray7981]

Great! Thank you.

[@MagendraVaradhan]

Thank you so much Sir, any way to extract the tags in a pdf and alternative texts

[@ishdeepsingh3313]

The table has a line above it- A sample table to extract. Is there a way I can extract that line along with the table as well using PDF plumber or any other library?

[@lavanyan7260]

How to tag pdf links using python

[@abigailmapuladikobo9941]

Thanks for the video. How can we extract text data from multiple pdf files(more than 100)? I want to extract the “abstract “ which is a paragraph, in every pdf file

[@ahowl7mx]

Looks like a cool demo but both pdfplumber and pymupdf doesn't work on my pdf. Wonder if my file is broken, it isn't complicated. :/ Any text is '', no result.

[@saaraff5599]

I have a solution initially ocr the pdf from any online ocr pdf website then use the pdf the these code am sure it will help

[@vaibhavshinde6419]

are these pip packages free for commercial use?

[@PANDURANG99]

is it possible to read read pdf from online location like google drive, sharepoint using python without download pdf

[@oguve278]

That sounds quite sneaky, but I’d take screenshots of your screen and utilize some sort of Computer Vision detection or OCR…

[@PANDURANG99]

@@oguve278 Great, but there is so difference between OCR,cv and pdf, in pdf you will get exact text but in cv it confused between zero and O , I and 1 so much complicated without predefined text format.

[@henr22]

Thank you for the video 👍

[@ROKKor-hs8tg]

How can geometric shapes be extracted?

[@giacomobonomelli]

thank you!

[@gvenagas]

I found that by opening a pdf file with Mozilla Firefox and inspecting it with the developer tools you can collect its text (with the help of JavaScript) after the web browser has converted it to HTML and maybe save it for further processing with someone programming language.

[@ROKKor-hs8tg]

Pypdf2 Pdfreader Not work How all pages with fitz

[@dodgewagen]

Thank you!

[@salemsalem4329]

where the pdf file is ,you need to provide this file

[@vasupatel7013]

Hi is there any way to make some thing that can identify how many pages in a PDF are having image and how many pages are non Image using python or any other language

[@ilianos]

I'm sure you can do that somehow with PyMuPDF. As it allows you to process a single page. The question remains, how you would extract only pages (or rather the page number = "pno" in PyMuPDF) when there's an image that was extracted from that page. Maybe ask GPT-4, it was able to help me set up some basic Python code for PyMuPDF.

[@Julian-tf8nj]

In a test, I had POOR results with pdfplumber : It failed to detect multiple columns, and treated them as 1 row! It also failed a number of times at detecting blank spaces in words - and they get all smushed together. *Copy-and-pasted appalling scan results:* Themovementofoceanwaterisoneofthetwoprinci- shapeofthebasininwhichthecurrentisrunning,extentand pal sources of discrepancy between dead reckoned and location of land, and deflection by the rotation of the earth. PyMuPDF, by contrast, did just fine.

[@Pythonology]

Thank you for the comment Julian. In most cases I prefer PyMuPdf, in general it is the best choice

[@higiniofuentes2551]

Thank you for this very useful video!

[@higiniofuentes2551]

If something is a columnar text (3 or 4), like a banking extract account in pdf which import would be the best? Thank you!

[@MedoHamdani]

Will it work for Arabic?

[@FactoidFreak]

Yes

[@MedoHamdani]

@@FactoidFreak although now there is a gui for it, but will try this way

[@Bos_Taurus]

I would need to get 2 words from a pdf file but the program would have to do that for 100 pdf files

[@aneesh2002]

pymupdf is more faster and advanced

[@manny7662]

Better support for it as well.

[@impradeepx]

wtf r u doing