Redact/Replace

[Singrig]

Hi ,
Pymupdf is really a great package to work with PDF's and other type of formats..

I have a quick question, I need to redact the sensitivity information from PDF, Is there any function related to redact or can we replace the words in PDF while highlighting?

[JorjMcKie]

The new MuPDF version 1.16.0 also supports Redact annotations.
I haven't decided yet, whether / when to include support for this also in PyMuPDF.
Currently I am still in v1.16.0 development.
Quite a big update to the current v1.14.x unfortunately.

[JorjMcKie]

What do you mean by "sensitivity"? Encryption?
PyMuPDF v1.16.0 will definitely fully support password-based encryption / decryption and permission levels.

[Singrig]

Sorry I'm confused Mupdf as redacting the specific words in PDF ? Can you please share some link?

Let's consider a eg, if PDF has Customer name either I need to redact that customer name or I need to replace with some junk letters

[JorjMcKie]

No, we are talking about different things. Redact annotations are part of the most recent PDF specification.
What you want is not supported at all by MuPDF - sorry.
I have written an anonymizer going in that direction some time ago. This was covering Base-14 fonts only, too. This is significant effort going far beyond what this repository has to offer.

[Singrig]

Sorry for bothering you, if I have list of can't we delete/replace those specific words like we highlighting those words using pymupdf?

[JorjMcKie]

No, this already goes fairly deep into how text is coded in a PDF. In the absolutely most simple cases you might even treat the PDF as a text file and use some editor.

Of course you can always cover sensitive things with - say - a black rectangle. But that is cosmetics only: the information is still there.
A text like "Mr. Anonymous" might be coded in a plethora of ways in unfortunate circumstances: hexadecimal, each single letter being separated from each other and not in natural reading sequence, the text may be split across several lines, ... and what not.

[Singrig]

Can you please share some code or some links that can help to achieve as u suggested like making it into not in natural reading seq or split across line...

[JorjMcKie]

Read chapter text, specifically about textbox text extraction to see how a natural reading order can be re-established.
And of course you should look at the original PDF spec manual to see examples for how complex a simingly simple text may be coded. For example, the PDF manual for this repo on page 9 looks like:

and the highlighted text is coded like this in the PDF file:

BT\n/F38 9.9626 Tf 72 462.176 Td [(PyMuPDF)]TJ/F52 9.9626 Tf 45.81 0 Td [(is)-270(a)-270(Python)-270(binding)-270(for)]TJ

I cannot share the code I developed in the above mentioned case, because that was paid work and I thus do not own the copyright.

[Singrig]

Thanks this will help me... Let me check what I can do... The difficulty would be I'm very new to this Field

[Singrig]

Hi
I'm trying to add text in pdf i'm getting error "name 'Py_RETURN_NONE' is not defined

My code

import fitz
doc = fitz.open('PyMuPDF.pdf') # new or existing PDF
page = doc[0] # new or existing page via doc[n]
p = fitzPoint(50, 72) # start point of 1st line
text = "Some text,\nspread across\nseveral lines."

[JorjMcKie]

you are not using the current version - please switch to 1.14.20

[JorjMcKie]

@Singrig - the new v1.16.11 supports redaction annotations.

[Singrig]

Thanks McKie,hope the more details I can find in PyMupdf docs

…

On Sun 23 Feb, 2020, 5:31 PM Jorj X. McKie, ***@***.***> wrote: @Singrig <https://github.com/Singrig> - the new v1.16.11 supports redaction annotations. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#344?email_source=notifications&email_token=AIV3AO26W2EUTARRTGEOR43REJQRBA5CNFSM4IKLCXPKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMVZ6RA#issuecomment-590061380>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIV3AO5DYWPEDFFOPSEQZRTREJQRBANCNFSM4IKLCXPA> .

[JorjMcKie]

@Singrig - sure you will. The new documentation is already uploaded. I am about to also populate PyPI with the installation material.

[Singrig]

Cool that will help a lot

…

On Sun 23 Feb, 2020, 5:54 PM Jorj X. McKie, ***@***.***> wrote: @Singrig <https://github.com/Singrig> - sure you will. The new documentation is already uploaded. I am about to also populate PyPI with the installation material. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#344?email_source=notifications&email_token=AIV3AO2CQD247LSRYR4VM73REJTIJA5CNFSM4IKLCXPKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMV2M5Y#issuecomment-590063223>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIV3AO2MZUOUCKUP5IYPCU3REJTIJANCNFSM4IKLCXPA> .