Magic Methods for PyPDF2

[MartinThoma]

I would like to use magic methods for PyPDF2 and hear your ideas about it:

#1: PdfFileReader.__length__ ❌

Usage example:

>>> from PyPDF2 import PdfFileReader
>>> reader = PdfFileReader("path/to/my.pdf")
>>> len(reader)   # the number of pages
42

#2: PageRange.__add__ ✔️ - #948

Usage example:

>>> from PyPDF2 import PageRange
>>> PageRange("1:3") + PageRange("2:5")
PageRange('1:5')

#3: PdfFileReader.__rshift__ ❌

Put all contents of one pdf in a reader:

from PyPDF2 import PdfFileReader, PdfFileWriter
reader = PdfFileReader("path/to/my.pdf")
writer = PdfFileWriter()

# New
reader >> writer

# Equivalent to the old:
for page in reader.pages:
    writer.addPage(page)

#4: PdfFileReader.__getitem__ ❌

from PyPDF2 import PdfFileReader
reader = PdfFileReader("path/to/my.pdf")

# New
page = reader[42]

# Equivalent to the old:
page = reader.pages[42]

Thoughts?

Those are the ones that pop to my mind, but maybe the heavy PyPDF2 users have other operations in mind that you would like to do? What do you think about those?

[MartinThoma]

This is what I imagine: #759 - there is still a lot of functions / attributes I would need to rename.

[sekrause]

• PageRange.__add__ looks nice.
• PdfFileReader.__rshift__ looks very un-pythonic and confusing. I would argue against something like that.
• I would also argue against PdfFileReader.__length__ and PdfFileReader.__getitem__. I think it's better to only allow those operations on reader.pages. Having two different ways of doing that is against the Zen of Python: "There should be one-- and preferably only one --obvious way to do it."

[MasterOdin]

I agree with the point about PdfFileReader.__length__ and PdfFileReader.__getitem__ as I'd expect those (or any magic method) on PdfFileReader to operate directly onto the reader, and not an underlying property of it (pages).

[exiledkingcc]

I prefer adding a PdfPageView class to present all pages in pdf, then add __length__, __getitem__, __iter__, __next__ for it.