RFC 3986 (https://tools.ietf.org/html/rfc3986#section-2.1) allows lower hexadecimal digits for percent encoding, but urllib.parse.quote and its variants convert into only UPPERCASE digits [A-F].

I will create a PR for fix.

I opened a pull request with an invalid bpo issue number so I unlinked it. Then a completely unrelated PR was is shown as linked by me, but I didn't do that. Unlinking that as well.

PRs keep getting linked to this issue under my name, without me doing anything.

I think that may be happening because you are still refering to this issue in the title of
https://github.com/python/cpython/pull/19888?

Not sure how the linkage is made, but if this keeps happening you may open an issue in the core-workflow repo in the Python org.

In the title -> in the description

After Serhiy unlinked the linked PR, it did not happen again, so it may have been an one-off hiccup. I updated the description GH-19888 nevertheless.

What is wrong with the current behavior? Why does the case of hexadecimal digits matter?

There are no options to repr() to control the case of hexadecimal digits in <object object at 0x7f7a16dc4580> and '\ufffe', and I doubt there are such options in other functions producing hexadecimals. So it is a very uncommon request.

I require lower hex digits for percent encoding, which is allowed by the RFC.  Small modifications of the codes can achieve it.

I'm -0 on the new argument. Why do you require a lower case string here?

But they are not required, right?

No, they are not required. RFC 3986 recommends upper case and says that upper and lower case percent encoding are equivalent:

   The uppercase hexadecimal digits 'A' through 'F' are equivalent to
   the lowercase digits 'a' through 'f', respectively.  If two URIs
   differ only in the case of hexadecimal digits used in percent-encoded
   octets, they are equivalent.  For consistency, URI producers and
   normalizers should use uppercase hexadecimal digits for all percent-
   encodings.

> Why do you require a lower case string here?

Well, I want to use urllib.parse.quote to escape special characters, such as ()[]#*/\, in filenames, and for visibility lower hex digits are preferable.
PR 19766 does not break compatibility.

There is no need to add a new flag to quote. You can easily archive your goal with a simple regular expression substitution:

>>> import re, urllib.parse
>>> s = urllib.parse.quote("Example|#?")
>>> s
'Example%7C%23%3F'
>>> re.sub("%[0-9A-F]{2}", lambda q: q.group(0).lower(), s)
'Example%7c%23%3f'

Concur with Christian. Heh, I was going to propose the same.