base85 z85 variant encoding

[underrun]

BPO	31116
Nosy	@pitrou, @Dobatymo, @matan1008
PRs	bpo-31116: Add Z85 variant to base64 #30598

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = None
created_at = <Date 2017-08-03.22:32:22.945>
labels = ['type-feature', 'library', '3.9']
title = 'base85 z85 variant encoding'
updated_at = <Date 2022-01-14.10:46:27.835>
user = 'https://bugs.python.org/underrun'

bugs.python.org fields:

activity = <Date 2022-01-14.10:46:27.835>
actor = 'matan1008'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['Library (Lib)']
creation = <Date 2017-08-03.22:32:22.945>
creator = 'underrun'
dependencies = []
files = []
hgrepos = []
issue_num = 31116
keywords = ['patch']
message_count = 3.0
messages = ['299724', '361991', '362026']
nosy_count = 4.0
nosy_names = ['pitrou', 'underrun', 'Dobatymo', 'matan1008']
pr_nums = ['30598']
priority = 'normal'
resolution = None
stage = 'patch review'
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue31116'
versions = ['Python 3.9']

[underrun]

Issue bpo-17618 introduced base85 options into the base64 module including b85 and a85/adobe variants.

There has since been introduced a variant that is easier to work with specifically as it avoids " ' \ characters which avoids quoting/escaping issues when working with base85 literals in python, json, xml, etc.

https://rfc.zeromq.org/spec:32/Z85/

there is a reference implementation in c for which an extension could easily be built.

alternately it would be very easy to add a _z85alphabet etc. to the base64 module to mirror what has been done with a/b85.

1. is there any interest in this being in the language
2. if so do we want python impl or c extension around ref impl?

[Dobatymo]

base64.b85encode does not use ", ' or \ as well.

That's the _b85alphabet

_b85alphabet = (b"0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"
                b"abcdefghijklmnopqrstuvwxyz!#$%&()*+-;<=>?@^_`{|}~")

[underrun]

z85 avoids all of:

` \ " ' _ , ;

Backtick and semicolon increase the number of places that z85 can be used
without issue over b85. All of these might not be as big an issue in python
as in other tools/languages but interoperability would still be improved
through the addition of z85.

• Derek

On Fri, Feb 14, 2020, 09:18 Dobatymo <report@bugs.python.org> wrote:

Dobatymo <the-real-hazzard@outlook.com> added the comment:

base64.b85encode does not use ", ' or \ as well.

That's the _b85alphabet

_b85alphabet = (b"0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"
b"abcdefghijklmnopqrstuvwxyz!#$%&()*+-;<=>?@^_`{|}~")

----------
nosy: +Dobatymo

---

Python tracker <report@bugs.python.org>
<https://bugs.python.org/issue31116\>

---