"\t" no longer supported as an indent

[ZLJasonG]

Python 3.7.6
ujson 4.0.0

import ujson
ujson.dump(output, f, indent="\t")

TypeError: an integer is required (got type str)

This behaviour has changed since ujson-3.2.0

[hugovk]

Confirmed, here's a self-contained example:

import ujson

print(ujson.__version__)

# ok
ujson.dumps([{"key": "value"}])

# TypeError: an integer is required (got type str)
ujson.dumps([{"key": "value"}], indent="\t")

[hugovk]

git bisect says 5b979ee is the first bad commit, from PR #426:

commit 5b979eeebf31f3bc8d5de914e578935cce8571e1
Author: Chen-Han Hsiao (Stanley)
Date:   Thu Sep 17 20:39:38 2020 +0800

    Fix indent and add test case

 python/objToJSON.c  | 2 +-
 tests/test_ujson.py | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

@chenhan1218, please could you check this? Thank you!

[ZLJasonG]

To be fair, it doesn't look like it actually worked before, it used a single space instead of the tab, it just didn't complain about it.

[hugovk]

Good point, and there's already an issue for it: #319 (edit: and #317).

The TypeError happened in version 2, and then commit fe0e88d from PR #327 (version 3.0.0) replaced it with spaces:

[
 {
  "key": "value"
 }
]

It would be good to support the same API as stdlib json:

If indent is a non-negative integer or string, then JSON array elements and object members will be pretty-printed with that indent level. An indent level of 0, negative, or "" will only insert newlines. None (the default) selects the most compact representation. Using a positive integer indent indents that many spaces per level. If indent is a string (such as "\t"), that string is used to indent each level.

https://docs.python.org/3/library/json.html#basic-usage

[chenhan1218]

@hugovk
Yes. I could confirmed.
And this issue has also happened on v2.0.0, v1.34

As you mentioned, the issue is same as #319 .

( In v3.0.0, this TypeError issue doesn't happen but the behavior is also not correct. )

[hugovk]

Thanks all, let's close this as a duplicate of #319.