read_json & read_ndjson do not support scientific notation with + symbol

[StijnKas]

Polars version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of Polars.

Issue description

It seems read_json and read_ndjson do not currently support scientific notation with + symbol. Notation with the - symbol is supported.

Reproducible example

#imports
import polars as pl
from io import StringIO

#works
pl.read_ndjson(StringIO('{"Value":1.1e-10}'))

shape: (1, 1)
┌────────────┐
│ Value      │
│ ---        │
│ f64        │
╞════════════╡
│ 1.1000e-10 │
└────────────┘

#does not work
pl.read_ndjson(StringIO('{"Value":1.1e+10}'))


thread '<unnamed>' panicked at 'called `Result::unwrap()` on an `Err` value: ExternalFormat("MissingComa(43)")', /Users/runner/work/polars/polars/polars/polars-io/src/ndjson_core/ndjson.rs:161:90
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
---------------------------------------------------------------------------
PanicException                            Traceback (most recent call last)
/var/folders/bn/mzxdq739615fffnwsnzkjkgjfplznp/T/ipykernel_52657/1504185214.py in <cell line: 1>()
----> 1 pl.read_ndjson(StringIO('{"Value":1.1e+10}'))

/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/polars/io.py in read_ndjson(file)
   1025 
   1026     """
-> 1027     return DataFrame._read_ndjson(file)
   1028 
   1029 

/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/polars/internals/dataframe/frame.py in _read_ndjson(cls, file)
    835 
    836         self = cls.__new__(cls)
--> 837         self._df = PyDataFrame.read_ndjson(file)
    838         return self
    839 

PanicException: called `Result::unwrap()` on an `Err` value: ExternalFormat("MissingComa(43)")

Expected behavior

I would expect it to support the plus symbol while reading, seeing as it's supported by the official json spec https://www.json.org/json-en.html.

Installed versions

---Version info---
Polars: 0.15.1
Index type: UInt32
Platform: macOS-12.6.1-x86_64-i386-64bit
Python: 3.10.7 (v3.10.7:6cc6b13308, Sep  5 2022, 14:02:52) [Clang 13.0.0 (clang-1300.0.29.30)]
---Optional dependencies---
pyarrow: 10.0.1
pandas: 1.2.5
numpy: 1.23.3
fsspec: 2022.8.2
connectorx: 0.3.0
xlsx2csv: 0.8
matplotlib: 3.6.1

[universalmind303]

I can take a look at this one. I expect it is an upstream issue with simd-json

[universalmind303]

The root cause is actually from the infer schema which uses json-deserializer.

[universalmind303]

once arrow2 is updated with the latest json-deserializer version, this should be resolved. See jorgecarleitao/arrow2#1321

[StijnKas]

Closed by #5781