如何让json.loads处理Names-Values结构的DataTable序列化JSON？若无则求替代解析器

Absolutely! Python's built-in json.loads() has no problem parsing this structure at all—because it's valid standard JSON; the unique Names/Values organization is just a space-efficient data layout choice, not a deviation from JSON syntax.

How to parse and use the data #

Here's a quick example of loading the structure and converting it into a more familiar row-based format:

import json

# Your sample JSON string
sample_json = '''{ "Names" : ["summaryDate","count"], "Values" : [["2020-01-15T00:00:00",10],["2020-01-16T00:00:00",12],["2020-01-17T00:00:00",16]] }'''

# Parse the JSON into a Python dict
parsed_data = json.loads(sample_json)

# Extract column names and row values
column_names = parsed_data["Names"]
row_values = parsed_data["Values"]

# Convert to a list of dictionaries (matching standard row-wise JSON)
table_rows = [dict(zip(column_names, row)) for row in row_values]

print(table_rows)
# Output:
# [{'summaryDate': '2020-01-15T00:00:00', 'count': 10}, {'summaryDate': '2020-01-16T00:00:00', 'count': 12}, {'summaryDate': '2020-01-17T00:00:00', 'count': 16}]

For tabular data handling #

If you're working with this data as a table, libraries like pandas make conversion even simpler:

import pandas as pd

df = pd.DataFrame(parsed_data["Values"], columns=parsed_data["Names"])
print(df)
# Output:
#           summaryDate  count
# 0  2020-01-15T00:00:00     10
# 1  2020-01-16T00:00:00     12
# 2  2020-01-17T00:00:00     16

Alternative parsers for better performance #

If you need faster parsing (for very large datasets), third-party JSON libraries work just as well with this structure:

• ujson: A fast, lightweight alternative to the standard library
• orjson: Even faster than ujson, with support for more Python types

Example using ujson:

import ujson

parsed_data = ujson.loads(sample_json)
# Same processing steps as above apply

All of these libraries will correctly parse your Names/Values structure because it adheres strictly to JSON standards.

内容的提问来源于stack exchange，提问作者CestLaGalere