'ValueError: Trailing data' error Resolution

Introduction

Whenever we read a dataset from the .json file, we often get a 'ValueError: Trailing data' error. Whenever we try to import a JSON file into a pandas dataframe, this type of error can take place even if the data is written in separate lines by using '\n'.

For example, below is the content of the JSON file, and we will try to read the data normally using pandas.

{"Id": 1, "Language": "Java", "Inventor": "Jame\nGosling" }
{"Id": 2, "Language": "C", "Inventor": "Dennis\nRitchie" }
{"Id": 3, "Language": "C++", "Inventor": "Bjarne\nStroustrup"}
{"Id": 4, "Language": "Python", "Inventor": "Guido\nvan Rossum"}
{"Id": 5, "Language": "PHP", "Inventor": "Rasmus\nLerdorf"}
{"Id": 6, "Language": "Javascript", "Inventor": "Brendan\nEich"}

You can also try this code with Online Python Compiler

Run Code

Now, we will import this JSON file into pandas.

data = pd.read_json('data.json')
print(data.head())

You can also try this code with Online Python Compiler

Run Code

We will get the following ValueError.

This is because the “Inventor” item in our JSON file contains ‘\n’ to indicate endlines, so we get a ValueError.

Also, see -  Locally Weighted Regression.

Fixing Error

To fix this ValueError, we need to add an additional attribute, "lines" while reading the dataset. We need to make this "lines" attribute true so that pandas read the file as a JSON object per line.

The below code will perfectly read the dataset that is present in JSON format.

data = pd.read_json('data.json', lines = True)
print(data.head(6))

You can also try this code with Online Python Compiler

Run Code

Id    Language            Inventor
0   1        Java       Jame\nGosling
1   2           C     Dennis\nRitchie
2   3         C++  Bjarne\nStroustrup
3   4      Python   Guido\nvan Rossum
4   5         PHP     Rasmus\nLerdorf
5   6  Javascript       Brendan\nEich