在 processing.py 中加载 csv 文件

Question

I am trying to load a csv file in processing.py as a table. The Java environment allows me to use the loadTable() function, however, I'm unable to find an equivalent function in the python environment.

Answer 1

You can parse your csv file in an array list with this :

import numpy as np
array_list = np.genfromtxt('file.csv',delimiter=';',dtype=None)

Answer 2

The missing functionality could be added as follows:

import csv

class Row(object):
    def __init__(self, dict_row):
        self.dict_row = dict_row

    def getFloat(self, key):
        return float(self.dict_row[key])

    def getString(self, key):
        return self.dict_row[key]

class loadTable(object):
    def __init__(self, csv_filename, header):
        with open(csv_filename, "rb") as f_input:
            csv_input = csv.DictReader(f_input)
            self.data = [Row(row) for row in csv_input]

    def rows(self):
        return self.data

This reads the csv file into memory using Python's csv.DictReader class. This treats each row in the csv file as a dictionary. For each row, it creates an instance of a Row class which then lets you retrieve entries in the format required. Currently I have just coded for getFloat() and getString() (which is the default format for all csv values).

Answer 3

You could create an empty Table object with this:

from processing.data import Table
t = Table()

And then populate it as discussed at https://discourse.processing.org/t/creating-an-empty-table-object-in-python-mode-and-some-other-hidden-data-classes/25121

But I think a Python Dict as proposed by @martin-evans would be nice. You load it like this:

import csv
from codecs import open # optional to have the 'enconding="utf-8"' in Python 2

with open("data/pokemon.csv", encoding="utf-8") as f:
    data = list(csv.DictReader(f))  # a list of dicts, col-headers as keys