CSV parsers for Striplog

csv_DictReader.py

fname = './log-1_tops_only.csv'
import csv
from striplog import Striplog
from striplog import Lexicon
from striplog import Interval
lexicon = Lexicon.default()

with open(fname) as f:
    data_dict = {}
    for ii in csv.DictReader(f):
        for k,v in ii.items():
            try:
                v = int(v)
            except ValueError:
                try:
                    v = float(v)
                except ValueError:
                    v = str(v)
            try:
                data_dict[k].append(v)
            except KeyError:
                data_dict[k] = [v,]

for ii in Striplog._build_list_of_Intervals(data_dict, lexicon=lexicon):
    print(ii)

csv_genfromtxt.py

fname = './log-1_tops_only.csv'
import numpy as np
from striplog import Striplog
from striplog import Lexicon
lexicon = Lexicon.default()

names, dtype = True, None

# Reading this in now as bytes. That means that we need to convert them
# to more useful types later. Do we expect anything more other than
# `int`, `float` and `str`?
# We can also read a sequence of dtypes, which might be more useful, but
# it might be too much to ask for from users?
data = np.genfromtxt(fname, dtype=dtype, names=names, delimiter=',', encoding=None)

data_dict = {}
if names:
    print('checking names')
    for name in data.dtype.names:
        # We expect `top` and `base`, but we might get `tops` and `bases`.
        if name == 'tops': # There might be other cases worth checking?
            data_dict.update({'top': data[name].astype(float)})
        if name == 'bases': # There might be other cases worth checking?
            data_dict.update({'base': data[name].astype(float)})
        else: # Everything else is handled here, which is quite nice.
            if dtype:
                data_dict.update({name: data[name]})
            else:
                # This feels pretty nasty, to be honest.
                try:
                    data_dict.update({name: data[name].astype(int)})
                except ValueError:
                    try:
                        data_dict.update({name: data[name].astype(float)})
                    except:
                        data_dict.update({name: data[name].astype(str)})

print(data_dict)

for ii in Striplog._build_list_of_Intervals(data_dict, lexicon=lexicon):
    print(ii)

log-1_tops_only.csv