def test_queries_empty(self):
df0 = Table(cols, [], 'Empty')
q_create = df0.to_sql_create()
q_values = df0.to_sql_values()
df2 = Table.from_sql_create(q_create)
df2.add_sql_values(q_values)
self.assertEqual(df2.columns, cols)
self.assertEqual(df2.rowcnt, 0)
def parse_xls(self, bs):
import xlrd
def xlrd_parse_date(value):
# XLRD returns dates as strange timestamps which must be properly parsed
return datetime.datetime(*xlrd.xldate_as_tuple(float(value), book.datemode)).date()
book = xlrd.open_workbook(file_contents=bs)
first_sheet = book.sheet_by_index(0)
self._parse_init()
self.row_index = 0
_typefuns = {}
_typefuns.update(typefuns)
_typefuns['date'] = xlrd_parse_date
while True:
try:
row = first_sheet.row_values(self.row_index)
self.parse_row(row, _typefuns)
self.row_index += 1
if self.row_limit > 0 and self.row_index > self.row_limit:
break
except IndexError:
break
return Table(self.res_columns, self.res_rows)
def parse_csv(self, fl, delimiter=None): self._parse_init() csvreader = csv.reader(fl, delimiter=delimiter, quotechar='"') for self.row_index, row in enumerate(csvreader): self.parse_row(row) if self.row_limit > 0 and self.row_index > self.row_limit: break return Table(self.res_columns, self.res_rows)
def test_from_columns(self):
columns = [
df1['type'],
df1['weight'] + df1['size']**2,
]
title = 'Random test title'
df2 = Table.from_columns(columns, title)
self.assertEqual(df2.rowcnt, df1.rowcnt)
self.assertEqual(df2.colcnt, len(columns))
self.assertEqual(df2.title, title)
def test_queries(self):
q_create = df1.to_sql_create()
q_values = df1.to_sql_values()
df2 = Table.from_sql_create(q_create)
df2.add_sql_values(q_values)
self.assertEqual(df2.columns, df1.columns)
self.assertEqual(df2.rowcnt, df1.rowcnt)
self.assertEqual(df2['size'].sum(), df1['size'].sum())
self.assertEqual(df2['weight'].sum(), df1['weight'].sum())
def test_concat(self): df3 = Table.concat([df1, df2]) self.assertEqual(df1.columns, df3.columns) self.assertEqual(df3.rowcnt, df1.rowcnt+df2.rowcnt) self.assertEqual(df3['size'].sum(), df1['size'].sum()+df2['size'].sum()) df1.append(df2) # Now, df1 must be equal to df3 self.assertEqual(df1.columns, df2.columns) self.assertEqual(df1.rowcnt, df3.rowcnt) self.assertEqual(df1['size'].sum(), df3['size'].sum()) trash = '1234567890' df3.rows[1][3] = trash df3.rows[4][3] = trash # It must not affect df1 or df2 self.assertEqual((df1['tel']==trash).sum(), 0) self.assertEqual((df2['tel']==trash).sum(), 0) self.assertEqual((df3['tel']==trash).sum(), 2)
from lemuras import Table cols = ['type', 'size', 'weight', 'tel', 'when'] rows = [ ['A', 1, 12, '+79360360193', '1983-09-14'], ['B', 4, 12, 84505505151, '13.08.2018'], ['A', 3, 10, '31415926535', '2019-02-23 15:20'], ['B', 6, 14, '', '2011/11/11'], ['A', 4, 15, '23816326412', '25/03/95'], ['A', 2, 12, None, None], ] df1 = Table(cols, rows, 'Sample')
def __init__(self, *args, **kwargs): super(TestLemurasTableGroupby, self).__init__(*args, **kwargs) df1 = Table(cols, rows) self.gr = df1.groupby(['type', 'weight'])
def test_load(self): df1 = Table.from_html(html_table, title='Numbers') self.assertEqual(df1.columns, ['Name', 'Value']) self.assertEqual(df1.rowcnt, 3) self.assertEqual(df1['Value'].sum(), 7.4779084706)
def test_json_dict_pretty(self):
txt = df1.to_json(as_dict=True, pretty=True)
df2 = Table.from_json(txt)
self.assertEqual(df2.columns, cols)
self.assertEqual(df2.rowcnt, df1.rowcnt)
self.assertEqual(df2['weight'].sum(), df1['weight'].sum())
def test_json_list(self):
txt = df1.to_json(pretty=False)
df2 = Table.from_json(txt)
self.assertEqual(df2.columns, cols)
self.assertEqual(df2.rowcnt, df1.rowcnt)
self.assertEqual(df2['weight'].sum(), df1['weight'].sum())
def test_outer(self): df3 = Table.merge(df1, df2, ['type', 'size'], 'outer') self.assertEqual(df3.rowcnt, cnt_inner+cnt_left+cnt_right) self.assertEqual(df3.columns, ['type', 'size', 'weight', 'tel', 'cost']) self.assertEqual(df3.nunique()[1].sum(), 23) self.assertEqual(df3['weight'].sum(), 74)
def test_inner(self): df3 = Table.merge(df1, df2, ['type', 'size'], 'inner') self.assertEqual(df3.rowcnt, cnt_inner) self.assertEqual(df3.columns, ['type', 'size', 'weight', 'tel', 'cost']) self.assertEqual(df3.nunique()[1].sum(), 13) self.assertEqual(df3['weight'].sum(), 34)
import unittest from lemuras import Table cols1 = ['type', 'size', 'weight', 'tel'] rows1 = [ ['A', 2, 12, '+79360360193'], ['B', 1, 12, 84505505151], ['D', 1, 10, '31415926535'], ['B', 2, 14, ''], ['A', 1, 15, '23816326412'], ['D', 2, 11, 0], ] df1 = Table(cols1, rows1) cols2 = ['type', 'size', 'cost'] rows2 = [ ['A', 2, 3.1415], ['B', 1, 2.7182], ['C', 2, 1.2345], ['D', 1, 1.2345], ] df2 = Table(cols2, rows2) cnt_inner = 3 cnt_left = 3 cnt_right = 1 class TestLemurasTableMerge(unittest.TestCase): def test_inner(self): df3 = Table.merge(df1, df2, ['type', 'size'], 'inner')
def test_load_results(self):
df2 = Table.from_sql_result(sql_result)
self.assertEqual(df2.columns, ['id', 'name', 'q1', 'q2', 'q3'])
self.assertEqual(df2.rowcnt, 8)
self.assertEqual(df2['id'].sum(), 19169)
def test_save(self): txt = Table.from_html(html_table).html() df2 = Table.from_html(txt, title='Numbers') self.assertEqual(df2.columns, ['Name', 'Value']) self.assertEqual(df2.rowcnt, 3) self.assertEqual(df2['Value'].sum(), 7.4779084706)
import unittest
from lemuras import Table
cols = ['x', 'y', 'val']
rows = [
[0, 'c', 3],
[2, 'b', 7],
[1, 'a', 2],
[4, 'c', 2],
]
df1 = Table(cols, rows)
cols = ['x', 'y', 'val']
rows = [
[1, 'a', 3],
[0, 'c', 3],
[2, 'b', 7],
[0, 'c', 1],
[1, 'a', 2],
[4, 'c', 2],
]
df2 = Table(cols, rows)
class TestLemurasTablePivot(unittest.TestCase):
def test_pivot_main(self):
res = df1.pivot('x', 'y', 'val', 0)
self.assertEqual(res.colcnt, df1['x'].nunique() + 1)
self.assertEqual(res.rowcnt, df1['y'].nunique())
self.assertEqual(
res.loc(res['y'] == 'c').row().sum(),
import unittest from lemuras import Table cols = ['type', 'size', 'weight', 'tel'] rows = [ ['A', 1, 10, '+79360360193'], ['B', 4, 12, 84505505151], ] df1 = Table(cols, rows, 'First') cols = ['type', 'size', 'weight', 'tel'] rows = [ ['A', 2, 10, '31415926535'], ['B', 6, 14, ''], ['A', 3, 12, 0], ] df2 = Table(cols, rows, 'Second') class TestLemurasTableConcat(unittest.TestCase): def test_concat(self): df3 = Table.concat([df1, df2]) self.assertEqual(df1.columns, df3.columns) self.assertEqual(df3.rowcnt, df1.rowcnt+df2.rowcnt) self.assertEqual(df3['size'].sum(), df1['size'].sum()+df2['size'].sum()) df1.append(df2) # Now, df1 must be equal to df3 self.assertEqual(df1.columns, df2.columns) self.assertEqual(df1.rowcnt, df3.rowcnt) self.assertEqual(df1['size'].sum(), df3['size'].sum())