Remove Python 2 support and modernize required libraries.

.gitignore

@@ -15,3 +15,5 @@ eval*
 .pytest_cache
 .tox
 docs_api
+.idea
+venv

examples/advanced.py

@@ -1,5 +1,3 @@
-from __future__ import print_function
-
 from koala.ExcelCompiler import ExcelCompiler
 from koala.Spreadsheet import Spreadsheet
 from koala.excellib import xsum

examples/basic.py

@@ -1,5 +1,3 @@
-from __future__ import print_function
-
 from koala.ExcelCompiler import ExcelCompiler
 from koala.Spreadsheet import Spreadsheet
 

koala/Cell.py

@@ -1,13 +1,9 @@
 # cython: profile=True
 
-from __future__ import absolute_import, division
-
 from koala.CellBase import CellBase
 from koala.Range import RangeCore
 from koala.utils import *
 
-from openpyxl.compat import unicode
-
 
 class Cell(CellBase):
     ctr = 0
@@ -62,11 +58,7 @@ def __init__(
             self.__row = None
             self.__col_idx = None
 
-        # `unicode` != `str` in Python2. See `from openpyxl.compat import unicode`
-        if type(formula) == str and str != unicode:
-            self.__formula = unicode(formula, 'utf-8') if formula else None
-        else:
-            self.__formula = formula if formula else None
+        self.__formula = formula if formula else None
 
         self.__value = value
         self.python_expression = None

koala/ExcelCompiler.py

@@ -1,15 +1,5 @@
-from __future__ import print_function
 # cython: profile=True
 
-import os.path
-
-import networkx
-
-from koala.reader import read_archive, read_named_ranges, read_cells
-from koala.utils import *
-from koala.ast import graph_from_seeds, shunting_yard, build_ast, prepare_pointer
-from koala.Cell import Cell
-from koala.Range import RangeFactory
 from koala.Spreadsheet import Spreadsheet
 import warnings
 

koala/Range.py

@@ -1,12 +1,7 @@
-from __future__ import absolute_import, division, print_function
-
 from koala.CellBase import CellBase
 from koala.ExcelError import ErrorCodes, ExcelError
 from koala.utils import *
 
-from openpyxl.compat import unicode
-
-
 # WARNING: Range should never be imported directly. Import Range from excelutils instead.
 
 ### Range Utils ###
@@ -58,7 +53,7 @@ def check_value(a):
         return ExcelError(a)
 
     try:  # This is to avoid None or Exception returned by Range operations
-        if isinstance(a, (unicode, str)):
+        if isinstance(a, str):
             return a
         elif float(a):
             return a
@@ -570,7 +565,7 @@ def multiply(a, b):
     @staticmethod
     def divide(a, b):
         try:
-            return old_div(float(check_value(a)), float(check_value(b)))
+            return float(check_value(a)) / float(check_value(b))
         except Exception as e:
             return ExcelError('#DIV/0!', e)
 
@@ -584,9 +579,9 @@ def power(a, b):
     @staticmethod
     def is_equal(a, b):
         try:
-            if not isinstance(a, (str, unicode)):
+            if not isinstance(a, str):
                 a = check_value(a)
-            if not isinstance(b, (str, unicode)):
+            if not isinstance(b, str):
                 b = check_value(b)
 
             return is_almost_equal(a, b, precision=0.00001)
@@ -596,9 +591,9 @@ def is_equal(a, b):
     @staticmethod
     def is_not_equal(a, b):
         try:
-            if not isinstance(a, (str, unicode)):
+            if not isinstance(a, str):
                 a = check_value(a)
-            if not isinstance(a, (str, unicode)):
+            if not isinstance(a, str):
                 b = check_value(b)
 
             return a != b

koala/Spreadsheet.py

@@ -1,4 +1,3 @@
-from __future__ import absolute_import, print_function
 # cython: profile=True
 
 from koala.Range import get_cell_address, parse_cell_address
@@ -17,9 +16,6 @@
 import networkx
 from networkx.readwrite import json_graph
 
-from openpyxl.compat import unicode
-
-
 class Spreadsheet(object):
     def __init__(self, file=None, ignore_sheets=[], ignore_hidden=False, debug=False):
         # print("___### Initializing Excel Compiler ###___")
@@ -56,7 +52,7 @@ def __init__(self, file=None, ignore_sheets=[], ignore_hidden=False, debug=False
             else:                       # assume file path
                 archive = read_archive(os.path.abspath(file))
             # Parse cells
-            self.cells = read_cells(archive, ignore_sheets, ignore_hidden)
+            self.cells, self.sheets = read_cells(archive, ignore_sheets, ignore_hidden)
             # Parse named_range { name (ExampleName) -> address (Sheet!A1:A10)}
             self.named_ranges = read_named_ranges(archive)
             self.range = RangeFactory(self.cells)
@@ -306,7 +302,7 @@ def cell_set_formula(self, address, formula):
             for index, c in enumerate(cell.range.cells): # for each cell of the range, translate the formula
                 if index == 0:
                     c.formula = formula
-                    translator = Translator(unicode('=' +    formula), c.address().split('!')[1]) # the Translator needs a reference without sheet
+                    translator = Translator('=' + formula, c.address().split('!')[1]) # the Translator needs a reference without sheet
                 else:
                     translated = translator.translate_formula(c.address().split('!')[1]) # the Translator needs a reference without sheet
                     c.formula = translated[1:] # to get rid of the '='

koala/__init__.py

@@ -1,5 +1,3 @@
-from __future__ import absolute_import
-
 from openpyxl import *
 from .ast import *
 from .Cell import *

koala/ast/__init__.py

@@ -1,12 +1,9 @@
-from __future__ import absolute_import
 # cython: profile=True
 
 import collections
-import six
 
 import networkx
 from networkx.classes.digraph import DiGraph
-from openpyxl.compat import unicode
 
 from koala.utils import uniqueify, flatten, max_dimension, col2num, resolve_range
 from koala.Cell import Cell
@@ -118,7 +115,7 @@ def shunting_yard(expression, named_ranges, ref = None, tokenize_range = False):
         for index, token in enumerate(tokens):
             new_tokens.append(token)
 
-            if type(token.tvalue) == str or type(token.tvalue) == unicode:
+            if type(token.tvalue) == str:
 
                 if token.tvalue.startswith(':'): # example -> :OFFSET( or simply :A10
                     depth = 0
@@ -139,7 +136,7 @@ def shunting_yard(expression, named_ranges, ref = None, tokenize_range = False):
                         if depth == 0:
                             new_tokens.pop() # these 2 lines are needed to remove INDEX()
                             new_tokens.pop()
-                            expr = six.next(rev).tvalue + expr
+                            expr = next(rev).tvalue + expr
                             break
 
                     expr += token.tvalue
@@ -377,13 +374,11 @@ def cell2code(cell, named_ranges):
 
     else:
         ast = None
-        if isinstance(cell.value, unicode):
-            code = u'u"' + cell.value.replace(u'"', u'\\"') + u'"'
-        elif isinstance(cell.value, str):
-            raise RuntimeError("Got unexpected non-unicode str")
+        if isinstance(cell.value, str):
+            code = '"' + cell.value.replace('"', r'\"') + '"'
         else:
             code = str(cell.value)
-    return code,ast
+    return code, ast
 
 
 def prepare_pointer(code, names, ref_cell = None):

koala/ast/astnodes.py

@@ -1,25 +1,21 @@
-from __future__ import print_function
 # cython: profile=True
 
 from networkx import NetworkXError
 
-from openpyxl.compat import unicode
-
 from koala.excellib import FUNCTION_MAP, IND_FUN
 from koala.utils import is_range, split_range, split_address, resolve_range
 from koala.ExcelError import *
 
 
 def to_str(my_string):
-    # `unicode` != `str` in Python2. See `from openpyxl.compat import unicode`
-    if type(my_string) == str and str != unicode:
-        return unicode(my_string, 'utf-8')
-    elif type(my_string) == unicode:
+    if isinstance(my_string, bytes):
+        return my_string.decode("utf-8")
+    elif isinstance(my_string, str):
         return my_string
     else:
         try:
             return str(my_string)
-        except:
+        except Exception:
             print('Couldnt parse as string', type(my_string))
             return my_string
     # elif isinstance(my_string, (int, float, tuple, Ra):
@@ -45,7 +41,7 @@ def __getattr__(self, name):
     def children(self, ast):
         try:
             args = ast.predecessors(self)
-            args = sorted(args, key=lambda x: ast.node[x]['pos'])
+            args = sorted(args, key=lambda x: ast.nodes[x]['pos'])
         except NetworkXError:
             args = ''
         return args

koala/excellib.py

@@ -6,10 +6,9 @@
 
 # source: https://github.com/dgorissen/pycel/blob/master/src/pycel/excellib.py
 
-from __future__ import absolute_import, division
-
 import itertools
 import numpy as np
+import numpy_financial as npf
 import scipy.optimize
 import datetime
 import random
@@ -18,8 +17,6 @@
 from calendar import monthrange
 from dateutil.relativedelta import relativedelta
 
-from openpyxl.compat import unicode
-
 from koala.utils import *
 from koala.Range import RangeCore as Range
 from koala.ExcelError import *
@@ -427,7 +424,7 @@ def irr(values, guess = None):
         raise ValueError('guess value for excellib.irr() is %s and not 0' % guess)
     else:
         try:
-            return np.irr(values)
+            return npf.irr(values)
         except Exception as e:
             return ExcelError('#NUM!', e)
 
@@ -729,7 +726,7 @@ def randbetween(bottom, top):
 
 def right(text,n):
     #TODO: hack to deal with naca section numbers
-    if isinstance(text, unicode) or isinstance(text,str):
+    if isinstance(text, str):
         return text[-n:]
     else:
         # TODO: get rid of the decimal

koala/reader.py

@@ -1,5 +1,3 @@
-from __future__ import print_function
-
 from io import BytesIO
 import re
 import os
@@ -108,6 +106,8 @@ def read_cells(archive, ignore_sheets = [], ignore_hidden = False):
 
     cells = {}
 
+    sheets = []
+
     functions = set()
 
     cts = dict(read_content_types(archive))
@@ -127,6 +127,9 @@ def read_cells(archive, ignore_sheets = [], ignore_hidden = False):
 
         if sheet_name in ignore_sheets: continue
 
+        if sheet_name not in sheets:
+            sheets.append(sheet_name)
+
         root = fromstring(archive.read(sheet['path'])) # it is necessary to use cElementTree from xml module, otherwise root.findall doesn't work as it should
 
         hidden_cols = False
@@ -225,7 +228,7 @@ def read_cells(archive, ignore_sheets = [], ignore_hidden = False):
     #     if f not in existing:
     #         print('== Missing function: %s' % f)
 
-    return cells
+    return cells, sheets
 
 
 def read_rels(archive):

koala/serializer.py

@@ -1,13 +1,9 @@
-from __future__ import absolute_import, print_function
-
 import json
 import gzip
 import networkx
 
 from networkx.classes.digraph import DiGraph
-from networkx.readwrite import json_graph
 from networkx.drawing.nx_pydot import write_dot
-from openpyxl.compat import unicode
 
 from koala.Cell import Cell
 from koala.Range import RangeCore, RangeFactory
@@ -51,10 +47,10 @@ def parse_cell_info(cell):
     for cell in simple_cells:
         parse_cell_info(cell)
         value = cell.value
-        if isinstance(value, unicode):
+        if isinstance(value, str):
             outfile.write(cell.value.encode('utf-8') + b"\n")
         else:
-            outfile.write((str(cell.value) + u"\n").encode('utf-8'))
+            outfile.write((str(cell.value) + "\n").encode('utf-8'))
         outfile.write(b"====" + b"\n")
 
     outfile.write(b"-----" + b"\n")
@@ -205,41 +201,15 @@ def dump_json(self, fname):
 
 def load_json(fname):
 
-    def _decode_list(data):
-        rv = []
-        for item in data:
-            if isinstance(item, unicode) and unicode != str:
-                item = item.encode('utf-8')
-            elif isinstance(item, list) and unicode != str:
-                item = _decode_list(item)
-            elif isinstance(item, dict):
-                item = _decode_dict(item)
-            rv.append(item)
-        return rv
-
-    def _decode_dict(data):
-        rv = {}
-        for key, value in data.items():
-            if isinstance(key, unicode) and unicode != str:
-                key = key.encode('utf-8')
-            if isinstance(value, unicode) and unicode != str:
-                value = value.encode('utf-8')
-            elif isinstance(value, list):
-                value = _decode_list(value)
-            elif isinstance(value, dict):
-                value = _decode_dict(value)
-            rv[key] = value
-        return rv
-
     with gzip.GzipFile(fname, 'r') as infile:
-        data = json.loads(infile.read().decode('utf-8'), object_hook=_decode_dict)
+        data = json.loads(infile.read().decode('utf-8'))
 
     return data
 
 
 ########### based on dot #################
-def export_to_dot(self,fname):
-    write_dot(self.G,fname)
+def export_to_dot(self, fname):
+    write_dot(self.G, fname)
 
 
 ########### plotting #################