#!/usr/bin/env python
# -*- coding: UTF-8 -*-
#
# Copyright 2016-2018 European Commission (JRC);
# Licensed under the EUPL (the 'Licence');
# You may not use this work except in compliance with the Licence.
# You may obtain a copy of the Licence at: http://ec.europa.eu/idabc/eupl
"""
It provides Operand classes.
"""
# noinspection PyCompatibility
import regex
import functools
import schedula as sh
from . import Token
from ..errors import TokenError
from .parenthesis import _update_n_args
maxcol = 16384
maxrow = 1048576
[docs]class XlError(sh.Token):
pass
NULL = XlError('#NULL!')
DIV = XlError('#DIV/0!')
VALUE = XlError('#VALUE!')
REF = XlError('#REF!')
NUM = XlError('#NUM!')
NAME = XlError('#NAME?')
NA = XlError('#N/A')
[docs]class Operand(Token):
[docs] def ast(self, tokens, stack, builder):
if tokens and isinstance(tokens[-1], Operand):
raise TokenError()
super(Operand, self).ast(tokens, stack, builder)
builder.append(self)
_update_n_args(stack)
[docs]class String(Operand):
_re = regex.compile(r'^\s*"(?P<name>(?>""|[^"])*)"\s*')
[docs] def compile(self):
return self.name.replace('""', '"')
[docs]class Error(Operand):
_re = regex.compile(
r'^\s*(?P<name>\#(?>NULL!|DIV/0!|VALUE!|REF!|NUM!|NAME\?|N/A))\s*',
regex.IGNORECASE
)
errors = {str(k): k for k in (NULL, DIV, VALUE, REF, NUM, NAME, NA)}
[docs] def compile(self):
return self.errors[self.name]
[docs]class Number(Operand):
_re = regex.compile(
r'^\s*(?P<name>[0-9]+(?>\.[0-9]+)?(?>E[+-][0-9]+)?|TRUE|FALSE)'
r'(?!([a-z]|[0-9]|\.|\s*\:))\s*',
regex.IGNORECASE
)
[docs] def compile(self):
return eval(self.name.capitalize())
_re_range = r"""
(?>
(?>
(?>
'(\[(?P<excel>[^\[\]]+)\])?
(?P<sheet>(?>''|[^\?!*\/\[\]':"])+)?'
|
(\[(?P<excel_id>[0-9]+)\])(?P<sheet>[^\W\d][\w\.]*)
|
(?P<sheet>[^\W\d][\w\.]*)
|
'(?P<sheet>(?>''|[^\?!*\/\[\]':"])+)'
)!
)?
(?>
(?>
(?>
\$?(?P<c1>[A-Z]+)?\$?(?P<r1>[1-9]\d*)?
(?>:\$?(?P<c2>[A-Z]+)?)(\$?(?P<r2>[1-9]\d*)?)?
)
|
\$?(?P<c1>[A-Z]+)\$?(?P<r1>[1-9]\d*)
|
\$?(?P<c1>[A-Z]+):\$?(?P<c2>[A-Z]+)
|
\$?(?P<r1>[1-9]\d*):\$?(?P<r2>[1-9]\d*)
)
|
(?>
(?>
R(?P<r1>[1-9]\d*)C(?P<n1>[1-9]\d*)
(?>:R(?P<r2>[1-9]\d*)C(?P<n2>[1-9]\d*))?
)
|
R(?P<r1>[1-9]\d*)C(?P<n1>[1-9]\d*)
|
R(?P<r1>[1-9]\d*):R(?P<r2>[1-9]\d*)
|
C(?P<n1>[1-9]\d*):C(?P<n2>[1-9]\d*)
)
|
(?P<ref>[A-Z_\\]+[A-Z0-9\.\_]*)
)
|
(?>
(?>
R\[(?P<rr1>[\+-]?[1-9]\d*)\]C\[(?P<rc1>[\+-]?[1-9]\d*)\]
(?>:R\[(?P<rr2>[\+-]?[1-9]\d*)\]C\[(?P<rc2>[\+-]?[1-9]\d*)\])?
)
|
R\[(?P<rr1>[\+-]?[1-9]\d*)\]C\[(?P<rc1>[\+-]?[1-9]\d*)\]
|
R\[(?P<rr1>[\+-]?[1-9]\d*)\]:R\[(?P<rr2>[\+-]?[1-9]\d*)\]
|
C\[(?P<rc1>[\+-]?[1-9]\d*)\]:C\[(?P<rc2>[\+-]?[1-9]\d*)\]
)
)
(?![\(\w])
"""
_re_range = regex.compile(
r'^(?>(?P<indirect>INDIRECT\("{0}?"\))|{0})'.format(_re_range),
regex.IGNORECASE | regex.X | regex.DOTALL
)
def _index2col(index):
index = int(index) - 1
if index < 0:
return ''
d = index // 26
chr1 = _index2col(d) if d > 0 else ''
return '%s%s' % (chr1, chr(ord('A') + index % 26))
def _col2index(col):
index, power = 0, 1
for ch in col.upper()[::-1]:
index += (ord(ch) - ord('A') + 1) * power
power *= 26
return index
@functools.lru_cache(None)
def _maxcol():
return _index2col(maxcol)
@functools.lru_cache(None)
def _maxrow():
return str(maxrow)
def _build_cel(c, r):
r = str(r and int(r) or '')
return c != _maxcol() and c or '', r != _maxrow() and r or ''
def _build_ref(c1, r1, c2, r2):
(c1, r1), v2 = _build_cel(c1, r1), '{}{}'.format(*_build_cel(c2, r2))
v1 = '{}{}'.format(c1, r1)
if v1 == v2 and c1 and r1:
if v1:
return v1
raise ValueError()
return '%s:%s' % (v1, v2)
_re_build_id = regex.compile(r'^[0-9]+$')
def _build_id(ref, sheet='', excel=''):
if excel:
sheet = "[%s]%s" % (excel, sheet.replace("''", "'"))
if not _re_build_id.match(excel):
sheet = "'%s'" % sheet
return '!'.join(s for s in (sheet, ref) if s)
def _sum(*args):
return sum(map(float, args))
@functools.lru_cache(None)
def _range2parts(inputs, outputs):
dsp = sh.Dispatcher(raises=True)
dsp.add_data(data_id='cr', default_value='1')
dsp.add_data(data_id='cc', default_value='1')
dsp.add_function('relative2absolute', _sum, ['cr', 'rr1'], ['r1'])
dsp.add_function('relative2absolute', _sum, ['cc', 'rc1'], ['n1'])
dsp.add_function('relative2absolute', _sum, ['cr', 'rr2'], ['r2'])
dsp.add_function('relative2absolute', _sum, ['cc', 'rc2'], ['n2'])
dsp.add_function(function=_index2col, inputs=['n1'], outputs=['c1'])
dsp.add_function(function=_index2col, inputs=['n2'], outputs=['c2'])
dsp.add_function(function=_col2index, inputs=['c1'], outputs=['n1'])
dsp.add_function(function=_col2index, inputs=['c2'], outputs=['n2'])
dsp.add_function(function=sh.bypass, inputs=['c1'], outputs=['c2'])
dsp.add_function(function=sh.bypass, inputs=['r1'], outputs=['r2'])
dsp.add_function(function=lambda x, y: x[y],
inputs=['external_links', 'excel_id'], outputs=['excel'])
dsp.add_function(function=sh.bypass, weight=1,
inputs=['excel_id'], outputs=['excel'])
dsp.add_data(data_id='excel', filters=(str.upper,))
dsp.add_data(data_id='sheet', default_value='', filters=(str.upper,))
dsp.add_data(data_id='ref', filters=(str.upper,))
dsp.add_data(data_id='name', filters=(str.upper,))
dsp.add_data(data_id='n1', default_value=0, initial_dist=100)
dsp.add_data(data_id='r1', default_value='0', initial_dist=100)
dsp.add_data(data_id='c2', default_value=_maxcol(), initial_dist=100)
dsp.add_data(data_id='r2', default_value=_maxrow(), initial_dist=100)
dsp.add_function(None, _build_ref, ['c1', 'r1', 'c2', 'r2'], ['ref'])
dsp.add_function(None, _build_id, ['ref', 'sheet', 'excel'], ['name'])
func = sh.DispatchPipe(dsp, '', inputs, outputs)
func.output_type = 'all'
return func
_keys = {'r1', 'r2', 'excel', 'c1', 'c2', 'n1', 'n2', 'sheet', 'ref', 'name'}
[docs]def fast_range2parts(**kw):
inputs = sh.selector(_keys, kw, allow_miss=True)
for func in (fast_range2parts_v1, fast_range2parts_v2, fast_range2parts_v3):
try:
return sh.combine_dicts(kw, base=func(**inputs))
except TypeError:
pass
else:
raise ValueError
[docs]def fast_range2parts_v1(r1, c1, excel, sheet=''):
n1, sheet = _col2index(c1), sheet.upper()
ref, excel = '{}{}'.format(*_build_cel(c1, r1)).upper(), excel.upper()
return {
'r1': r1, 'r2': r1, 'c1': c1, 'c2': c1, 'n1': n1, 'n2': n1, 'ref': ref,
'sheet': sheet, 'name': _build_id(ref, sheet, excel), 'excel': excel
}
[docs]def fast_range2parts_v2(r1, c1, r2, c2, excel, sheet=''):
sheet = sheet.upper()
ref, excel = _build_ref(c1, r1, c2, r2).upper(), excel.upper()
return {
'r1': r1, 'r2': r2, 'c1': c1, 'c2': c2, 'n1': _col2index(c1),
'n2': _col2index(c2), 'ref': ref, 'sheet': sheet,
'name': _build_id(ref, sheet, excel), 'excel': excel
}
[docs]def fast_range2parts_v3(r1, n1, r2, n2, excel, sheet=''):
sheet = sheet.upper()
c1, c2 = _index2col(n1), _index2col(n2)
ref, excel = _build_ref(c1, r1, c2, r2).upper(), excel.upper()
return {
'r1': r1, 'r2': r2, 'c1': c1, 'c2': c2, 'n1': n1, 'n2': n2, 'ref': ref,
'sheet': sheet, 'name': _build_id(ref, sheet, excel), 'excel': excel
}
[docs]def range2parts(outputs, **inputs):
if 'excel_id' in inputs:
inputs.pop('excel', None)
elif 'excel' not in inputs:
inputs['excel'] = ''
try:
return fast_range2parts(**inputs)
except ValueError:
keys = tuple(sorted(inputs))
outputs = outputs and tuple(outputs) or outputs
return dict(_range2parts(keys, outputs)(*(inputs[k] for k in keys)))
[docs]class Range(Operand):
_re = _re_range
[docs] def process(self, match, context=None):
d = super(Range, self).process(match)
if len(d) <= 1 and 'indirect' not in d and 'ref' in d:
try:
from .function import Function
if Function(self.source).name == d['ref']:
return {}
except TokenError:
pass
if 'ref' in d:
self.attr['is_reference'] = True
return range2parts(None, **sh.combine_dicts(context or {}, d))
def __repr__(self):
if self.attr.get('is_ranges', False):
from ..ranges import Ranges
return '{} <{}>'.format(self.name, Ranges.__name__)
return super(Range, self).__repr__()
[docs] def compile(self):
if self.attr.get('is_ranges', False):
from ..ranges import Ranges
return Ranges().push(self.attr['name'])
return sh.EMPTY