Get one liner formula from graph node

[dberardo-com]

It would be useful to get a one liner string representing a DAG from the dispatcher.

so basically having one call like "dsp.get_one_liner(origin = 'A2')" which basically returns all subnodes of the graph starting from A2. Example:

A2: "=SUM(A1,A3)"
A3: "=SUM(1,3)"
A1: "=A4"
A4: "=SUM(1,1,1,1,1)"

dsp.get_one_liner(origin = 'A2') --> returns: "=SUM(SUM(1,1,1,1,1),SUM(1,3)"
dsp.get_one_liner(origin = 'A1') --> returns: "=SUM(1,1,1,1,1)"

and so on...

[vinci1it2000]

You can use the following script to extract the online formula.

>>> import functools
>>> import schedula as sh

>>> xl_model = formulas.ExcelModel().from_dict({
...     "A3": 1,
...     "A2": "=SUM(A4,A13)",
...     "A13": "=SUM(1,3)",
...     "A1": "=A4",
...     "A4": "=SUM(A3,1,1,1,1)"
... })

>>> @functools.lru_cache()
... def get_one_line_formula(xl_model, node):
...     try:
...         formula = next(iter(xl_model.dsp.dmap.pred[node]))
...     except StopIteration:
...         value = xl_model.dsp.default_values[node]['value']
...         return str(value)
...     for k in sorted(xl_model.dsp.dmap.pred[formula], key=lambda x: (len(x), x),
...                     reverse=True):
...         if not isinstance(k, sh.Token):
...             formula = formula.replace(k, get_one_line_formula(xl_model, k))
...     return formula[1:]

>>> get_one_line_formula(xl_model, 'A1')
'SUM(1, 1, 1, 1, 1)'

>>> get_one_line_formula(xl_model, 'A2')
'SUM(SUM(1, 1, 1, 1, 1), SUM(1, 3))'

[dberardo-com]

i guess the function above does not work with ranges, right ? like A1:C3

[nopria]

In #142 I posted some code that expands ranges, hope you can test it for your case.

[dberardo-com]

thanks for the mutual cooperation, i am kind of curious to see what do you use this formula for, as i believe i have quite a different use case for it :D

will have a deeper look at your code over there to figure that out

[nopria]

i guess the function above does not work with ranges, right ? like A1:C3

Actually it seems that ranges are handled, or at least xl_model.dsp.dmap.pred dictionary contains the expantions of ranges and also takes care of unused cells. For example in xl_model.dsp.dmap.pred dictionary of my Excel document I just noted the item

{"='[test.xlsx]WS'!M10:N14": {"'[test.xlsx]WS'!M10": {}, "'[test.xlsx]WS'!M11": {}, "'[test.xlsx]WS'!M12": {}, self: {}}}

which is precisely the expantion of range M10:N14 to the only 3 cells of the entire range used in my document.