LibreOffice Calc Python UDF 的基本包装器

Question

如何使用非整数电子表格范围形成对运行 LibreOffice Calc Python 用户定义函数的 Basic 包装器调用？

我不清楚如何为 values 和 excel_date 数组正确声明函数，这两个数组都是实数数组，在我理解的基本术语中键入 Variant。

我关注了Calling a python function from within LibreCalc and ranges can be passed as integer arguments without defining them but these are real values. Both arrays are single-dimension so the multi-dimensional array issues do not apply as discussed in How can I call a Python macro in a cell formula in OpenOffice.Org Calc?

基本代码为：

Function xnpv_helper(rate as Double, values() as Variant, excel_date() as Variant) as Double
    Dim scriptPro As Object, myScript As Object
    scriptPro = ThisComponent.getScriptProvider()
    myScript = scriptPro.getScript("vnd.sun.star.script:MyPythonLibrary/Develop/math.py$xnpv_helper?language=Python&location=user")
    xnpv_helper = myScript.invoke(Array(rate, values, excel_date), Array(), Array() )
end function

电子表格数据：

Date    Amount  Rate    xnpv
31/12/19    100 -0.1    
31/12/20    -110

xnpv_helper 在底部。实际 python 脚本基于 https://github.com/tarioch/xirr from financial python library that has xirr and xnpv function?

# Demo will run in python REPL - uncomment last line
import scipy.optimize

DAYS_PER_YEAR = 365.0

def xnpv(valuesPerDate, rate):
    '''Calculate the irregular net present value.

    >>> from datetime import date
    >>> valuesPerDate = {date(2019, 12, 31): -100, date(2020, 12, 31): 110}
    >>> xnpv(valuesPerDate, -0.10)
    22.257507852701295
    '''
    if rate == -1.0:
        return float('inf')
    t0 = min(valuesPerDate.keys())
    if rate <= -1.0:
        return sum([-abs(vi) / (-1.0 - rate)**((ti - t0).days / DAYS_PER_YEAR) for ti, vi in valuesPerDate.items()])
    return sum([vi / (1.0 + rate)**((ti - t0).days / DAYS_PER_YEAR) for ti, vi in valuesPerDate.items()])


from datetime import date
def excel2date(excel_date):
    return date.fromordinal(date(1900, 1, 1).toordinal() + int(excel_date) - 2)

def xnpv_helper(rate, values, excel_date):
    dates = [excel2date(i) for i in excel_date]
    valuesPerDate = dict(zip(dates, values))
    return xnpv(valuesPerDate, rate)

# valuesPerDate = {date(2019, 12, 31): -100, date(2020, 12, 31): 110}
# xnpv_helper(-0.10, [-100, 110], [43830.0, 44196.0])

Answer 1

感谢 @arturaz for explaining the Basic array comes to python as a tuple and needs converting to a list (best match for an array in python). Further details are given in his answer at 建立在问题和其他答案的基础上。

APSO python 控制台显示：

什么从 Basic 解析为 python
需要什么
需要带日期格式的数据

APSO python console [LibreOffice]
3.8.10 (default, Aug 10 2021, 19:39:20) [MSC v.1928 64 bit (AMD64)]
>>> 
1: ((43830.0,), (44196.0,)) ((100.0,), (-110.0,)) -0.1
2: [43830.0, 44196.0] [-100, 110] -0.1
3: {datetime.date(2019, 12, 31): -100, datetime.date(2020, 12, 31): 110} -0.1

这两个数组在 python 辅助函数中使用之前被重新格式化：

def xnpv_helper(rate, values, excel_date):
    excel_date0 = [a for b in excel_date for a in b]
    values0 = [a for b in values for a in b]
    dates = [excel2date(i) for i in excel_date0]
    valuesPerDate = dict(zip(dates, values0))
    return xnpv(valuesPerDate, rate)

LibreOffice Calc Python UDF 的基本包装器

Basic Wrapper for LibreOffice Calc Python UDF

python

libreoffice

openoffice.org

libreoffice-calc

libreoffice-basic