将 xlsx 文件读入 python 而不会丢失信息

Question

The goal is to read in an xlsx file into python. All information provided is just to give context to the problem.

In these examples, the date read in should be 2004/07/01

How the data looks in "LibreCalc"

Here's what librecalc displays

12/30/99
12/30/99
...

Here's what the actual cell formulas are defined as.

=Date(2004,7,1)
=Date(2004,7,1)
...

Pandas

CASE 1 ( pd.read_excel )

Displayed results

00:00:00
00:00:00
...

values function returns

array([datetime.time(0, 0), datetime.time(0, 0), datetime.time(0, 0), ...,
       datetime.time(0, 0), datetime.time(0, 0), datetime.time(0, 0)],
      dtype=object)

CASE 2 ( pd.read_excel(..., parse_date)

Displayed results

2020-04-14
2020-04-14
...

values function returns

array(['2020-04-14T00:00:00.000000000', '2020-04-14T00:00:00.000000000',
       '2020-04-14T00:00:00.000000000', ...,
       '2020-04-14T00:00:00.000000000', '2020-04-14T00:00:00.000000000',
       '2020-04-14T00:00:00.000000000'], dtype='datetime64[ns]')

Which is today's date, not 2004/7/1

---

So the problem seems to beyond a simply reformatting the values. It looks python doesn't have any idea how to parse the dates.

---

Answer 1

my dudes, if I run "find and replace" and just find "=Date(" and replace it with "=Date(" then everything works. I have like 40 files I need to do that to but it's manageable. That's insane. Why in the world would this work?