read_excel reads big numbers incorrectly when they use a cell reference

[breanna-gream]

Checks

• I have checked that this issue has not already been reported.
• I have confirmed this bug exists on the latest version of Polars.

Reproducible example

import polars as pl

df1 = pl.read_excel("test.xlsx", sheet_name="Sheet1")
df2 = pl.read_excel("test.xlsx", sheet_name="Sheet2")

df1.item(), df2.item()

# Returns
# (-8086931554011838357, -8086931554011838464)

The contents of test.xlsx (also attached below) are:

col1
'-8086931554011838357

in Sheet 1 (note the apostrophe to store this number as text). And then Sheet2 looks the same, but the number references Sheet1!A2 instead of being a static value.

test.xlsx

Log output

dataframe filtered
dataframe filtered

Issue description

I think this might be an issue with the calamine engine? Using the xlsx2csv engine seems to fix it, however I would prefer not to use it as I have found that calamine handles floating point precision better.
Apologies if this was not the correct place to raise this - please let me know.

Expected behavior

Returns correct number even if the cell uses a formula instead of a static value. I.e. output would be:

(-8086931554011838357, -8086931554011838357)

Installed versions

--------Version info---------
Polars:              1.9.0
Index type:          UInt32
Platform:            Windows-10-10.0.19045-SP0
Python:              3.11.9 (tags/v3.11.9:de54cf5, Apr  2 2024, 10:12:12) [MSC v.1938 64 bit (AMD64)]

----Optional dependencies----
adbc_driver_manager  <not installed>
altair               <not installed>
cloudpickle          <not installed>
connectorx           <not installed>
deltalake            <not installed>
fastexcel            0.11.6
fsspec               <not installed>
gevent               <not installed>
great_tables         <not installed>
matplotlib           <not installed>
nest_asyncio         <not installed>
numpy                2.1.1
openpyxl             <not installed>
pandas               <not installed>
pyarrow              17.0.0
pydantic             <not installed>
pyiceberg            <not installed>
sqlalchemy           <not installed>
torch                <not installed>
xlsx2csv             0.8.3
xlsxwriter           <not installed>

[PrettyWood]

@breanna-gream It indeed comes from calamine (from https://github.com/tafia/calamine/blob/master/src/xlsx/cells_reader.rs#L372 to be precise)
I don't have enough xml knowledge to understand exactly why this logic is here. Probably best to open an issue on calamine side

EDIT: I opened tafia/calamine#472

[breanna-gream]

@PrettyWood looks like your PR tafia/calamine#472 has been merged (thank you for creating that!)

However I have tested again based on my original example (after re-installing polars[calamine]) and am still seeing the same issue - is there something else I need to do to see this fix flow through?

[PrettyWood]

We need to wait for a new calamine release. Once it's out we'll bump calamine in fastexcel and create a new fastexcel release.

[alexander-beedie]

FYI: closing this as it's not a Polars-specific issue (fix will come from upstream).