AttributeError: 'ElementTree' object has no attribute 'getiterator' when trying to import excel file

You could try to use an argument engine="openpyxl". It helped me to resolve the same problem.

The error occurs when pandas is used in python3.9+ because the code xml.etree.ElementTree.Element.getiterator() which had been deprecated with a warning previously, has now been removed.

A workaround is to install another engine openpyxl to read the excel file, and replace your code which reads the excel file.

First,

pip3 install openpyxl

Then, instead of pd.read_excel('student-results.xlsx'), write pd.read_excel('student-results.xlsx', engine='openpyxl')

Reference: Python bug tracker

I got the same error with xlrd (1.2.0) or xlrd3 (1.0.0) without pandas, but with Python 3.9. The following may interest those looking for an explanation:

It only happened when defusedxml was available (in that case, xlrd will use it). But it could be worked around, without changing any of the involved libraries:

import xlrd
xlrd.xlsx.ensure_elementtree_imported(False, None)
xlrd.xlsx.Element_has_iter = True

The second line ensures that Element_has_iter will not be reset when opening a workbook, so that it remains to True - as set in the 3rd line. When this is done, xlrd uses iter instead of crashing on the missing getiterator.

That said, I agree that moving to openpyxl in place of xlrd is a cleaner solution, at least untill xlrd or xlrd3 possibly gets fixed. Openpyxl appears to be more actively developed. In my case, I have to adapt direct calls to those libraries, it is probably more work than just typing openpyxl instead of xlrd to tell pandas about what it should do, but I'll consider it.

So ok with @corridda, use openpyxl, and others are right about the cause, but maybe this explains a little more on the causes.

You can check issue description here. You run Python 3.9 and use xlrd library which call removed getiterator method. You should find part of code in the file from your "Trace back" and replace getiterator with iter.

It might require to run Python file and replace that calls couple times.

After reading the comments I can confirm that this one below worked perfectly.
Make sure you have XLRD version 1.2.0 New version of xlrd does not have the xlsx option.

pip install xrld==1.2.0 

import xlrd
xlrd.xlsx.ensure_elementtree_imported(False, None)
xlrd.xlsx.Element_has_iter = True

pd.read_excel(path) #don't even need the engine=openpyxl

Worked in Python 3.9

This showed up for me when I upgraded to Python 3.9. The difference seems to be related to a combination of the compression format of xlsx files and as the deprecation of an iterator function.

For xlsx documents I need to specify the engine='openpyxl' keyword argument when opening.

This is not the case for csv or xls documents.

Install openpyxl

$ pip3 install openpyxl

Open xlsx and xls files with different engines.

from pathlib import Path
import pandas as pd

file_path = Path(file_name)

if file_path.suffix == '.xlsx':
    df = pd.read_excel(file_name, engine='openpyxl')

elif file_path.suffix == '.xls':
    df = pd.read_excel(file_name)

else:
    # handle other file types
    pass

Follow the below steps:

1. Go to /Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages/xlrd
2. open the xlsx.py in any text editor and replace the two getiterator() method with iter().
3. Reload your Jupiter notebook. It will work.

I had this problem while working on a course from stackskill. After looking around I found that the xlsx.py file itself has getiterator() and from what I had read it needs to be just iter().

There are only two sections in the xlsx.py file that require changing as far as I had noticed. It seemed to be an easy fix.

I will post the line of code that I have changed if you would like to modify them.

The first section of the code I came across is on line 266 and again on 312.

The original line for each should like this.

for elem in self.tree.iter() if Element_has_iter else self.tree.getiterator(): 

With the modification it should look like this.

for elem in self.tree.iter() if Element_has_iter else self.tree.iter():

Make sure you make a copy of the file before modifying it but as of right now it seems to be working better then adding engine="openpyxl" as this seemed to only work partially with certain sections of code as opposed to others.

I hope this helps everyone out looking for a fix for this. I 've also included a link of where I found the fix.

https://www.programmersought.com/article/61207614127/

To avoid messing with xlrd, you can also save your Excel file with .xls extension instead of .xlsx.