Improve error reporting when tabular_data is a dict of non-iterable values

I am trying to tabulate a counter from collections counter

print(tabulate(counter))

this approach gives an error message:

 File "/Users/wf/Library/Python/3.8/lib/python/site-packages/tabulate.py", line 1528, in tabulate
    list_of_lists, headers = _normalize_tabular_data(
  File "/Users/wf/Library/Python/3.8/lib/python/site-packages/tabulate.py", line 1089, in _normalize_tabular_data
    izip_longest(*tabular_data.values())
TypeError: 'int' object is not iterable

What else should i do - or would this be a feature request?

This is what it did so far to get my result - unfortunately floatfmt doesn't seem to work:

def showCounter(self,counter,title,headers=["#","key","count","%"],tablefmt='pretty',limit=50):
        '''
        show the counter contents
        '''
        bins=len(counter.keys())
        limit=min(bins,limit)
        total=sum(counter.values())
        table=[("total",bins,total)]
        for i,bintuple in enumerate(counter.most_common(limit)):
            key,count=bintuple
            table.append((i+1,key,count,count/total*100.0))
        print(title)
        print(tabulate(table,headers=headers,tablefmt=tablefmt,floatfmt=".2f"))

Hi,

I'd like to second this feature request.
It would be great to have built-in support for collections Counters, instead of writing my own wrapper function.

Best regards,
Ivan

@WolfgangFahl

Hi @WolfgangFahl,

When tablefmt='pretty' is set, tabulate appears to be treating each column as if it were a string, prior to applying floatfmt. So no float formatting is being applied, even though the floatfmt parameter is set. As a workaround, you can either not specify tablefmt, or you can apply the format builtin when calculating your data.

table.append((i + 1, key, count, format(count / total * 100.0, "0.2f")))

As for directly using a Counter object with tabulate, you can call the Counter.items() method, and it should work.

from collections import Counter
counter = Counter(['a', 'a', 'a', 'b', 'b', 'c', 'c', 'c', 'c', 'c', 'c', 'c', ])
print(tabulate(counter.items()))

# Expected output:
#
# -  -
# a  3
# b  2
# c  7
# -  -

@patrickdeanbrown

@patrickdeanbrown thank you for looking into this. Your workarounds are nice but do not solve the problem/fulfill the need of a simpler interface that this issue is requesting.

Tabulate can process an iterable (list) of dictionaries. It should work with a list of Counters too:

>>> c = Counter(a = 3, b = 4)
>>> print(tabulate([dict(c)], headers="keys"))
  a    b
---  ---
  3    4
>>> print(tabulate([c], headers="keys"))
  a    b
---  ---
  3    4

Using counter.items() is the intended way to print similar data as rows.

>>> print(tabulate(c.items()))
-  -
a  3
b  4
-  -

But the error message can be improved. The library should have suggested what data shape it expects.

Fix astanin#124. Improve error reporting when non-iterable data provi…

Merge pull request #233 from ilya112358/add_msg_when_not_iterable

Navigation Menu

Repository navigation

Improve error reporting when tabular_data is a dict of non-iterable values #124

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Improve error reporting when tabular_data is a dict of non-iterable values #124

Description

Activity

WolfgangFahl commented on May 3, 2021

ivan-radic commented on May 17, 2021

patrickdeanbrown commented on Nov 4, 2021

WolfgangFahl commented on Nov 5, 2021

astanin commented on Oct 19, 2022

astanin commented on Oct 19, 2022

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions