Django-import-export: semicolon/comma

[klabedz]

Hi!

I am working on import of csv file - in csv file, i've got field delimeter = semicolon, and when I am importing file i am getting an error:
KeyError: u'Column \'NumeroCommande\' not found in dataset. Available columns are: [u\'NumeroColis;"NumeroCommande";\']'

So it seems, that it takes all the columns and melting it in one.

If there is a comma set as a field_delimeter there is no problem with importing file.

Is there a way to replace semicolon with comma in your code while importing a file?

[bmihelac]

You can add new formats, but note that csv with delimeter ";" does not exists in tablib

see:

https://github.com/bmihelac/django-import-export/blob/master/import_export/formats/base_formats.py

[chhantyal]

Looks like tablib now supports custom delimiter, see here https://github.com/kennethreitz/tablib/commit/70716fdd216755dc4b542df74e95b0e5ac74f0ee

[mgrdcm]

As an example for @klabedz and anyone else, I just did this to add semicolon CSV support to a project I'm working on based on @chhantyal's and @bmihelac's comments above:

from import_export.formats import base_formats

class SCSV(base_formats.CSV):

    def get_title(self):
        return "scsv"

    def create_dataset(self, in_stream, **kwargs):
        kwargs['delimiter'] = ';'
        return super().create_dataset(in_stream, **kwargs)

# ...

# within the Admin subclass that uses ImportMixin or ImportExportMixin:
    def get_import_formats(self):
        return self.formats + (SCSV,)

So this is simple enough I'm not sure it warrants making an actual code change to django-import-export.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[avillapalos]

The answer of @mgrdcm it's almost good but it doesn't work for me, I've solved it with (based on @mgrdcm answer):
from import_export.formats import base_formats

class SCSV(base_formats.CSV):

    def get_title(self):
        return "scsv"

            def create_dataset(self, in_stream, **kwargs):
                    kwargs['delimiter'] = ';'
                    kwargs['format'] = 'csv'
                    return tablib.import_set(in_stream, **kwargs)

# ...

# within the Admin subclass that uses ImportMixin or ImportExportMixin:
    def get_import_formats(self):
        self.formats += [SCSV]
                return [f for f in self.formats if f().can_import()]

[aljimgar]

I'm having the following error when I try to import semicolom csv...
"NameError encountered while trying to read file:"