Use import_set

[mazing]

It seems it's not documented how to import a file and loop through the rows and columns.

Can someone help me out with a small example?

In the API the function import_set() is listed, but I don't know how to use this function.

[rabinnankhwa]

Read a file and set the contents as input to the function import_set.
For eg:

import tablib
data = open("test.csv").read()
dataset = tablib.import_set(data)
header = dataset.headers
rows = dataset._data
for row in rows:
    print row

[ghost]

i've tried the open().read() but only return None object from the tablib.import_set(). I'm pretty sure the filepath is correct because i can read the whole content of the file with "".join(open("path")).

            csvfiles = glob.glob( 'backup_db/{date}/*.csv'.format( date=args[0] ) ) # args[0] = '2015_05_02' (like year, month, day)

            for csvf in csvfiles : # csvf = 'backup_db/2015_05_02/SomeModel.csv'

                #print( 'importing ' , csvf , '\n' , ''.join(open(csvf,'r')) )

                dataset = tablib.import_set( open(csvf,'r').read() )

                print( 'dataset -> ' , dataset ) # None

                modelname = os.path.basename(csvf)[:-4]

                print( modelname+'Resource' )

                model = getattr(admin,modelname+'Resource')()

                print(dataset)

                model.import_data(dataset=dataset,dry_run=True) # Error - > 'NoneType' object has no attribute 'dict'

[ghost]

Same issue.

>>> import tablib
>>> print tablib.import_set(open("derp.xls", "rb"))
None

File is existing

[casimir]

Just encountered this issue with a CSV file. When using ; as separator tablib.import_set return None, whereas when using , it return the correct dataset.

[claudep]

Please reopen if you still have issues on the latest release.