Documentation for non-command line usage

[gerpsh]

Currently the documentation only references command line usage. Would it be possible to get some documentation/examples on how to use csvdiff within Python code?

[larsyencken]

Hi, sorry we don't have much in the way of Python docs.

The core of the module works on lists of dictionaries, which we call records. You should do the import like:

from csvdiff import patch

diff = patch.create(from_records, to_records, index_columns)

See also patch.apply() and patch.save() in the code. Things are at least documented there.

[karakutu001]

how should I imagine "index_columns"? what kind of value is index_colums? why except index_colums an array?

I try to use the module with the following line

diff = csvdiff.diff_files('output2.txt', 'input2.txt',[] )

the script cannot find all the line where was a change. And I can not find what is wrong?

Thanks

[larsyencken]

The index columns are the columns that form the primary key for this dataset. It could be one column like ['id'], or multiple, like ['firstname', 'lastname']. It's going to depend on your data, on what uniquely identifies a row.

We use this to determine when two rows represent the same thing, so that we can detect changes.

[larsyencken]

I should add that the README has basic usage now, and using it without providing index columns raises an error on the current master.