mbellotti
commented
10 years ago It's because you've installed your submodules wrong :) Submodules are locked in to a certain commit, they do not update unless you specifically update them (or unless you install them incorrectly... perhaps by doing a git clone in the submodule directory instead of initializing the submodule itself) The parse function comes from the dataconverters library, which in the commit dataproxy depends on looks like this:
The CURRENT version of dataconverters, however, looks different:
https://github.com/okfn/dataconverters/blob/master/dataconverters/xls.py#L17
No **kwargs parameter to catch all the misc arguments, so when you give it the parameter "worksheet" it has no idea what you are talking about and errors
There are two ways to fix this:
- checkout the proper commits for all the repos under the vendor folder of dataproxy. A link to the commit each submodule is frozen to is next to the folder name on github for easy reference. Unfortunately this means you won't have any of the bug fixes for any of those repos
- remove the worksheet parameter from this call https://github.com/okfn/dataproxy/blob/9e4ad3972dd6cfef38f274af20abf550120c7acc/dataproxy/app.py#L246
Alternatively, I'm maintaining a fork of dataproxy here (https://github.com/mbellotti/dataproxy) that you might want to check out. I've updated the dependencies and rewrote some of it to run independently of AppEngine.
rufuspollock
So I've managed to get dataproxy running locally and it works for csv files. When I try with xls files, I get the following error. I get the same error whether I pass worksheet=1 or not. Any idea?
callback(
{
· error:
{
o message: "Data transformation failed. TypeError: parse() got an unexpected keyword argument 'worksheet'",
o title: "Data Transformation Error"
}
}
)
example url used below ( host changed to protect the innocent )
‘http://host.com:12000/?url=http://host.com:5100/dataset/a86b9d37-aef1-4425-adf5-09a6a0f2302d/resource/a97e5fea-c858-48dd-b56d-0f0f40355683/download/oldexcel.xls&type=XLS&worksheet=1&format=jsonp’