[ckan-discuss] lowercase file formats

Fri Mar 16 16:48:19 GMT 2012

Hello,

Yes that is fine, Tom added it and I approved without thinking through
all the consequences.

David

On Fri, Mar 16, 2012 at 4:23 PM, David Read
<david.read at hackneyworkshop.com> wrote:
> Rufus has just suggested to me that we simply remove (revert) the
> lowercase functionality for now. David Raznick, does that suit?
>
> David
>
> On 16 March 2012 14:46, David Read <david.read at hackneyworkshop.com> wrote:
>> On 16 March 2012 13:17, Rufus Pollock <rufus.pollock at okfn.org> wrote:
>>> On 16 March 2012 11:36, David Read <david.read at hackneyworkshop.com> wrote:
>>>> I notice that file formats are now automatically changed to lower
>>>> case, whatever you type in. I can see this is useful to stop the
>>>> database ending up with both 'CSV' and 'csv', but I think capitals are
>>>> often important. e.g. our database has RDFa, iCal, netCDF, Flash
>>>> Player and having this hard-coded in the logic layer is a rather
>>>> inflexible. And what if on our site we'd prefer CSV to csv? Also, it
>>>> doesn't seem right to do this conversion without telling you *after*
>>>> you submit the form.
>>>
>>> Got you. I guess the issue is you end up with a lot CSV vs csv in the
>>> DB. What about netCDF versus netcdf or Flash versus flash. At least
>>> for faceting it is clear we need to standardize. So options are:
>>>
>>> * Don't change the format you are given (but then standardize into the
>>> faceting in SOLR?)
>>> * Change stuff to lower case (not liked but standardized)
>>>
>>>> How about just having format autocomplete with some well-known
>>>> formats, or convert only well-known formats (e.g. csv, xls) to
>>>> lower/uppercase, rather than the whole lot?
>>>
>>> Autocomplete just runs off whole list of existing formats at the
>>> moment. Autocompleting based on specific formats would add more
>>> complexity (so if you wanted that change we might want to remove full
>>> autocomplete).
>>
>> Yes, exactly. Mostly we care about the top 10 formats. In the meantime
>> to us getting round to this different auto-complete option, do you
>> mind if I make the code that forces it to lower case format
>> configurable to off?
>>
>> Dave
>>
>>>
>>> Sounds like this is all ticketable and real questoin is then cost /
>>> benefit of this versus other features :-)
>>>
>>> rufus
>>>
>>>> Dave
>>>>
>>>> commit da3edd4371e381c2a631578938e878a2cf612bd9
>>>> Author: Tom Rees <zephod at gmail.com>
>>>> Date:   Tue Feb 28 18:26:31 2012 +0000
>>>>
>>>>    [#1506]: Force format field to be lowercase in resources.
>>>>
>>>> diff --git a/ckan/lib/dictization/model_dictize.py b/ckan/lib/dictization/model_
>>>> index 309350e..1e1475e 100644
>>>> --- a/ckan/lib/dictization/model_dictize.py
>>>> +++ b/ckan/lib/dictization/model_dictize.py
>>>> @@ -79,6 +79,7 @@ def extras_list_dictize(extras_list, context):
>>>>
>>>>  def resource_dictize(res, context):
>>>>     resource = table_dictize(res, context)
>>>> +    resource['format'] = resource.get('format','').lower()
>>>>     extras = resource.pop("extras", None)
>>>>     if extras:
>>>>         resource.update(extras)
>>>>
>>>> _______________________________________________
>>>> ckan-discuss mailing list
>>>> ckan-discuss at lists.okfn.org
>>>> http://lists.okfn.org/mailman/listinfo/ckan-discuss
>>>
>>>
>>>
>>> --
>>> Co-Founder, Open Knowledge Foundation
>>> Promoting Open Knowledge in a Digital Age
>>> http://www.okfn.org/ - http://blog.okfn.org/