Egon,

Forgot to tell there is another option.

POST to /compound should work, it will return compound URL.

Then you'll need to add the compounds URL to a new dataset:

curl -X POST -H "Accept:text/uri-list" -d
"compound_uris[]={host}/compound/666"  {host}/dataset

or to an existing one:

curl -X PUT -H "Accept:text/uri-list" -d
"compound_uris[]={host}/compound/666"  {host}/dataset/999

Hope this helps,
Nina

On 29 October 2010 15:01, Nina Jeliazkova <jeliazkova.nina at gmail.com> wrote:

> Hi Egon,
>
> On 29 October 2010 14:49, Egon Willighagen <egon.willighagen at gmail.com>wrote:
>
>> Hi all,
>>
>> I was just wondering on how to do something... when I add a molecule
>> to a data set, I would like to know the matching URI... is there a way
>> to achieve this? A POST to /dataset returns me the data set URI...
>>
>
> Yep, this is how we got it in the API.  Same for PUT.
>
>
>>
>> Is there an implicit assumption that molecules are numbered according
>> to in what order they are added... but that would only work if added
>> as SDF...
>>
>
> No, there is no such assumption.
>
>>
>> I almost feel myself needing to put one molecule in one data set, but
>> there has to be a better way... right?
>>
>
> Well, you can, but doesn't sound nice .
>
> If you only add a single molecule and get dataset URL from the POST
>
> curl -H "Accept:text/uri-list"  dataset_URL
>
> will return the list of compounds URLs.
>
> Is this sufficient?  In principle one can assign arbitrary identifier to
> the molecule and it will be available as a feature.
>
> I am not sure what is the best way to return "matching" compound uris after
> POST, just because POST allows adding multiple  compounds.
>
> Nina
>
>
>> Egon
>>
>> --
>> Dr E.L. Willighagen
>> Postdoctoral Research Associate
>> University of Cambridge
>> Homepage: http://egonw.github.com/
>> LinkedIn: http://se.linkedin.com/in/egonw
>> Blog: http://chem-bla-ics.blogspot.com/
>> PubList: http://www.citeulike.org/user/egonw/tag/papers
>> _______________________________________________
>> Development mailing list
>> Development at opentox.org
>> http://www.opentox.org/mailman/listinfo/development
>>
>
>