SMS: Create hash value from concatenated CSV columns?

Support
1

Hi folks,

I am using SMS to convert data from a CSV, following the “cars” example in the manual. I need to create an IRI for an interval that has a start date and an end date. In this case, the lifespan of a person.

Start date is birthdate: {#birthdate}
End date is deathdate: {#deathdate}

I am trying to create something like:

cdp1:Lifespan_HASH
rdf:type study:Lifespan ;
time:hasBeginning cdp:Date_{#birthdate};
time:hasEnd cdp:Date_{#deathdate} .

How can I create the Lifespan hash value hash(“birthdate”+“deathdate”) ?

I’m trying to avoid pre-processing the data.

Also, is there an SMS syntax reference available? I expect I am going to need it.

Cheers,

Tim

2

Hi Tim,

We don’t currently expose a facility to execute functions over inputs. We have a special value you can use in the field template to generate a UUID. It is _UUID_. So you can create the lifespan IRI as cdp1:Lifespan_{_UUID_}. The IRI contents itself shouldn’t be relevant, but serve to identify the relationships with what it represents.

We haven’t published a grammar specification for SMS mappings but it looks like you have the hang of it. Basically you use the curly braces as a placeholder for field values. Let us know if you have any specific questions.

Best,
Jess

3

_UUID_ will work perfectly.

Thanks a bunch!

Tim

1 Like
4

Ok. I did not think this through in my quick reply above.

I need to refer to the same IRI in other triples.

So, creating the UUID is fine here:

cdp1:Lifespan_{_UUID_}
  rdf:type study:Lifespan ;
  time:hasBeginning cdiscpilot01:Date_{#birthdate};
  time:hasEnd cdiscpilot01:Date_{#deathdate} ;

But then how can I refer to that same UUID to assign a PERSON to that lifespan. Won’t this second _UUID_ create a new, unique ID (not what I need) ?

cdiscpilot01:Person_{#uniquePersonId}
  study:hasLifespan cdiscpilot01:Lifespan_{_UUID_}.

That is why I was thinking about concatenation: So I could recreate the identical hash value in multiple places during the data conversion.

5

In that case it would make sense to just use the date values in the IRI template, e.g. cdp1:Lifespan_{#birthdate}_{#deathdate}.

Jess

1 Like
6

Well that just makes perfect sense! I was overthinking it. :crazy_face:

Case closed! :+1:

2 Likes
7

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.