Query dynamoDB with non hash key field (with boto / python)

Question

I'm using dynamoDB with boto, and having a bit of a problem in the design/query of my table.

I'd like my data to look something like

+---------------------------------------+
hash_key    account_id    mykey
-----------------------------------------
1           12345         myvalue1
2           12345         myvalue2
3           12345         myvalue3
4           123456        myvalue4
+---------------------------------------+

And then retrieve all data for account 12345. Looking at the boto docs, I always need to have the hash_key available. I know how I would query this standard SQL / MongoDB, but I can't find a solution for boto. I assume this is possible? Thanks!

EDIT: This seems to work

+---------------------------------------+
hash_key    range_key    mykey
-----------------------------------------
12345       12568        myvalue1
12345       53890        myvalue2
12345       12322        myvalue3
123456      23432        myvalue4
+---------------------------------------+

Followed by

> res = table.query(hash_key='12345')
> for item in res:
>    print i

Since I want to grab all the entries with account # 12345, regardless of the range_key, I need to query instead of get_item

Did you ever experience a 'UnknownFilterTypeError' when querying like this? — ingh.am
– ingh.am, Commented Oct 20, 2013 at 13:46

yadutaf · Accepted Answer · 2012-09-18 21:07:07Z

5

I would use the account_id as the hash_key along with some range_key to differentiate them.

In DynamoDB, the primary key is composed of a (hash_key, range_key) range_key being optional. This tuple needs to be unique. Note that you will need the whole tuple to access a given element with get_item.

Having an 'auto_increment' hash_key is a bad habit from the SQL world.

If you want to know more on this subject, I wrote some background do on modeling data in the dynamodb-mapper documentation: http://dynamodb-mapper.readthedocs.org/en/latest/api/model.html#auto-increment-when-to-use

edited Sep 18, 2012 at 21:07

answered Sep 18, 2012 at 20:47

yadutaf

7,2201 gold badge41 silver badges48 bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

L-R Over a year ago

that's what I wanted originally, but the hash_key needs to be unique!

yadutaf Over a year ago

Yeah, use range_key to differentiate them. If you have more background on the model, I can try to suggest one.

L-R Over a year ago

I see - just tested, and indeed dynamoDB allows the same hash_key if you add the range_key. I guess the value I give to range_key is irrelevant (as long as it's unique), since I only use it to differentiate the entries? The data are actually GPS points that are inserted periodically. Thanks!

yadutaf Over a year ago

You're welcome. I updated my answer to reflect your comment. Note that you will need the whole (hash_key, range_key) tuple to access a random element. Btw, If this solves your issue, you may consider accepting this answer :)

L-R Over a year ago

Also added an edit - looks like query is what I was after - not sure of the performance implications here though.

|

Collectives™ on Stack Overflow

Query dynamoDB with non hash key field (with boto / python)

1 Answer 1

6 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

6 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related