DFS DQL SOAP Query question (caching results when using startingIndex)

jnichols1971

I am posting a DQL query to queryService, let's say something like this:

<?xml version="1.0" ?>
<S:Envelope xmlns:S="http://schemas.xmlsoap.org/soap/envelope/">
    <S:Header>
        <ServiceContext ...
                        token="temporary/jsdi9-23850uvj9023j4509u">
            <Identities xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                        xsi:type="RepositoryIdentity"
                        repositoryName="myrepo"
                        password="me"
                        userName="blah"/>
            <Profiles xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                      xsi:type="ns4:ContentTransferProfile"
                      isProcessOLELinks="false"
                      allowAsyncContentTransfer="false"
                      allowCachedContentTransfer="false"
                      transferMode="BASE64"/>
        </ServiceContext>
    </S:Header>
    <S:Body>
        <ns7:execute ...
            <query xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                   xsi:type="ns8:PassthroughQuery"
                   queryString="select * from dm_document order by r_object_id">
                <ns8:repositories>myrepo</ns8:repositories>
            </query>
            <execution maxResultPerSource="0" maxResultCount="500" startingIndex="0"/>
        </ns7:execute>
    </S:Body>
</S:Envelope>

If the query gets 5000 results, I'll be sending the query back 10 times (incrementing maxResultCount by 500).  All works great if I have an "order by" clause in the query.  But if I don't order the results at the query level, I sometimes get repeating rows betweeen each iteration - which makes me think that the query is being run new each time I run it - so with no order some of the results will overlap.  Would registering @ ServiceContext to get a registered security token session-ize this enough to remove this issue - or is there another solution?  Seems a waste to re-run the query every time I'm incrementing the startingIndex.  FWIW, I'm creating the SOAP request manually and posting/parsing it myself (it's a test script written in TCL for QA to validate source data against Documentum).

Thanks,

Jamie

jnichols1971

Looks like I get to answer my own question.

After more research, I found what I was looking for on page 171-172 of the ECS documentation.

<execution maxResultPerSource="-1" maxResultCount="200" startingIndex="0" cacheStrategyType="BASIC_FILE_CACHE_STRATEGY"/>

Changing that rather than having the default seems to cache the DQL results properly.

However, the next question:

Since the example is Java/C#, I don't know what else happens "under the hood" that might affect the SOAP request (since I make mine manually).  Doesn't look like the Profiles attribute allowCachedContentTransfer="false" needs to be changed either.

Hopefully that might help someone else if they are running large queries.