SQL

I am tasked with using CACHE to insert data retrieved from a CACHE data base and insert it into an sql database. The columns in the sql table that I am trying to insert data into have names that contain underscores such as "ACCESSION_DATE" I found a utility in CACHE to connect to the sql data base and perform inserts. I have the data I need to insert. I need to pass the  data into the sql utility referencing the column names.

I tried using indirection to set the data into underscored variables, but that isn't allowed  in cache object script.

Anyone have suggestions as to how to do that?

Thank you,

Mike Minor

Has anyone use any other setting then -1 on the Stay Connected setting on a SQL.OutboundAdapter? I am having issues with a particular database and wonder if it would be beneficial since I only connect once a day to read the data from the Adapter to create an HL7 message.

Thanks

Scott Roth

Hi all.

I have some problem with getting data from external source (PostgreSQL) via JDBC and load it in DeepSee and I resolved some problems while writing this question :)

So..Below I proved my steps of working

I came across How to import a tab separated text file into a SQL table programmatically?, which appears to indicate that I can pass the filename and all records will be imported. However, when I put use the Import method only a single record gets imported. However, all records are imported if I use the management console to import the file by selecting it and choosing the options.

Has anyone ran into this in the past?

I have several BPL's that act like SQL Server Integration Services (SSIS) where they are transferring data. When I do this they are all being written under 1 session id, is it possible to make the BPL create multiple session ID's as it is cycling through a Snapshot, so it is easier to read in the trace?

Thanks

Scott Roth

It is often necessary to sort the results of a query on a string field containing a combination of alphabetic and numeric characters. In cases like this the default string collation may not always return the data in the expected sequence.

An example of this may be where a select from Samples.Person should order the results by the street address, but firstly ordered by the street number part as numeric, and then by the street name.

The default query will return the results as follows:

SELECT Name,Home_Street

FROM Sample.Person

I am working my way through some SQL documentation.

The documentation follows and my question comes after.

A query supplied to %Prepare() can contain input host variables, as shown in the following example:

  SET minage = 80

  SET myquery = 3

  SET myquery(1) = "SELECT %ID AS id, Name, DOB, Home_State"

  SET myquery(2) = "FROM Person WHERE Age > :minage"

  SET myquery(3) = "ORDER BY 2"

  SET qStatus = tStatement.%Prepare(.myquery)

This talks about input host variables

I thought that input host variables only work with Embedded SQL, not Dynamic SQL

Thank you

Hi,

I am afraid I am stuck again. I have ensemble connected with a MsSQL database and I have a service that calls a store procedure. All is working, and I can get strings from the database in the service. The problem is that the data the store procedure returns is suppose to be an HL7 message so I need to convert the string into HL7 format. I have been using the ##class(EnsLib.HL7.Message).ImportFromString  method but although I have a valid HL7 message from the string received from the DB...apparently the content of the message is being modified

I noticed that the subscript of the index map is actually the collated field (%SQLUPPER).

Is String the only data type going through this transformation? Any other data type would also go through this transformation?

Thanks. 

Hi, I am trying to execute a store procedure within an ensemble operation to connect to aMsSQL database and I am having problem to pass the input arguments... I have tried several things but only one seems to work and it is not the ideal solution.

This is the code that works...as you can see the parameter is added to the sql query...not the best solution but the only that works at the moment

This might be more of a math problem than a Caché question.
I have a SQL query that joins two tables. I want to assign a unique ID to each row of the product table.

1. I could append the GUIDs of the rows in the two tables, but there are a number of clients that expect a maximum length of 50 on this unique ID. Two GUIDs appended make 72 characters.
2. I could append the two GUIDs and then truncate the result, but now I'm worried about collision.

What's the chance of collision if I append the GUIDs and truncate the result to 50 characters? Is there a good way to solve this without updating the clients?

As I was going though and trying to figure out why our CACHE.dat has increased in size over the past 18 days, I found that EnsLib_HL7.Message is still retaining messages dating back to 2014 even though we have our purge set to 10 days. Has anyone else experienced this?

Thanks

Scott Roth

The Ohio State University Wexner Medical Center

Hi,

My doubt is about the SQL Query in Caché:

I don't want to take all results from the table and orgainze them manually, for example: I have a table with 50 records, but I only want to select 10 records, being from tenth until the twentieth and this without knowing their IDs.

So, how can I do this, without losing performance, there is any way?

NOTE: In other SQL languages I can do, for example, SELECT * FROM extbl Limit 10 OFFSET 10

Thanks,

Andrei L. Nenevê

I'm trying to create an sal connections from one cache system to another. I've created the connection part  and I think that is working. When trying to link a table through the link table wizard, I'm getting an error "<MAXSTRING>zCreateOneLinkTable+52^%CSP.UI.SQL.LinkWizardResultPage1.

Is this due to my global block size in the new system not being large enough? It is currently set at 8192.

Thank you

Mike