A quick question regarding to SQL Query Plan:
Why these two requests have different plans, in particularly, why second request needs temp file? To me, temp file is a bad thing which should be avoided, right?

1. select * from Test.Log where cdate = 1
2. select * from Test.Log where cdate > 1

Plan for 1:

Read index map Test.Log.cdateIndex, using the given %SQLUPPER(cdate), and looping on ID.
For each row:
Read master map Test.Log.IDKEY, using the given idkey value.
Output the row.

Plan for 2:

I've recently encountered a HS Caché that won't start informing that Collation 30 is not available, but I have not found an easy way of knowing what collation is 30.

I've found that the following command returns the ones from the current locale, but not from all locales:

Set Rset = ##class(%ResultSet).%New("%Library.Collate:CollationList")

d Rset.Execute()

While (Rset.Next()) {zw Rset.Number_": "_Rset.Name}

Thank you

First,

Forgive me if this is a rather low-level question - I am a physician informaticist who is learning cache/mumps for the first time. I am sure this probably relates to some setting but I can't find a similar issue anywhere.

When I create a list, and then try and return that list - I get a bunch of symbols back along with my results, or sometimes I just get symbols.

I pasted a screenshot below. Thank is advance for the help!

Anthony

Dear All,

I'm looping through a list with characters (Parenthesis, square brackets etc) and I wish to remove these from the list.

The for loop and IF logic are effectively removing these items from a list.

However the variable I use for the maximum number of loops carries on beyond the number of items of the list after removal.

In the output value for these I get the integer of the for loop instead of a null for the empty list fields.

I can see in the documentation that the end variable is not editable once the loop is started.

Hello,

We have been reasearching how could we parse a string with a JSON structure like the following:

When using CallableStatement to call a stored procedure, the log shows an error. SqlCode-400.

This is my calling statement.

CALL web_DHCENS_CDSS.patadviceinfo_patadviceinfo(?,?,?)

After the database is restored, the WIJ file may be incompatible, so I deleted the WIJ file, and it needs to be set manually at startup

message in cconsole.log:

Enter Cache' with 
     csession CACHE2016 -B
and Do ^STURECOV for help recovering from the errors

after command " Do ^STURECOV":

I have a class that writes to an external SQL Server database. When the field is empty, it does not write the column and the column has the value NULL. In some cases, I want the column to be the value of the empty string instead of NULL.

Hi,

I've started to use Task Schedule function in Caché. But I have two questions about it:

A client stated to me that they are trying to eliminate SMTP servers from their network and want all email to be sent through their Microsoft Exchange Server. Admittedly I am not very knowledgable of the workings of Exchange Server.

Are there customers sending email notifications through Exchange Server? How is this accomplished? Does Exchange Server act as a SMTP server allowing use of the Email adapter or is there some other mechanism that must be used to communicate with a Microsoft Exchange Server?

Hello,

is there any mechanism available in DeepSee to keep values of AGE dimension up-to-date without rebuidling cube? Can age be determined at the mdx query execution time?

Thank you!

Dan Kutac

So calling this lookup manually from the console works as expected:

However, calling it from a method with some concatination to build the key is giving me problems:

I know that 1-1 relationships are not officialy supported by intersystems cache/iris so I want to know the best way to store data with this kind of data model.

Currently I have two classes that where implemented some time ago:

Table A with a relationship type one on table B

Table B with a relationship type one on table A

To compile I have a double compile with qualifyer U.

What is the best way to implement a data model with 1-1 relationships?

Thank you

We are beginning a project to switch over to a differnt EMR for our inpatient system, and we are wondering if others out here have worked on a similar project and how did you handle your Ensemble environment?

Currently, we use one namespace for all of our HL7 interfaces. Has anyone used a separate namespace for a project to keep the build separate until go-live? What were the pros/cons for you using a different namespace? Or did you use one and what issues did you have because of it?