#Data Import and Export

Hi everyone, 

How do you alter primary key on a table that has data?

Any example much appreciated.

Cheers,

Tom

Is there a command that will loop through the flat files of a given Linux/Unix folder?  I can write the code to open and read each file. But the file names are unknown. I am looking for a way to access each file given a named Linux folder. The files have differing structures so a record map will not work.

Thank you for reading and thank you even more for answering!

There will be serious errors and garbled code when I use the Data Migration to connect to MySQL through JAVA gateway. In fact , there are always many problems when I import data to the Caché database.

Hi,

I have an odd problem that occurs when I'm trying to import lookup tables in the XML format into Studio.

I store my files in in C:\CodeRepository.

I can manually import these lookup tables using "Import local..." option in the InterSystems Studio. 

However, when I'm trying to import them using command in terminal: do $system.OBJ.LoadDir("C:\CodeRepository\","ck")

I'm getting the following error:

Loading file C:\CodeRepository\Title.lut.xml as xml
ERROR #5863: Another user has 'Title.LUT' open for editing.
  > ERROR #6315: Errors reporting importing XML subelement in file

Exported contents of a namespace on one server (classes, include files and lookup tables). Importing that code into a newly created namespace on another server. Both servers Ensemble 2018.1, same build. Export was via InterSystems Studio. Export is around 18Mb in total (XML file sizes).
When importing and compiling on the new server, getting errors as below - with #6301: SAX XML Parser error prominent - on a number of the imported files, all containing data transformations or business processes.

hi,I meet a problem,i want to move huge data from ensemble to oracle.it`s about 3years business data.i have tried some case！

1. EnsLib.SQL.OutboundAdapter

i parse the data one by one,and translate it into inser or update sql statement then write to Oracle，it·s most slowly，and take more disk space。

2.save the data to a new table，and then use dbeaver transport data to oracle。

3.user linktable. save object as insert，but，i can·t user %openid method.

hope to find a high-efficiency way！

您好，我想问一下为什么我无法通过Iris中的内置转换模板将从HAPI-FHIR服务器下载的患者资源转换为SDA格式。它总是表明缺少某个默认值，或者这就是我们需要更改格式吗？我想问一下我们如何快速创建自己的映射，以及如何通过json将所需的数据转换为sda并将其存储在iris中，然后我们可以通过齐柏林飞艇进行调用，非常感谢您解决我的问题问题

Transaltion for the question:

Hello, I would like to ask why I am unable to convert patient resources downloaded from the HAPI-FHIR server to SDA format via the built-in conversion template in IRIS. It always indicates that some default value is missing, or is this the format we need to change?

Hi

We have Edge Production where we uploading flat files clinical data, and during testing we do this for several rounds load data, test them and reload them. So everytime we have to generate differnet patient data or change MRNs etc. So request is is there a command to delete patient records that will remove from the tables, and delete reference in the registery as well.

Any guidance that you follow for testing backload data and delete loaded data, pls share

I am exporting data from Cache using the SQL export wizard. This is on a docker image of Cache hosted on a Linux server.
I select my database, schema, and table, make sure that all columns are being exported, check other settings, and then finish the export.
All server settings look correct and permissions are rwe on the directory. I have tried various directories, including a system level directory. I tried creating the file before I do the export in case there was an issue with creating the files, and I also made sure the created files had rwe permisisons.

when doing ETL process from cache database to mysql database(ods) and to another mysql database(dwd).

There are  lots of DR columns in one table.

In the first step we need two databases to be totally consistent(ods=cache database)

However, the DR column in another database lost its meaning and become to normal number.

In the last step we do the data governance in ODS to DWD. My question is how to write sql without many `join` in this step.

In terms of general through-put design and long term support, I'm considering what would be a "best approach" for needing to create multiple batch files in a few different layouts from the same data-sets.

I need to iterate over a few very large tables to generate consistent batch outputs for different partner applications that need to be working on the same "version" of the source data.