|
History
| Log In
|
|
|
|
Bharathi, I am just about ready (this weekend perhaps) to release the next version. I believe if you agree with my suggestion about handling array types, that it could be included. I think it is only a few lines of code. I do not agree with your original suggestion due to the likely complexity it may introduce - unless you can convince we its not. I still don't know how you propose to handle the case of multiple array fields.
Bob,
Thanks so much for your reply. I agree that my proposal might introduce some complexity. I would like to go ahead with your suggestion of handling arraytypes. Thank you once again Hi Bob.
Thank you so much for such a quick build. I will test it and let you know if I find any issues. Regards Bob,
How should this work? The following are the query and result displayed: {sql:dataSource=PhsDev|multipleResults=false|output=html} select measured_depth from survey where unique_identifier = '174000000' {sql} Result was one row in table format: From a beanshell the following results in table of measured_depths: while (result.next()) int arrayLength = mdFloatArray.length; out.println("measured_depth array length = " + arrayLength); for (int index = 0; index < arrayLength; index++) { out.println("measured_depth[" + index +"]=" + mdFloatArray[index]); }... Results: What database are you using? I should have investigated differences in how databases handle array types. If it is not consistent between databases its going to be a real pain. I tested only with postgresql.
Bharathi, did you do any investigation on this. Also, forgot to mention earlier, a new parameter will be needed to turn on this type of processing instead of the current behavior. This is NOT implemented in b3 but will be needed for general release to retain upward compatibility with existing behavior. Bob,
This morning I was having trouble setting up the Datasource for standalone confluence in my local PC. Actually , I work with Quentin and we tried to deploy the new plugin in one of our linux servers. The result is as described by Quentin. We are using a proprietary JDBC driver to a middle tier server. driver class is com.phs.client.jdbc.PhsJDBCDriver. I will work on this more tomorrow. Thanks The current implementation is dependent on http://java.sun.com/j2se/1.5.0/docs/api/java/sql/Array.html
 and http://java.sun.com/j2se/1.4.2/docs/api/index.html Bob,
Could you please tell me what new parameter i should use for displaying array type? Thanks Bharathi Bob,
We tried to use this modified plugin for Oracle instance. It looks good. I would like to make modifications to this so that it works for PowerHub datasources also. Could you please help me with that? Thanks Bharathi Bharathi, array processing on always on with b3 (see my edited note above). Glad it work with a standard database driver. Regarding PowerHub, I don't know what that is. It sounds like it is a non-standard jdbc driver. Regarding, help, my priorities are to get the next version out and do not have much time for non-standard support.
Bob,
Thanks for the help. I would like to know if the new version that is going to be released with have this array processing or not. We are thinking of using Oracle database directly instead of using PowerHub middle tier server so it will be really help to have this new functionality in the next version. I really appreciate. Bob, I checked out the new version , do you think if the expandArray is defaulted to false , the value should be treated as string and display the array as string instead of null in the colunm? I think it should be displayed as string. I am seeing the data as below where COL1 is of array type, i think this might be misleading.
COL2 COL1 thanks Interesting. That was with Oracle using standard oracle jdbc driver? Postgres automatically shows it as {1,2,3,4}. Basically it uses a getString function and probably up to the driver as to what to show. I would be inclined to default to true if nothing (or null) is shown for various DBs. Then at least people would have the option of getting back to the previous behavior.
Yes, I tested it in Oracle jdbc . After seeing your comments ,i am surprised too. I agree , may be we have to keep the parameter default to true.
Thanks |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
Improvement
Resolved
Major
SQL Macro need to handle array data type
Could you please help me with this issue? I really need this functionality. Could you please help me to modify this plugin code?
I could not find the latest source code after 3.3.0. I see the beta version as jars but no source code.
Help !
Thanks
Bharathi