Jira issue originally created by user @plaflamme:

Jira issue originally created by user @plaflamme:

The Opal Key SQL schema is somewhat difficult to navigate and understand. There is a need to revisit the Hibernate Annotations to reduce the resulting SQL schema's complexity.

A discussion about this is available on obiba-dev group. Please refer to this discussion before investigating this issue.

Jira issue originally created by user nemond:

Jira issue originally created by user dspathis:

Jira issue originally created by user @plaflamme:

Based on semantic mapping of DataItems, a user should be able to define a report that selects and filters DataItem's properties.

Jira issue originally created by user tdebat:

Currently onyx-import cannot handle importing repeating variables such as blood tubes in to the Participant Key database. Onyx-import must be able to handle repeating variables.

Jira issue originally created by user dspathis:

Includes implementation of:

• OpalKeyClient
• DecryptCommand
• DecryptCommandOptions
• opalkey.bat (script to launch opalkey)

DecryptCommand will have an associated strategy (see earlier sub-tasks) that it will use to acquire an unencrypted InputStream of the targeted file. It will then read that InputStream and write out the data to another file.

Note that most of the work here involves pulling together code implemented in the other sub-tasks.

Jira issue originally created by user nemond:

Create/update participant as necessary
-Decrypt file
-verify digest

Jira issue originally created by user dspathis:

This tasks includes both the implementation of the FactoryBean as well as the corresponding Spring configuration.

Jira issue originally created by user nemond:

See comments for code review notes.

Jira issue originally created by user @plaflamme:

Jira issue originally created by user dspathis:

Jira issue originally created by user @plaflamme:

Jira issue originally created by user @plaflamme:

Jira issue originally created by user dspathis:

Jira issue originally created by user @plaflamme:

Jira issue originally created by user nemond:

Jira issue originally created by user dspathis:

Jira issue originally created by user @plaflamme:

Jira issue originally created by user @plaflamme:

Jira issue originally created by user @plaflamme:

Jira issue originally created by user nemond:

Jira issue originally created by user @plaflamme:

The startup scripts use java.ext.dir to setup the classpath. The JVM's ext dir is different when using a JDK or JRE:

JDK:
$JAVA_HOME/jre/ext

JRE:
$JAVA_HOME/ext

The use of ext dir is probably not the solution here. We should build the classpath by concatenating jar files located within the lib dir.

This is very easy to implement for bash. Windows batch files are more tricky. Solution is to look at how Tomcat builds its classpath in startup.bat file: it uses an external batch file to build the classpath.

Jira issue originally created by user @plaflamme:

Using the basic semantic mapping, Opal should allow extracting data based on DataItem selection and filtgering

See report definition in wiki to determine the requirements.

Spring Batch should be used to define the extraction jobs. An «export» command should be added to the list of available cli commands.

Jira issue originally created by user @plaflamme:

When loading occurrences from onyx, the DataPoint occurrence IDs should be normalized, starting at 1.

That is, the first occurrence of a variable should have 1 as its occurrence ID, the second 2 and so on.

Jira issue originally created by user @plaflamme:

The default storage engine in MySQL is configurable (--default-storage-engine). As such, jdbc-mart cannot rely that the default engine is MyISAM. It should force it to MyISAM.

This could be conditional, depending on the amounts of columns being created.

Jira issue originally created by user dspathis:

The strategy interfaces are:
IOnyxDataInputStrategy
IChainingOnyxDataInputStrategy

The strategy context class is:
OnyxDataInputContext

For details, refer to
http://wiki.obiba.org/confluence/display/CAG/DecryptionofImported+Data.

Jira issue originally created by user @plaflamme:

Each variable must have a unique code. This code must be automatically generated by Opal and should be sequential. It should have a configurable prefix and a series of digits.

For example:
CAG-2947
OHS-17880

Notes:

• The variable code's string length will vary. If we try to prefix the number with zeroes, we end up either limiting the quantity of variables the system can accept or having to manage the number of zeroes to add to codes. This complicates the issue needlessly.
• Advancing the sequence of codes should probably be done outside the transaction that is creating the variables. Unless the database supports sequences, this is required because two processes creating variables must not end up generating the same code.

Jira issue originally created by user nemond:

Jira issue originally created by user @dwainelson:

When creating reports for SAS the variable names can not be any longer then 32 characters.

Jira issue originally created by user tdebat:

Ensure that the onxy-import command exits when finished

Jira issue originally created by user @plaflamme:

Jira issue originally created by user dspathis:

Jira issue originally created by user @plaflamme:

Some variables are marked as "keys" in the Onyx catalogue. These are variables for which there should be no DataPoint as they contain identifiable values.

The value should be registered within the key database, but no DataPoint should be produced.

Jira issue originally created by user dspathis:

Variable --> DataItem
DataItem --> DataPoint
VariableAttribute --> DataItemAttribute

Jira issue originally created by user @dwainelson:

The JAVA_HOME should point to were the jre is located.

Jira issue originally created by user nemond:

defaultParticipantServiceImpl

Jira issue originally created by user tdebat:

The balsac barcode was being export as a boolean. Ensure that the fix made in onyx is now exporting a unique barcode and that it is correctly imported into the Participant Key database.

Jira issue originally created by user nemond:

Jira issue originally created by user @plaflamme:

The onyx-import job does not handle occurrences yet. That is, some variables allow repeated values for the same participant. Each repetition is an occurrence of the variable. When storing these values, we must store the repetition's ordinal (sequence within the repetition).

Given the following variable structure:
BP
|-SYSTOLIC
`-DIASTOLIC

Values will be provided as follows:

766
767



120


120




100


100

In this example, there are 2 occurrences of the BP variable, which creates 2 occurrences for each of its child variable. The DataPoints produced should have their "occurrence" property set. The first should have "1" and the second "2" (and not "766" and "767").

Jira issue originally created by user @plaflamme:

Keystore credentials are now passed through an instance of OnyxDataInputContext. The credentials should actually be setup so that a console prompt appears when required.

This is achieved by using the KeyStore.CallbackHandlerProtection class during KeyStore.load().

Jira issue originally created by user @plaflamme:

Jira issue originally created by user nemond:

-Define method for updating and querying participant key map
-method for generating dcc participant id (configurable)?
-DefaultParticipantServiceImpl

Jira issue originally created by user @plaflamme:

Jira issue originally created by user nemond:

Jira issue originally created by user tdebat:

If the environment variables OPALHOME or JAVAHOME are not set then the DOS batch file commands "opal" and "opalkey" mistakenly close the DOS terminal window.

Jira issue originally created by user @plaflamme:

Using Upgrade Manager, add the following install steps:

• Upgrade Manager schema creation (for persisting opal version)
• Hibernate Schema Creation (see LocalSessionFactoryBean for programatically creating the schema)
• Spring Batch Schema Creation.

Jira issue originally created by user nemond:

Jira issue originally created by user tdebat:

The "opalkey decrypt" command no longer works. This occurred after the file reorganization intended to separate onyx specific code from opal core. In order to fix this problem we will need a way to "lazily" load Spring configuration. For example "opalkey decrypt" does not need the database, so there is no need to load database beans in order to run this command.

Jira issue originally created by user @plaflamme: