Data::Conveyor::Manual::Instruction - About payload instruction objects
version 1.103130
Partial updates to business objects have necessitated instruction objects. The parser takes the request (EPP frame or mail) and creates the payload. For example, if the request specifies that a person's organization should be set the parser creates a person payload item object and sets the person handle on the data sub-object. It then creates an instruction container for that person payload item object and, using the instruction factory, adds a "change organization" instruction object to the container.
The policy stage needs to make sure that the new values are valid for the respective value objects, so when the payload runs the check() method, it also asks the instruction container to check itself. The instruction container asks each instruction object to check itself, and each instruction object asks its value object (or other data object) to check itself.
The existing values of the business object (for example, the person as it is stored in the database) could have invalid values (for example, an invalid email address). These values should not be checked, because a change to the person's address should not trigger a check on the person's email address. If a value was accepted previously, it should be accepted now as well, unless it is being changed.
However, we could also imagine the situation where a person that is an owner of a domain (i.e., a registrant) has a P.O. box as his address, which would trigger an exception. If we only want to change that person's phone number, we don't want to trigger an exception because of the P.O. box address. So we need to use the value objects' dirty flag. After reading the payload from the storage, make sure that none of the business objects' value objects are dirty. Each business object has a clear_dirty method (by way of inheriting from Class::Scaffold::HierarchicalDirty) which clears the dirty flag of all its constituent value objects. That way each object that has value objects (or has sub-objects which eventually have value objects) will have a kind of virtual dirty flag which is connected to said value objects. Therefore:
$ticket->payload->clear_dirty;
When the policy has checked the instruction container's values it will copy the new values over to the data object (e.g., the person object with the existing values). That way the new values will have the dirty flag set. The policy must only check that a registrant doesn't have a P.O. box address if the address is dirty. An address object can be said to be dirty if any of its constituent value objects is dirty.
Certain instructions may only occur as a set. For example, you cannot change the zip code of an address alone; you need to change the whole address. On the other hand, you can change a person's organization without having to specify other fields. When carrying out the instructions, a business objects also needs to enforce those instruction sets and, if necessary, throw an exception.
The transaction selector (txsel) for NICAT is also somewhat simplified. When trying to determine whether there is a person.set-owner transaction, it only has to ask the instruction container whether there is a change to the person's name. Previously it had to read the old person object and check whether there was a change.
txsel
NICAT
The instruction mechanism still needs to work with other registries that don't allow partial updates of objects; in this case, have a "clear" instruction which clears the object before setting all the new values.
See perlmodinstall for information and options on installing Perl modules.
No bugs have been reported.
Please report any bugs or feature requests through the web interface at http://rt.cpan.org/Public/Dist/Display.html?Name=Data-Conveyor.
The latest version of this module is available from the Comprehensive Perl Archive Network (CPAN). Visit http://www.perl.com/CPAN/ to find a CPAN site near you, or see http://search.cpan.org/dist/Data-Conveyor/.
The development version lives at http://github.com/hanekomu/Data-Conveyor and may be cloned from git://github.com/hanekomu/Data-Conveyor. Instead of sending patches, please fork this project using the standard git and github infrastructure.
Marcel Gruenauer <marcel@cpan.org>
Florian Helmberger <fh@univie.ac.at>
Achim Adam <ac@univie.ac.at>
Mark Hofstetter <mh@univie.ac.at>
Heinz Ekker <ek@univie.ac.at>
This software is copyright (c) 2004 by Marcel Gruenauer.
This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.
To install Data::Conveyor, copy and paste the appropriate command in to your terminal.
cpanm
cpanm Data::Conveyor
CPAN shell
perl -MCPAN -e shell install Data::Conveyor
For more information on module installation, please visit the detailed CPAN module installation guide.