So we have an app that needs to be saved and loaded from its disk.
We have currently found a state in an object and that object is being directly serialized in XML with XML serializer object. I went to this project with another developer and it seems that this is the right way to go about it. He claimed that ideal models on the disc represent ideal in memory.
I think the way data is saved in the disk, its memory should be decoupled by its representation so that we can modify the breakdown pattern for one or the other without saving the format. Not to mention that the XML serializer needs to read / write everything, where the object may need to be read only in many cases.
The best practice regarding serializing in question is XML data to XML, whether it is to directly hydrate your object directly from the XML, or reduce the process by using an object whose only purpose is XML serializer Should be feed in?
It seems that you are pointing the negative of your solution, but in fact there is a good option Are not giving up.
Either:
a) Use your solution until it is now
or
b) your own solution Who address issues raised by you.
One thing that might be suitable to explore: You can see object databases like Debian 4O, which may be able to save / load the object (saving) You can make most of the changes, but can not be ideal for the case you want to be able to edit the on-disk status directly (you can do this, but you have to jump through some hoops).
Comments
Post a Comment