Hi Phil,
I am sorry I don’t quite understand your question. Yes, the SampleDB class defines the data model and the sample will create such a database if it does not exist. In the ddl sample function a column is programatically added to a table (by using ALTER TABLE …) and then a DBRecord object is used to read / modify records of that table. The creation of the DB is handled by the MySQL driver. If you feel that you need more control over the DDL creation process simply derive a class from DBDatabaseDriverMySQL and override getDDLScript(). And if you think that your improvements might be of public interest then you can send us your code and we might publish it. In order to let other users benefit from our conversation I have sent a copy to the empire-db-user mailing list. Regards Rainer Von: phil barrett [mailto:[email protected]] Gesendet: Montag, 14. Dezember 2009 20:16 An: Rainer Döbele Betreff: Re: AW: Samples Problem Dear Rainer, I also checked the advanced sample. For your coding you require that you have a class that defines the DB as in SampleDB, however this would be invalidated by your statements to add fields, is this correct? Do you have some way round it? With MySQL is there some way to set types such as InnoDB, MYISAM? I'll send my questions thru to the mailing list also when it replies to me.. Regards Phil ________________________________ From: Rainer Döbele <[email protected]> To: phil barrett <[email protected]> Sent: Mon, 14 December, 2009 10:41:13 Subject: AW: Samples Problem Dear Phil, I am terribly sorry that I have not been able to answer your e-Mail earlier. In fact I saw it when I received it but lost track of it since it arrived in the wrong e-mail folder. The e-mail address you have used is outdated - please let me know where you found it so I can remove it. Since we have become an Apache project all questions should be addressed to the empire-db user mailing list: [email protected] However to be able to post a message there, you first have to subscribe to the list, by sending a message to [email protected] You will then receive an e-mail which you have to confirm. In most cases you should get an answer within 24 hours. This is how things work at Apache. Now to your questions: Our latest official Apache release is 2.0.5. Unlike our previous releases this release is intended to be used with Apache Maven (see http://maven.apache.org). Maven is a build and dependency management tool and solves a lot of problems with distributions. Maven can even build your Eclipse project files. There is some info about that in the readme.txt file supplied with the distribution. But in short: 1. Install Maven 2. Open a command window 3. Go to the distribution source directory (/src) 4. Run “mvn clean install eclipse:eclipse” Then you will have your Eclipse project files generated. I have not been able to check whether there is a problem with the sample and MySql. If it does not work with autocommit set to true try changing it (although it should not really matter). If you receive an error please post it (to [email protected]). Last, there is an example for data model changes at runtime in the advanced sample (see function ddlSample() in class SampleAdvApp). I hope this answer is not too late. As mentioned above please adresse further questions to the Apache mailing list. Thanks, Rainer Von: phil barrett [mailto:[email protected]] Gesendet: Dienstag, 8. Dezember 2009 10:20 An: empire-db Betreff: Samples Problem Hi, downloaded and tried to run your samples, I have the latest version of Eclipse installed. From your directory structure there doesnt seem to be a project path or otherwise that Eclipse will import, the only way I could do it was to include the files as source. Might be better if you exported your eclipse project as a project archive. I am using MySql fresh defaults install, if you run the simple sample it fails as the autocommit is true, if I get any reply back from this mail, I can get you more info as to where. Might be useful to say where the jars that need to be included can be found, or include them in the project archive. You have some words about this 'which even allow data model changes at runtime' but no information as to how you achieve this, or would implement it as you have at least in the example static bean classes such as Employees which has attributes and indexes etc defined which would mean a change to a class that would then have to be redeployed, new values which have to be set in a table which may contain many millions of rows, index rebuilds etc, can you give me some brief insight into this and I will try and test it myself (code examples would be good :- ) Regards Phil
