A maintenance release with several bug fixes reported by 2 different people. They were edge cases, related to imports and different namespace.
I've posted a new release of Dingo 2.0. This release enhances the support imported schemas. In config file you can tell Dingo to generate the class for the imported element and complextypes.
I've fixed 3 bugs reported by Howard Shank.
I finally got around to cleaning the CVS stuff and decided to switch to subversion. Now users should be able to browse the subversion repository for Dingo.
I fixed a bug reported by an user for sequence with maxOccurs unbounded and schemas with "using" reserved word. "using" is now escaped with "@" like other reserved words.
I've posted a new release 1.1.5, which fixes 2 bugs reported on 9/15 by Pascal Fluck - crabouif.
For the last few months, I've been busy with my rule engine on http://www.sourceforge.net/projects/ruleml-dev. I plan on enhancing dingo this summer.
I posted a new release with 2 bug fixes over the weekend.
I was busy last week with paying work and didn't get a chance to make a release with the new feature requested by an user. If someone find a need to compile a schema which doesn't declare any elements and only imports other schemas, CVS has been updated. I'm going to hold off on a new release until I get a basic implementation of sql support done.
An user of dingo reported an issue today and I was able to enhance dingo. The issue was caused by a schema that didn't declare any top level elements. Both .NET and Mono fails to compile the schema. the enhancement only took 30 minutes to write and test, so I plan to make a new release in the next 2 days. the code is already checked in to CVS for those who want to build dingo for themselves.
I've checked in some code for generating SQL. once the support is done, I plan to release 1.2 of Dingo. That probably will be a month or more. I plan to work on it slowly. Haven't decided which databases I want to support at the moment, but if I stick to the ANSI SQL, it should work on most databases.
I've started working on support for generating database tables to Dingo. this is all part of my original goal of supporting hiberate/nhibernate. once the sql support is done, it should be straight forward to generate classes, sql and hibernate mapping.
Currently I'm working on a reporting tool for JMeter, so I haven't had much time to work on Dingo. Once I have time again, I'm toying with the idea of writing a SqlGenerator to generate sq tables from the schema.
An user was able to recompile Dingo with .NET 2.0 beta2 and get it to work. I was able to recompile .NET 2.0 beta2, but I am getting a reflection error, which is caused by "order" xml serialization attribute.
I purchased a new machine, since running .NET 2.0 on my old P3 450mhz desktop with 512Mb of ram was painfully slow.
An user of dingo was kind enough to report an issue with .NET 2.0. I was able to track the issue down to changes in the new version. Apparently, there is a known bug which requires adding order="x" to the fields. This bug is apparently fixed and will be in the full release. In the mean time, I would suggest using .NET 1.1 to generate the source and compile with .NET 2.0. There's a link to the bug entry here... read more
Lately, I've been busy with JMeter related stuff, so I haven't had much time to explore XMI and Sql support. I will eventually get around to it. Right now I'm concentrating on writing a reporting component for JMeter.
Right now I'm contemplating supporting XMI and basic SQL generation without constraints like "not null". not sure when I'll have time to implement these features, but it's on my list of things to do.
implementing support for sql would be a nice move towards supporting hibernate mapping. I still need to decide on how I want to support enumerations in java and still don't like any of the options. I also still want to implement support for XStream converters.... read more
I am happy to announce the final release of Dingo 1.1.3. Here are the release notes.
1. JavaBuilder class now works correct for most cases, except for simpleType enumeration.
2. JavaCompiler class added. It's a modified version of Compiler class.
3. Updated ConversionUtil with language specific lookup and check for types
4. fixed a bug in some cases unbounded localElement doesn't use the right variable name... read more
I ran the unit tests tonight and fixed some bugs with the tests. I found one bug for jaxb style compilation and fixed it. tomorrow I will add one more test and release the 1.1.3.
I've posted a new releace candidate. A minor bug was found by Luke-a-tron when a sequence has no elements. the patch was minimal. If all goes well and I get time to run through the unit tests next weekend, I will make a final release of 1.1.3.
I'm also happy to announce that I've added Luke-a-tron as a developer.
I've fixed 2 bugs with reserved works lock and case. I've started adding support base64binary, which dingo currently doesn't support.
I've uploaded a new release candidate that fixes a bug with unbounded elements. In some cases the variable name is incorrect. it is fixed now.
I've posted 1.1.3 release candidate 2. I haven't updated the docs yet, but I will in the coming week.
I made some progress on the javabuilder tonight. I am able to run ipo.xsd on it and generate java code. Still need to go through the unit tests and samples to make sure it all works correctly. the code has been checked into CVS. tomorrow I will post a new release candidate.
it took less work to add the javaBuilder than I expected. Since java and C# are so similar, I started with the C# builder and make minor changes. I also created a new class called JavaCompiler. the JavaCompiler now builds to an executable named Jingo.... read more
I plan to work on the java builder a bit this weekend. If I manage to finish it, I will make a new release. Now that sourceforge has finally released their new statistics package, it's much nicer. Kudos to the sourceforge developers for all their hard work. It great to be able to slice and dice the stats.