This is the third time this has happened to me. I have performing
many edits to rules. So I will try something one way, check out
the generated code, then perhaps cut part of the rule, recreate it,
check out the generated code, *loop*. At some point, my code stops
generating. I will make a modification, but the generation of
source code stops.

I am including a file that I cannot get to generate code. I believe
I have sent another one previously. I even move this file to a
new Metascape project, and Metascape is still unable to generate code.

ev

Post edited by: ecasotti, at: 2008/01/19 20:23

Hi,

These aren't corrupt files, they are invalid in that there is something not specified that causes the generator not to be able to complete its generation -- just as a compiler will not produce bytecode if there is a syntax error in the source code. As I mentioned in an earlier post there is not yet support for markers like syntax error markers yet -- I'm still working out the best way to accomplish that. In the meantime you can..

1. Save early and often and look at your last changes. In this case, you were missing the projection definition in the Build Grids Act. The model built for me (with errors) after changing that.

2. Eclipse has very nice support for local histories and difs. Right-click on the .metaabm file, select Teamhow Local History. Check the Eclipse docs for more info.

3. As mentioned, the best clue is to take a look at the console log. This currently goes to your OS console I think, but I'll change to make it go to the Eclipse console.

Miles

Just an FYI, I have been careful to use Cut and not Delete. Since doing so, I have not invalidated a file in a few days. Watch me now jinx myself but I am wondering if the two Delete & Invalid files, are not related.

ev

Yep, that is very likely. Luckily, I've fixed that -- will be in the next update.

Again, let's distinguish corrupt from incorrectly defined. In the former case, the editor won't even be able to open the file.

Miles

I used the term Invalid, because you did in your previous post to distinquish from corrupt.

ev

Yep. Corrupt cases were probably caused by delete issue, and in next release.

Anyway, I'm happy to say that I have now implemented semantic checking of metaabm files so we will be able to recognize invalid issues as well. And actually Invalid is an overloaded term here, because you can also have files that can be loaded and built, but that aren't technically valid based on the XSD, and vs. vs!