DSL_textParser was not declared in htis scope

[Greg]

Hi,

I'm trying to port my Visual Studio 6.0 code to Linux and am getting the following errors:

LearningTest.cpp: In function ‘int main(int, const char**)’:
LearningTest.cpp:9: error: ‘DSL_textParser’ was not declared in this scope
LearningTest.cpp:9: error: expected `;' before ‘parser’
LearningTest.cpp:23: error: ‘EnableXdslFormat’ was not declared in this scope
LearningTest.cpp:25: error: ‘parser’ was not declared in this scope
make: *** [LearningTest.o] Error 1

I am using the Jan 16,2008 libraries, Is there a namespace defined that I need to add? Any help clearing up these errors would be appreciated.

Thanks,
Greg

[Greg]

I see now that these have been removed from the API.

DSL_dataset::ReadFile seems to fail on files with variable states which include an underscore. Is there another delimiter I can use? Currently I am using a space to separate the variable state names in the datafile.

Thanks.

[shooltz[BayesFusion]]

DSL_dataset::ReadFile seems to fail on files with variable states which include an underscore. Is there another delimiter I can use?

I can't reproduce the problem. I assume you have the variable state like "abc_123", which causes ReadFile to return false.

Maybe you're comparing the result returned from ReadFile to DSL_OKAY? This won't work, because DSL_OKAY is #defined as zero, while ReadFile returns bool.

[Greg]

Yes, I was using DSL_OKAY!!!

Thanks for the quick response.

[parseval]

I was making some experiments with the smilearn lib.
When I wounder why DSL_textParser is not recognized. When I realized that it is removed from the API !

But I'm still wondering, why it is part of the documentation in the wiki and is also used as example in the tutorials.

May be we can update this. Thank you.

[shooltz[BayesFusion]]

May be we can update this. Thank you.

Yes, you can update this. GeNIe/SMILE wiki is open to user edits.

[parseval]

Yes, you can update this. GeNIe/SMILE wiki is open to user edits.

Thank you... I didn't know this and I make some additional entries nows. Many thanks for this very useful fundament in bayesian network analysis. May be we can even cooperate to deliver some new features, like "online learning" .