Static Analysis for SAP2000 API

  • Thread starter Thread starter person123
  • Start date Start date
AI Thread Summary
SAP2000 is a widely used structural analysis software with an API that allows for model creation beyond its GUI. Users have found the API beneficial for managing complex structures, as it enables record-keeping, parameter updates, and direct geometric calculations. However, challenges arise with Python's lack of static analysis capabilities, making it difficult to access component classes, methods, and their arguments directly within the IDE. This leads to a cumbersome process of referencing documentation separately and managing method arguments manually.A proposed solution involves parsing the documentation to create Python classes that mirror the API's structure, allowing for static analysis features like tooltips in IDEs. This approach, while initially perceived as complex, is considered a viable solution to enhance usability. The discussion also clarifies that the term "static analysis" may have been misused, as the desired functionality aligns more with IDE tooltips than traditional static analysis tools. An update reveals progress in implementing this solution, although some documentation inconsistencies remain.
person123
Messages
326
Reaction score
52
TL;DR Summary
I found the SAP2000 API useful for modeling a complex structure, but without static analysis there's a lot of friction. Is there a way it could be added?
For background, SAP2000 is a structural analysis software used in industry a lot and it has an API in several languages that seems to talk to a SAP2000 executable. It be used to create structural models as an alternative (though I think infrequently used) to the GUI.I started using the SAP2000 API for modeling a complex structure and I found it very helpful. I could have a record of everything that made the model, loop through repetitions in the structure, quickly update it with different parameters, and have calculations for the structure's geometry in the code directly. However, while there is good documentation, I found a big source of friction was that python (the language I'm using) couldn't do a static analysis with that documentation. (I can't see the component classes and methods of a component class, nor the arguments of a method in the IDE itself like most libraries). This means I have to search the documentation separately, copy over the method I want (often there's 3 or 4 layers of component classes), and count the arguments to check what they each refer to (there's often many arguments, some irrelevant). Then to review the arguments I passed I have to either wrap the call in a function I wrote, add a comment or just remember it. If this friction was removed, I think using the API would be faster and more robust than using the GUI in general.

Could there be a way to get static analysis of the classes and methods?

One approach I thought of is parsing through the documentation (a chm file) to get name, description and arguments of each method. Then I would generate python classes with an identical composition hierarchy who have methods that just call an API method with an identical name. It would then appear identical to calling the API method directly but static analysis would come with it. It sounds like a crazy idea to me (writing code to write code and wrapping an API in essentially another API), but I think it should work.

Would there be a better and simpler way of doing it? Are there issues with the way I'm proposing? Might there already be a way of getting static analysis?

Thanks!
 
Technology news on Phys.org
I don't understand the term "static analysis" in the way you are using it. To me static analysis is performed by tools like pylint and I don't think that is what you want.

person123 said:
Then I would generate python classes with an identical composition hierarchy who have methods that just call an API method with an identical name. It would then appear identical to calling the API method directly but static analysis would come with it. It sounds like a crazy idea to me (writing code to write code and wrapping an API in essentially another API), but I think it should work.
No, it's not a crazy idea, that's exactly what you (or someone else) needs to do (if it has not been done already by the vendor or someone else).
 
pbuk said:
I don't understand the term "static analysis" in the way you are using it. To me static analysis is performed by tools like pylint and I don't think that is what you want.
Thank you!

Yes, I'm probably not using the term static analysis correctly. What I want to see are the list of component classes and methods of a class, and the arguments of a method, you often see in IDEs. Would they just be called tooltips instead? (I was calling it static analysis because it tells you things about your program without running it).

pbuk said:
No, it's not a crazy idea, that's exactly what you (or someone else) needs to do (if it has not been done already by the vendor or someone else).

Sounds good, I think I'll look around a bit more to see if it's been done, and if not, I'll do it myself. I've never really done anything like it and it felt like a strange approach, so it's very good to know it's fairly standard. (I'm not contacting CSI directly since I only have a work account and I'm not sure if I can use it to ask about things only tangentially related to my job).
 
Update:

I added the tooltips with the method I described, with a demo attached. The documentation isn't perfect -- small format mistakes prevented me from constructing about 10% of the methods, and you can even see typos in the demo (e.g. SapMdel instead of SapModel). I could fix these issues manually though.

 
Dear Peeps I have posted a few questions about programing on this sectio of the PF forum. I want to ask you veterans how you folks learn program in assembly and about computer architecture for the x86 family. In addition to finish learning C, I am also reading the book From bits to Gates to C and Beyond. In the book, it uses the mini LC3 assembly language. I also have books on assembly programming and computer architecture. The few famous ones i have are Computer Organization and...
I have a quick questions. I am going through a book on C programming on my own. Afterwards, I plan to go through something call data structures and algorithms on my own also in C. I also need to learn C++, Matlab and for personal interest Haskell. For the two topic of data structures and algorithms, I understand there are standard ones across all programming languages. After learning it through C, what would be the biggest issue when trying to implement the same data...

Similar threads

Back
Top