Presenter: Svetlozar Nestorov
Moderator: Ramana Yerneni
Note-taker: Joachim Hammer

Items of discussion:

1. Schema information is currently hard-coded into the function NativeQuery(). The obvious improvement is to separate the schema from the function and store it in a text file that is read in during run-time.
2. Use consistent syntax and notation, e.g., replace STRUCT's , etc.
3. Avoid cryptic "C"-style syntax and replace with more "user-friendly" synatax, e.g., use (parsedQuery != NULL) instead of (!parsedQuery), etc.

• In TM_Query:
  1. Modify matcher to return value of type condlisttype
  2. Fix memory leak caused by not freeing queryResult.
     Two issues that we need to address:
     • Specific fix: The queryResult variable gets overridden when the function PostProcess returns (queryResult is input and output. Rename one of the parameters (e.g., NativeQueryresult), and then free up the memory accessed by both variables right before calling PostProcess (this happens everytime TM_Query is called).
     • General fix: We need to devise a general scheme for allocating and de-allocating memory that is consistent with SSL/CSL.

Joachim Hammer