Engineering Notebook Luis Mora

Engineering Notebook -- Luis Mora

11/23/20 - 11/29/2020

• coded solution to solve trim internal nodes bug and repetitive edges bug
• helped aaron solve more bugs
• working on documentation
  • worked on SRS and SDS revision
  • working on paper
• will continue working on paper

11/9/20 - 11/20/20

• Implemented the -L functionality for the configurator
  • used a filehandler
  • there was a problem (race?) trying to use filehandler on the fly
    • sometimes the filehandler was not created when setting level
  • found a way to format log files
  • figured out how to modify gradle's clean function to remove log files
• Corrina helped me establish better levels for logger and in better places
• fixed bug that was introduced in the configurator -- logger should log everything while the handlers decide what to print
• added check for empty log file passed
• helped aaron with using streams to handle concatenate our lists of nodes
• Aaron's node problem could be solved by handling the creation of unwanted nodes in the cleanupNodes function and cleanupEdges function instead of dynamically (leads to complicated logic/hardcoding/null pointers)-- i'll pitch it to him in the morning - it's 1 am
• created a -o option for cmd line arguments

11/2/20 - 11/6/20

• investigating examples of Logger implementations in Java
• root handler is giving issues in setting a level for the other loggers.
  • trying to figure out how to use the root handler for loggers
  • tried avoiding using the root logger -- now facing issues of double printing as a result
  • removed the root logger successfully
• adding a logger implementation to the Configurator caused some tests to fail
  • surrounded code with try catch to allows tests to pass
• current implementation of logger has a behavior in which the verbosity should be the first argument passed to have information about file/directory validation printed -- could be intended

10/26/2020 - 10/29/2020

• Reviewed code
• Recommended changes to Corrina's branch to optimize code
• Found a mistake within my code that would lead to problems later on
  • nodes are unique in the global array but not their ids
  • fixed this by giving unique ids to nodes before adding to the global array
  • fixed that the edge clean up function -- it relies on the global node array being fixed already -- order matters
• testing module's toString function led to the discovery of similar module names (difference is UPPER/LOWER case)
  • confirmed to be (diag/algo/cans) cfg files
  • fixed
• private functions have not been renamed by doxygen to reflect upper/lowercase convention
  • needs a fix
  • probably running doxygen again
  • it has not been done since aaron rewrote the modules according to convention
• found out that the number of nodes in the project are a lot less than previously thought (300s vs 1200s)
• starting to look at logger examples in java to implement in sprint 3
  • sprint 3 should add a -o and -v to the configurator and log appropriately

10/19/2020 - 10/23/2020

• have been working on changing the design of graph - node to node - edge
• merged second draft of parser --- parser absorbed job of graph manipulator
• graph objects were removed and will be handled with Node and Edge type array lists
• handling edge cases with aaron for module prefixes --- ex/ RTOS functions do not follow the XXX_function naming convention ---- ex/ __attribute__ is ignored as it produces an empty string
• removed duplicate nodes from global arraylist
• am and will work on finishing parser with team

10/12/20 - 10/17/20

• worked on initial implementation of parser
• handled major cases except edge case
• helped in refactoring the parser -- suggested using maps instead of loops to speed up process
• brainstorming/designing next step in the logic flow -- removing internal connections within manipulator
• finding and solving errors in logic flow --- are nodes added unique? how are we saving connections?
• design - should the parser make modules or the manipulator?
• thinking of test cases for parser

Sometime in the beginning

• worked on configurator -- our class that handles the command line arguements
• tried breaking it various different ways till we came up with a solid command line handler
• helped create test cases for our initial classes

Things I have learned this semester

• Markdown
• Git Links/ Wiki's
• how to work with Gradle
  • build.gradle contains the settings needed to automate the build process
  • settings.gradle specifies where the project starts
  • gradle run --args="provide java application arguments here"