heechan.yang

IntroductionCITRUS: C++ unIt Testing for Reliable and Usable SoftwareCITRUS is an automated unit testing tool for C++ programs. This tool analyzes the target software, generates and mutates test cases (a.k.a test harness, drivers, method sequence) to explore diverse states of a program in the goal of increasing test coverage. Existing QuestionWhen a test case is generated, citrus only tests this..

Thoughts while TestingTargetting a FunctionWith 1 process of 6 hours of test, GraphFuzz only covered 50.9% line coverage.Some endpoint drivers are called a significantly smaller amount of times than others (due to randomness).I have experimented by removing other endpoint driver in schema in hopes of testing certain function more.By doing so, GraphFuzz was able to cover more lines in that specif..

LibFuzzer is a coverage guided fuzzing engine. This engine repeatedly executes a target with fuzzer input.LibFuzzer selects a sample input from a corpus.It mutates the input. [Optional: user can include custom implemented cross over function and a mutator function.]It runs the target driver with the mutated input.If it produces new coverage, the input is put back to the corpus, otherwise, it is ..

ContentsOverviewEndpoint & GraphsHow it WorksSimple Demo: CarPersonal ThoughtsReferences1. OverviewGraphFuzz is a software testing tool. It is focused to API testing by invoking API functions at valid order that are mutated from a structure of a graph. With this mechanism, GraphFuzz is able to find bugs that arise from invoking API functions in a certain order. The main objective of GraphFuzz is..

ContentsIntroductionInstallationhello_graphfuzz ReproductionSkia ReproductionReferences1. Introduction [1]GraphFuzz is an API-based testing tool to effectively measure test coverage. "GraphFuzz models sequences of executed functions as a dataflow graph, thus enabling it to perform graph-based mutations both at the data and the execution trace level" [1]. Using model (structure or grammar) to inf..

ContentsIntroductionInstalling DependenciesApproach to SolutionUnderstanding the Concept of Branch CoverageSummaryReferences1. IntroductionMeasuring coverage of a software at source code level is to test how much of existing branches have been fallen through with given test inputs. Therefore, existing branches are first identified. Execution of software comes after for measuring the branch cover..