Posts by Chang Park

Types of apps and frameworks

Screen Shot 2018-06-30 at 7.45.44 PM
Native Apps are built with native APIs from specific platform such as IOS, Android, and BlackBerry.

  • Pros
    • High performance.
    • Supports push notification and full device APIs.
    • Interactive and intuitive.
  • Cons
    • Platform dependent. Developer should work with different codebase for different platform.
    • Hard to maintain.
    • Expensive developing cost.
  • Frameworks that support cross-platform

 

Web Apps use html, css and java scripts and can be accessed via a web browser over a network such as chrome and safari.

  • Pros
    • Low developing cost.
    • Easy to build and maintain.
    • Cross-platform because it is accessed through internet browser.
  • Cons
    • Performance issues such as lagging.
    • Less interactive and less intuitive.
    • Limitations of device utilities such as bluetooth, NFC, camera, biometrics and sensors.
    • No push notification.

 

Progressive Web Apps are advanced version of Web Apps. These apps support push notification and offline functionality. This support is important to keep people using your app after initial usage. Still access through internet browsers. 

 

HybridApps are built with html, css, and java scripts and run inside WebView. Also, they have native source codes. These apps cover limitations of web apps. There are multiple frameworks that support to build hybrid apps(e.g. PhoneGap, Ionic, Cordova).

  • Pros
    • Support push notification and have access to device utilities using APIs.
    • No browser needed.
    • Faster development than native apps.
  • Cons
    • Still have performance issues.
    • More expensive than web apps.

 

 

Shells

bashsh

Shell is a program taking commands from a keyboard and giving them to operating system.

Below are three popular shells on Linux.

bash

  • Bourne Again Shell. The most commonly used shell on Linux.

tcsh

  • Enhanced C shell(csh).

zsh

  • The newest of the shells. Compatible with the Bourne shell. Has powerful completion features.

 

Virtual Consoles

linux terminal

Virtual Consoles enable multi-user logins on a single machine. For example, you can login with your account and also as a root.

 

By default, most Linux versions provide virtual consoles with these keys.

Ctrl + Alt + F1 – F6: Virtual Consoles

Ctl + Alt + F7: A console where X server is running.

 

X

  • is an application that manages one or more graphical displays and input devices(mouse, keyboard, etc.)
  • works as a server.

You have probably heard about X11, X server, or X window.

All of these mean the same, and X11 just means eleventh version of X.

 

You can use a command ‘w’ which shows a list of currently logged-in users. 

w result

tty represents regular terminal.

pts represents pseudo terminal slave(e.g. ssh connection).

 

 

Methodology: Qualitative vs. Quantitative

qualitative-vs-quantitative

These two terms can be used as research methods.

Qualitative

  • used to gain an understanding of underlying reasons, opinions, and motivations
  • provides insights into the problem or helps to develop ideas or hypotheses

Quantitative

  • used to quantify the problem by generating numerical data
  • uses measurable data to formulate facts and uncover patterns in research.
  • much more structured than qualitative.

 

How above methodologies can be applied in research project:

  1. Formulating hypotheses (Qualitative) – Can initiate a research by discovering the problems or opportunities people are thinking.
  2. Validating your hypotheses (Quantitative) – Can get numbers to apply statical analysis for validating hypotheses.
  3. Finding general answers (Quantitative) – Helps to get definite answers for broad questions.
  4. Incorporating the human element (Qualitative) – People’s own words can help to uncover blind spots.

 

 

Down Call vs. Up Call

downcall_upcall

Down Call

  • User code(unprivileged) enters to kernel with trap(interrupt).

Up Call

  • Kernel(privileged) enters to user code with IPC(Interprocess Communication).

 

White Box, Black Box, and Gray Box

black_and_white

In software testing, there are three different types of testing depending on how much knowledge the testers have on the software being tested.

White Box Testing

  • With full knowledge of the software. (internal structures, source codes, etc.)
  • Usually tests how the software performs.
  • Suitable for software developers.

Black Box Testing

  • Without having any knowledge of the software. (internal structures, source codes, etc.)
  • Usually tests for functionality of the software.
  • Suitable for third-party testers.

Gray Box Testing

  • Test with partial knowledge of the software.

Testing Methods

test

These are currently popular methods to test software.

Fuzz Testing

  • Input massive invalid, unexpected, or random data(Fuzz) to find errors or vulnerabilities in the software.
  • Effective on blackbox testing
  • Pros
    • Simple
    • Provides results with little effort
  • Cons
    • May miss bugs that don’t trigger a full program crash
    • May be less likely to trigger bugs that are only triggered in highly specific circumstances

 

Stochastic Testing

  • Input random sequence of datas 

 

Model-based Testing

  • Generates meaningful test case models which can detect errors on the desired behavior.

 

Symbolic Testing

  • Replace inputs of program with symbolic variables.
  • Track symbols rather than concrete input
  • When execution path diverges, fork and add constraints on symbolic values
  • Find the feasibility of all paths in the program.
  • Quite expensive for finding feasibility of path.
x = read();
if (x > 3) {
  y = 1;
  if (x < 0)
    y = 2;
} else y = 3;

After changing with symbolic variables, it checks the feasibility of paths. On above example, the expression (y = 2; on line #5) will not be reached.

  • Challenges
    • path explosion
      • Exponential in branching structure
      • Loop on symbolic variables
    • Complex code and environment dependencies
      • System calls
      • Library calls
      • Recursive calls
  • Possible Solutions
    • For “path Explosion”
      • Random search (e.g., randomly restart in a while loop, pick next path randomly), but check coverage implementation is required
      • Combined search
    • For “complex code and environment dependencies”
      • Simulate system call
      • Build simple versions of library calls
      • Run library an system calls with a concrete value
      • Summarize the loops
  • Applications:
    • Detecting infeasible paths
    • Generating test inputs
    • Finding bugs and vulnerabilities
    • Proving two code segments are equivalent (Code Hunt)

 

Concolic Testing (Concrete + Symbolic)

  • Input actual concrete value to the program, then generate new input by changing the path condition.
  • Less expensive than Symbolic testing.

 

A/B (or split, bucket) Testing

  • Comparing two versions of software which performs better
  • Statistical analysis which performed better

e.g.) randomly showing two different versions to users, and see which one had more uses

 

Boundary Testing

  • Extreme ends like Start- End, Lower- Upper, Maximum-Minimum, Just Inside-Just Outside values are called boundary values and the testing is called “boundary testing”.
  • Input variables are selected at their:
    • Minimum
    • Just above the minimum
    • A nominal value
    • Just below the maximum
    • Maximum
  • Boundary Testing comes after the Equivalence Class Partitioning.

 

Equivalence partitioning or equivalence class partitioning (ECP)

  • Software testing technique that divides the input data of a software unit into partitions of equivalent data from which test cases can be derived
  • In principle, test cases are designed to cover each partition at least once.
  • This technique tries to define test cases that uncover classes of errors, thereby reducing the total number of test cases that must be developed.
    • Pros: reduction in the time required for testing software due to lesser number of test cases

 

 

References

  • https://www.guru99.com/equivalence-partitioning-boundary-value-analysis.html
  • https://en.wikipedia.org/wiki/Equivalence_partitioning

What does the symbol ‘&’ mean in C and C++?

ampersand

‘&’ symbol has two meanings in C and C++.

  • Address of Operator: returns address of a variable.
    • int a = 3;
    • printf(“%d”, a);
      • will return value of a: 3
    • printf(“%d”, &a);
      • will return address of variable a.

 

  • Bit-wise Operator &: operates on bitlevel
    • Example:table_bitwise_operator
    • int a = 13&4;
      • 13 = 1101
      • 4  =0100
      • 13&4 = 0100
      • a = 4

compileSdkVersion, minSdkVersion, and targetSdkVersion

Screen Shot 2017-11-15 at 12.37.43 AM

compileSdkVersion

  • can choose which version of Android SDK to compile the app with
  • strongly recommended to compile with latest SDK (the app can use latest API)
  • this compile SDK will not be included in the app.
  • If using support library, the version should be matched with this compileSdkVersion.

 

minSdkVersion

  • lower bound of your app (minimum version of user devices the app can be installed).
  • Google Play Store uses to determine which devices the app can be installed on.
  • If using support libraries, it should be higher than the version of the support libraries.
    • On July 2017, minSdkVersion of all support libraries were increased to 14.
  • Check DashBoard to see current distribution of all Android versions.

 

targetSdkVersion

  • Indicates which Android version the app designed for.
  • Informs that developer tested the app against target SDK version.
  • If target SDK version is lower than running API version, system enables compatible behaviors to maintain the app’s forward compatibility.

 

Tips

compileSdkVersion = targetSdkVersion = latest Android Version

  • so that the app can work on latest Android version devices.

Recommend to check latest SDK and update the app

  • Set targetSdkversion and compileSdkVersion to latest version.
  • In that way, you can check if there will be any error when running the app on newer version.
  • You don’t need to use any new features of newer versions.

Euclid’s Algorithm

Screen Shot 2017-10-02 at 10.32.14 PM

Euclid’s(Euclidean) Algorithm

  • Algorithm to find GCD of two large numbers

GCD(N1, N2), suppose that N1 is larger than N2

N1 = N2 * q1 + r1   <– Larger number divide by Small number

N2 = r1 * q2 + r2   <– Small number divide by remainder of previous

r1 = r2 * q3 + r3

…….

N = r * q + 0    <– When remainder becomes zero, then r is the GCD

 

example

  • GCD(1701,3768)
    • 3768 = 1701*2 + 366
    • 1701 = 366*4 + 237
    • 366 = 237*1 + 129
    • 237 = 129*1 + 108
    • 129 = 108*1 + 21
    • 108 = 21*5  + 3
    • 21 = 3*7 + 0       <– 3 is the GCD of 1701, 3768