Overview: The notebook:open_book: of junior software engineer:keyboard:

This notebook is and will be with me in the path of a software engineer working on 3D stuffs. This notebook and is constantly evolving.

The topic ranges from programming language like C#, C++, Python, to 3D Software I mostly use like Rhino, Blender, and even to some Algorithm.

Table of Content

• Overview
• Table of Contents
• 0.Installation
• 1.Computer System
  • Git
  • Shell
  • LaTex
• 2.Programming Language
  • C#
  • C++
    • Compile
      • CMake
      • GCC
    • C++ Libraries
      • Eigen
  • Python
• 3.IDE & Text Editor
  • Visual Studio IDE
  • Visual Studio Code
• 4.Devops
  • Documentation Tool
  • Docker
• 5.Geometry Kernel
  • RhinoCommon & Grasshopper
• 6.Algorithm
• 7.Robotics

0.Installation:gear:

Vulkan

​ 1.Install Vulkan SDK

https://vulkan.lunarg.com/sdk/home#windows

​ 2.Vulkan Driver

It is already inside Nvidia Driver.

https://www.nvidia.com/Download/index.aspx?lang=en-us

CUDA, cudnn

​ 1.Install CUDA

https://developer.nvidia.com/cuda-downloads

​ 2.Download cudnn

https://developer.nvidia.com/cudnn-download-survey

(follow the instruction and you should be good)

GCC

​ Install from Ubuntu

$ sudo apt update
$ sudo apt install build-essential
$ sudo apt-get install manpages-dev
$ gcc --version  #verified you have already installed

CMake

​ Install from Ubuntu

1. First you have to make sure you have already gcc which contains make command for C binaries.
2. Make sure you installed OpenSSL package

$ sudo apt-get install libssl-dev
$ wget http://www.cmake.org/files/v3.21/cmake-3.21.2.tar.gz
$ tar xzf cmake-3.21.2.tar.gz
$ cd cmake-3.21.2
$ ./configure --prefix=/opt/cmake
$ make  #for compilation
$ sudo make install  
$ /opt/cmake/bin/cmake -version  #check if installed

Qt

​ Download and Install

1. https://www.qt.io/
2. Click option 'Downloads for open sources users'
3. When select download package, MSVC 2019 64-bit
4. After install Add YOUR_PATH_TO\Qt\6.2.2\msvc2019_64\bin to Environment Variable

​ Configure in Visual Studio

1. In Visual Studio, 'Extension' - 'Manage Extension' , search and install 'Qt Visual Studio Tools'
2. Reboot Visual Studio
3. In Visual Studio, 'Extension' - 'Qt VS Tool'
   1. In 'Version' tab, add the C:\Qt\6.2.2\msvc2019_64\bin to the path where it is also the location of qmake.exe
   2. In 'Legacy Project Format' tab, change 'Ask before checking out files' from True to False

SQLite

​ Download

1. https://sqlitebrowser.org/

vcpkg

​ Download and Install

1. Navigate to C:\\
2. Recommend PowerShell for the following procedure
3. git clone the repo

> git clone https://github.com/microsoft/vcpkg
> .\vcpkg\bootstrap-vcpkg.bat

​ Remove Outdated Packages

C:\vcpkg> .\vcpkg remove --outdated --recurse

​ Build Dependency Visual Studio Can Find

C:\vcpkg> .\vcpkg integrate install

​ When Something is terribly slow

e.g. powershell core, download by yourself, put it in the download folder.

​ Install Default Architecture

e.g. Install only x64. Make a system variable named: VCPKG_DEFAULT_TRIPLET, value: x64-windows

​ Install like a Nuget Package

C:\vcpkg> .\vcpkg integrate remove

Then you can see something like:

Install-Package vcpkg..........

Copy the command and run it in Nuget Package Manager Command in Visual Studio.

1.Computer System:desktop_computer:

Git🕷️

0.Basic Concepts

📌VCSs, Git, Github/Gitlab

VCSs = Version control systems (VCSs)

Git = Git is the de facto standard for version control

Github/Gitlab/Gitee = the host of Git Repository

📌 Snapshots咔嚓

Git models the history of a collection of files and folders within some top-level directory as a series of snapshots.

<root> (tree)
|
+- foo (tree)
|  |
|  + bar.txt (blob, contents = "hello world")
|
+- baz.txt (blob, contents = "git is wonderful")

📌Modeling history: relating snapshots

In pro words: a history is a directed acyclic graph (DAG) of snapshots

In human words: each snapshot in Git refers to a set of “parents”, the snapshots that preceded it.

      this is a commit
            ↑
o <-- o <-- o <-- o
            ^
             \
              --- o <-- o

      new_feature      new_feature + bug_fix
            ↑               ↗
o <-- o <-- o <-- o <---- o
            ^            /
             \          v
              --- o <-- o
                        ↓
                     bug_fix

📌 Fast-forward and three-way merge

Fast-forward: the commit all points to a same parent commit

o <-- o <-- o <-- o <-- o

Three-way merge:

• master and sub no conflicts:heavy_check_mark:, different file
• master and sub no conflicts:heavy_check_mark:, different modification in the same file
• master and sub conflicts:x:

o <-- o <-- o <-- o <---- o
            ^            /
             \          v
              --- o <-- o

📌 Data model

The following mimics the data model in Git in pseudocode.

File: it is a bunch of bytes

type blob = array<byte>

Directory: It contains named files and directories

type tree = map<string, tree | blob>

Commit: It has parents, metadata, and the top-level tree

type commit = struct 
{
    parents: array<commit>
    author: string
    message: string
    snapshot: tree
}

Object: It could be a blob, tree, or commit.

type object = blob | tree | commit

Data Storage: In Git data store, all objects are content-addressed by their SHA-1 hash.

objects = map<string, object>

def store(object):
    id = sha1(object)
    objects[id] = object

def load(id):
    return objects[id]

References: They are pointers to commits. Convert SHA-1 hash to human-readable names.

references = map<string, string>

def update_reference(name, id):
    references[name] = id

def read_reference(name):
    return references[name]

def load_reference(name_or_id):
    if name_or_id in references:
        return load(references[name_or_id])
    else:
        return load(name_or_id)

e.g.

HEAD is the latest "where we currently are"

master refers to a particular snapshot instead of a bunch of hexadecimal string.

📌 Repositories

In short, a Git repository: it is the data objects and references.

📌 A diagram for Git

1.Frequently Used Commands

command	objective	example
git add <file_name>	Add file to staging area	git add README.md
git add .	Add any unstaged files to staging area
git add -p <file>	Interactively choose hunks of patch
git blame	see each commit with authors
git branch	see all the branches
git branch <branch_name>	create a new branch	git branch dev
git branch -d <the_local_branch>	Delete local branch	git branch -d PointDebug
git cat-file -p <SHA-1 hash>	Visualize data by SHA-1 hash
git checkout <file>	remove the unstaged changes and back to current stage
git checkout <branch>	change HEAD to such branch	git checkout dev
git checkout <commit_guid>	change HEAD to such commit
git checkout -b <branch>	create a new branch and checkout to it	git checkout -b dev
git clean -f -d	clean all the untracked files and folders
git clone <url>	clone the repo from url
git clone --shallow	clone the repo without any history
git clone --recursive <url>	clone the repo recursively
git commit	"Archive and confirm" the changes to the directory
git commit -m"<message>"	Same with 👆, but with a short message	git commit -m"Update README.md"
git commit -a -m"<message>"	-a stands for add; combine add and commit	git commit -a -m"Update README.md"
git commit --amend	append current commit with previous commit
git config -l	-l stands for listing the detail of config, e.g. user name
git diff
git diff --staged	compare the un-commit changes with HEAD
git diff <file>	see diff of file with HEAD
git diff <guid> <file>	see diff of file between current HEAD and guid
git diff <prev_guid> <now_guid> <file>	see diff of file between previous and now
git fetch <remote>	fetch the commits from remote to local
git help <command>	help menu of such command	git help commit
git init	Initialize a git repo
git log	the log of git history
git log --stat	the log of git history statistically
git log --all --graph --decorate	nice diagram of git log
git merge <branchX>	merge branchX into current branch	git merge origin/master
git mv <old_name> <new_name>	rename specific file (this has to be commited)
git push	push commits to remote(already configured remote)
git push <remote> <branch>:<branch>	push commits to remote	git push origin main:main
git push <remote> --delete <the_remote_branch>	Delete remote branch	git push origin --delete PointDebug
git pull	git pull = {git fetch; git merge <remote>/<branch>}
git rm --cached <file>	To stop tracking files which have already been tracked	git rm --cached main.3dm.bak
git reset	remove all the staged changes, green=>red
git reset --hard <commitGuid>	Destroy any local modification and reset to such commit	git reset --hard 0d1d7fc32
git reset HEAD --hard	Reset the modified files back to the HEAD
git revert <commit_guid>	Reverting undoes a commit by creating a new commit.
git show <guid>	check specific commit by guid	git show 721d6bd
git stash	hide current untracked changes
git stash pop	pop out the hidden untracked changes

2.:+1:Goooood resources of Git

.gitignore template

​ https://github.com/github/gitignore

Software for Git

​ SourceTree is a free software managing Git while it provides GUI to interact with Git. Highly recommend! You can download:

​ https://www.sourcetreeapp.com/

Book for git

​ https://git-scm.com/book/en/v2

3.Some Regular Procedure

📌 How to solve conflicts?

Suppose you are on master, and you want to merge new_feature branch, the conflict is on main.cpp. Then you can do:

git merge new_feature

You will see the feedback from the prompt saying there are conflicts. You open the main.cpp which was decorated with some notation from git . Here I use VS Code, you can also use Vim.

code main.cpp

You manually fix the conflicts. Then you should add it to the staging area!(because you literally fix it)

git add main.cpp

Finally you ask git to continue the merging job.

git merge --continue

📌 Setup for recursive clone

You want to combine several dependencies into one project when you work on a macro project.

1. create a .gitmodules in the root of your git folder.
2. define what git repo will be in which location of your main folder.

[submodule "<name_of_this_repo"]
	path = <location_of_this_submodule>
	url = <url_of_this_repo>

//e.g.
[submodule "deps/polyscope"]
	path = deps/polyscope
	url = https://github.com/nzfeng/polyscope.git
[submodule "deps/googletest"]
	path = deps/googletest
	url = https://github.com/google/googletest.git

📌Git submodules

git submodule add --depth 1 git@gitee.com:shanghai-dajie-robot/pocket_raichu.git submodules/pocket_raichu
git config -f .gitmodules submodule.pocket_raichu.shallow true

git config -f .gitmodules submodule.submodules/pocket_raichu.shallow false

git submodule foreach git pull origin master

git clone <repo_url>
git submodule init
git submodule update --depth 10

📌Setup a Github Access Token

1. Go to the Github account - Developer Setting - Generate Token
2. Git clone an arbitrary repo from your page
3. When the computer request credential, just close it until it appears on the command line for the following info
   1. user_name: the name of your Github account
   2. password: paste your token here

Shell:shell:

​ In short, shell is a type of user interface. It is either command-line interface (CLI) or graphical user interface (GUI). This section will mainly on (CLI) especially on Bourne Again SHell, or “bash” for short.

0.Basic Concept

📌 environment variable(PATH)

All the commands you run in the shell has already been added to the environment variable in your PC. In Windows, you can search edit the system environment variable => environment variable.. => Edit the PATH.

📌 / and \

A path on the shell is a delimited list of directories;

on Linux and macOS:

​ path separated by /

​ / means the "root"

on Windows:

​ path separated by \

​ C:\ means the "root"

📌 absolute path and relative path

​ a.Absolute path is with full path (literally). e.g.

​ In windows,

C:\Users\Xingxin\AppData\Local\Temp

​ In Linux and macOS:

​ path starts with / is absolute

/home/tutorial

​ b.Relative path takes the advantage of environment variable to form an absolute path. e.g.

 %USERPROFILE%\AppData\Local\Temp

📌 . , .. , ~ , - used in path

It is a must to know these 3 symbols for they appear a lot.

. means current directory, e.g. cd ./tutorial change dir to the tutorial folder which under current dir.

.. means the parent directory, e.g. cd .. change dir to its parent dir.

~ means the home directory. e.g. ssh-add ~/.ssh/id_rsa add the ssh key in the "HOME folder/.ssh/"

- means the previous directory in the prompt

📌 - , -- used in command

- indicates a flag which modify their behavior

-- indicates an options

example:

ls -l

The -l flag asks the ls command to list the files in a a long listing format.

example:

git log --stat

The --stat option indicates the git command to show the statistics from git log.

📌 , "" , \ used in file name and folder name

The space is used to separate the arguments in the command line. e.g.

mkdir my photo

This will create 2 folders which are "my" and "photo". If you want to create a folder with "my photo". You either use "" to concatenate or use \ to escape the space.

mkdir "my photo"
mkdir my\ photo

📌' ' and " "

Things inside single quote ' ' are literal string. Meaning what is inside is what is inside.

Things inside double quote " " are strings. They can be substituted.

📌 <, > , >> in data stream

< take the data stream out...

> take the data stream in...

>> take the data stream append in...

e.g.

cat <hello.txt >README.txt

It means stream out whatever inside hello.txt and stream in README.txt

cat <hello.txt >>README.txt

It means stream out content of hello.txt and append them at the end of README.txt

📌 | pipe

It means take the output from the left as the input of the right. <A=output> | <A'=input>

e.g.

ls -l | tail -n3

ls -l prints out all the files. This will be treated as the input of tail

tail -n3 receives the print by ls -l and filter the last 3 lines.

📌 r, w, x file/folder permissions

r , read

​ file: read

​ dir: allow? to see files in this dir

w, write

​ file: write

​ dir: allow? to rename/remove files in this dir

x, execute

​ file: execute

​ dir: allow? to enter this dir

d, directory

-, nope

In the beginning, there are 10 characters.

0: indicates if this a directory. d: directory; -: a file

1 to 3: the permission of file owner

4 to 6: the permission of the owner group

7 to 9: the permission of other users

e.g.

drwxr-xr-x 1 Xingxin 197121      0 Aug  5 01:15  Autodesk/
-rwxr-xr-x 1 Xingxin 197121   2475 Jul 28 20:01 'Unreal Engine.lnk'*
-rw-r--r-- 1 Xingxin 197121    282 Jul  1 01:43  desktop.ini

For Autodesk/ folder, drwxr-xr-x

​ 0=d, this is a directory:heavy_check_mark:, file:x:

​ 123=rwx, owner can read✔️ , write✔️ and execute✔️ .

​ 456=r-x, owner group can read✔️ , write❌ and execute✔️ .

​ 789=r-x, other users can read✔️ , write❌ and execute✔️ .

For desktop.ini file, -rwxr-xr-x

​ 0=-, this is a directory:x:, file:heavy_check_mark:

​ 123=rw-, owner can read✔️ , write✔️ and execute❌ .

​ 456=r--, owner group can read✔️ , write❌ and execute❌ .

​ 789=r--, other users can read✔️ , write❌ and execute❌ .

📌 $ , #, sudo in prompt

$ indicates this is a normal user permission to this shell

# indicates this is a super user permission to this shell

sudo means super user do

​ running echo in normal user mode

$ echo hello

​ update the apt-get

$ sudo apt-get update

​ enter super user mode(in this mode, you don't need to add sudo anymore). su means shift user

$ sudo su
# echo "I am in the super user mode"
# exit

📌Wildcards matching by ? and *

? substitute with 1 character.

* substitute with following characters.

e.g. Suppose you have main.py main1.py main2.py main3.py

$ ls main?.py
main1.py  main2.py  main3.py  #It will substitute anything in `?` position

$ ls ma*
main.py main1.py main2.py main3.py  #It will match all the characters after *

📌 Use { } as a set

💡(It works very similar to the philosophy of list matching in Grasshopper.)

⭐You can see it as the command will iterate what is inside { }

$ mkdir {dev,src,master}  #create 3 folder at 1 time

$ mv *{.py,.sh} folder  # Will move all *.py and *.sh files

$ convert image.{png,jpg}  #This is equal to `convert image.png image.jpg`

$ touch main{1..28}.py  #It will create main1.py, main2.py ... all the way to main28.py

📌Use find to do recursive search

​ 1. Find folders in current dir

• . means current dir
• d means the search target is directory

$ find . -name dev -type d

​ 2. ⭐Find files in depth!!

• **/bin/*.dll means no matter what is the front, the most important pattern is **/bin/*.dll
• f means the search target is file

$ find . -path '**/bin/*.dll' -type f

​ 3. Find the files been modified

• -mtime means modified time
• -1 means last day

$ find . -mtime -1

​ 4. Find the files and delete them

• *.tmp all the temporary files
• -exec rm execute them with remove command

$ find . -name "*.tmp" -exec rm {} \;

​ 5. Find files by sizes

$ find . -size +500k -size -10M -name '*.tar.gz'  # Find all zip files with size in range 500k to 10M

📌What is shebang?

​ Different names:

It is also called hashbang, pound-bang, or hash-pling.

​ Typical Looking:

Always starts with #! at the beginning of a file.

​ Objective:

Increase the portability of the script. It will make use of the PATH environment and resolve to wherever the command lives in the system.

​ Example:

#!/usr/bin/env python3 , Execute with a Python interpreter, using the env program search path to find it.

📌Arguments

As you see the above case, there are other definitions of arguments in bash.

• $0 - Name of the script
• $1 to $9 - Arguments to the script. $1 is the first argument and so on.
• $@ - All the arguments
• $# - Number of arguments
• $? - Return code of the previous command. Similar to return 0 in C++
• $$ - Process identification number (PID) for the current script
• !! - Entire last command, including arguments. A common pattern is to execute a command only for it to fail due to missing permissions; you can quickly re-execute the command with sudo by doing sudo !!
• $_ - Last argument from the last command. If you are in an interactive shell, you can also quickly get this value by typing Esc followed by .

​ Example of using !!

Suppose you are at ~

$ mkdir /mnt/new  #the prompt will say you don't have permission
$ sudo !!  #Here means `sudo mkdir /mnt/new`

​ Example of using $_

In case you are in ~/Desktop/Dev

$ cd ..  #Now you are at ~/Desktop
$ cd $_  #Now you are at ~

​ Example of taking the variable

$ foo=$(pwd)
$ echo foo
~/Desktop

📌⭐Use grep to find codes

​ 1. grep basic

This is a super useful command! It simply check if A(string/text) is in B(file).

$ grep "import" main.py
import numpy as np
import sympy as sp
import pandas as pd

Btw, you can check if the return value of the last command

$ echo $?
0

0 means successfully executed.

​ 2. grep recursively with -R

You can recursively grep such pattern

$ grep -R "numpy" .  #it will find all the files contained "numpy"

  3.`grep` the history

The following is a pipe. Take the output from history command and grep if there is find in that history

$ history | grep find

📌 rg (R.I.P. grep)

​ Recursive grep by rg

rg <grep_pattern> -t <type_of-file> <search_location>

$ rg "import numpy as np" -t py ./dev

​ Find files does not match Regex pattern

e.g. find files not a shebang

$ rg -u --files-without-match "^#\!" -t sh

Find with stats

• -C 5 means search the surrounding 5 lines

$ rg "import requests" -t py -C 5 --stats ~/dev

📌 || and &&

​ A || B

First check A=True✔️ , runs A only.

First check B=False❌, run B

​ A && B

First check A=True✔️, runs B

First check A=False❌, does not run anything.

$ true || echo "This will not be printed"

$ false || echo "This will be printed"
This will be printed

$ true && "This will be printed"
This will be printed

$ false && "This will not be printed"

📌Use < to take the output as input

As mentioned before, the < means taking the data out. We can take advantage of this syntax.

$ diff <(ls dev) <(ls main)

The difference between the content of dev folder and main folder will be printed.

1.Frequently Used Commands

Command	Objective	Example
cat	catch whatever inside a file	cat hello.txt
cd	change directory
cp	copy a file
echo	like "echo", it simply prints out its arguments
find	find <folder> -name <name> -type <type>	find . -name main -type f
fd	shortcut for find(not installed yet)	fd "*.py"
history	list the history of your typed bash commands
ls	list all the files in current directory
man	manual of something	man rm
mkdir	make a directory/folder
mv	rename/move a file	mv xx.md yy.md
pwd	present working directory
rm	remove a file
rm -r	remove all the files recursively	rm -r ./img
rm -rf *	remove all the files at the current folder
rmdir	remove EMPTY folder	rm ./.vscode
rg	R.I.P. grep....:laughing: recursively grep	rg "import" -t py ~/dev
tail	print the last n lines	tail -n3
touch	create a file	touch main.cpp
shellcheck	Debug bash file	shellcheck mcd.sh
<command> --help	see the function of this command	git --help
man <command>	open the menu of this command	man ls
Ctrl+L	clean out the shell

2.Shell Scripting

⭐Big Picture: What is shell scripting for? Creating command pipelines, saving results into files, and reading from standard input.

📌Scripting 101

foo=bar
echo "$foo"
# prints bar
echo '$foo'
# prints $foo

​ 1.No space should insert when assigning value

foo=bar ✔️ , foo = bar❌ . Because space acts as delimiter.

​ 2.double quote and single quote have been explained.

\

​ 3. $ for interpolation.

It is very similar to C#. But $ in shell scripting has way more powerful functionalities.

int day = 28;
string word = $"Today is {day}-th days of this month.";

📌Scripting 102

​ The shell script is ended with .sh

$ vim mcd.sh

​ Make a script for 1.mkdir and 2. cd to it

​ Here $1 means the 1st argument the user put

mcd()
{
	mkdir -p "$1"
	cd "$1"
}

​ After complete the script, store it into the shell

$ source mcd.sh

​ Now you can use this command in shell

$ mcd dev

📌Scripting 103

Suppose you have dev.py main.py main1.py

Objective:

1. Find import numpy as np in all the files
2. If not, add it

Make a new bash file:

$ vim check_numpy.sh

Script it:

echo "Starting program at $(date)" # Date will be substituted

echo "Running program $0 with $# arguments with pid $$"

for file in "$@"; do
    grep "import numpy as np" "$file" > /dev/null 2> /dev/null
    if [[ $? -ne 0 ]]; then
        echo "File $file does not import numpy package, adding one"
        echo "import numpy as np" >> "$file"
    fi
done

Use it:

$ ./check_numpy.sh main.py main1.py dev.py

Output:

Starting program at Sun Sep 12 14:12:49     2021
Running program ./check_numpy.sh with 3 arguments with pid 431
File main.py does not import numpy package, now adding one
File main1.py does not import numpy package, now adding one
File dev.py does not import numpy package, now adding one

Explanation:

1. $(date) , take the output from command date and turn into string
2. $0 , the current bash script name
3. $# , num of arguments
4. $$ , current pid
5. for file in "$@" , loop over the all the arguments(the file names)
6. "$file" , turn file into plain text for comparing in grep
7. /dev/null 2> /dev/null , normally there are 2 output from grep which are STDOUT and STDERR. So this code means to store the OUT/ERR to the location.
8. [ [ ] ] , when comparing, use double brackets as recommended.
9. -ne , means not equal to , therefore if grep can't find numpy which is not 0, then it will execute the following.
10. echo XXX >> "$file" , append text into file

3.Others

📌 what is .\xxx？

Suppose we run dir in windows(≈ls in Unix). Then you will see:

2021/07/28  10:20    <DIR>          .
2021/07/28  10:20    <DIR>          ..
2021/07/28  09:41               141 .gitattributes
2021/07/28  09:41    <DIR>          .github
2021/07/28  09:41             5,559 .gitignore
2021/07/28  09:41                 0 .vcpkg-root
2021/07/28  09:41               100 bootstrap-vcpkg.bat
2021/07/28  09:41               109 bootstrap-vcpkg.sh
2021/07/28  10:06    <DIR>          buildtrees
2021/07/28  09:41         1,210,046 CHANGELOG.md
2021/07/28  10:21                93 CMakeDirectory.txt
2021/07/28  09:41             2,432 CONTRIBUTING.md
2021/07/28  09:41    <DIR>          docs
2021/07/28  10:06    <DIR>          downloads
2021/07/28  09:58    <DIR>          installed
2021/07/28  09:41             1,096 LICENSE.txt
2021/07/28  09:41             2,334 NOTICE.txt
2021/07/28  10:06    <DIR>          packages
2021/07/28  09:41    <DIR>          ports
2021/07/28  09:41            12,381 README.md
2021/07/28  09:41            15,908 README_es.md
2021/07/28  09:41            12,459 README_fr.md
2021/07/28  09:41            13,961 README_ko_KR.md
2021/07/28  09:41            11,716 README_zh_CN.md
2021/07/28  09:41    <DIR>          scripts
2021/07/28  09:41               923 shell.nix
2021/07/28  09:41    <DIR>          toolsrc
2021/07/28  09:41    <DIR>          triplets
2021/07/28  09:44         3,359,624 vcpkg.exe
2021/07/28  09:41    <DIR>          versions

Therefore, running the following

.\vcpkg 

means run the vcpkg.exe in current directory. You can even specify the params:

.\vcpkg search libigl

In this case, it means the author of vcpkg leave the argument for you to enter when they build this app.

To conclude, .\xxx means execute the xxx in current directroy.

📌How to install oh-my-zsh

1. Close VPN and install Windows Terminal and Ubuntu from Microsoft Store
2. sudo apt-get update
3. sudo apt-get upgrade
4. sudo apt install zsh
5. sh -c "$(curl -fsSL https://github.com/ohmyzsh/ohmyzsh/master/tools/install.sh)"
   1. If it failed. Go to Internet Properties , find out the Proxy and Port
   2. export https_proxy=http://127.0.0.1:7890 http_proxy=http://127.0.0.1:7890 all_proxy=socks5://127.0.0.1:7890
6. git clone --depth=1 https://github.com/romkatv/powerlevel10k.git ${ZSH_CUSTOM:-$HOME/.oh-my-zsh/custom}/themes/powerlevel10k
7. vim ~/.zshrc
8. Set ZSH_THEME="powerlevel10k/powerlevel10k"
9. Open the Setting.json of Windows Terminal, and paste the guid of Ubuntu to the default terminal.

4.Resources of Shell

ShellCheck - A shell script static analysis tool

​ https://github.com/koalaman/shellcheck

Unix env

​ https://www.man7.org/linux/man-pages/man1/env.1.html

tldr package

​ https://tldr.sh/

LaTex

Symbol and Letters

Appearance	Code
$\tilde{M}$	\tilde{}
$\equiv$ , equivalent to	\equiv
$\not\equiv$ , not equivalent to	\not\equiv
$\bar{A}$	$\bar{A}$
$\hat{A}$	$\hat{A}$
$\tilde{A}$	$\tilde{A}$

Vectors and Matrix

📌 vertical dashed lines inside a matrix

​ The c in {c:c} represents the column while : represents the dashed line.

​ Therefore, {c:c} means the dashed line start from 1 column from the left.

\begin{array}{c:c}
1 &  2 \\ 
3 & 4 \\
5 & 6 \\ 
\end{array}

$\begin{array}{c:c} 1 & 2 \ 3 & 4 \ 5 & 6 \ \end{array}$

📌 Matrix Template

\begin{bmatrix}
1&0&0\\
0&1&0\\
0&0&1\\
\end{bmatrix}

$\begin{bmatrix} 1&0&0\ 0&1&0\ 0&0&1\ \end{bmatrix}$

📌Determinant Template

\begin{vmatrix}
1 & 4 & 1 \\ 
0 & -2 & 1 \\ 
0 & 2 & 1
\end{vmatrix}

$\begin{vmatrix} 1 & 4 & 1 \ 0 & -2 & 1 \ 0 & 2 & 1 \end{vmatrix}$

Documentation

📌 Reference equation by tag

​ You create the equation with label and tag.

$ a = b \tag{a}\label{a} $

​ Then you can reference the equation:

$ \eqref{a} $

2.Programming Language:speaking_head:

C#

1.Types

1.1Value Types

1.2Reference Types

interface

📌define an interface and implement it

例如我定义了

interface IParam
{
    Vector3d WorldX { get; }
}

然后我也显示实现了WorldX

public class Param : IParam
{
    Vector3d IParam.WorldX
    {
        get => new Vector3d(1, 0, 0);
    }
}

❌ 但是为什么我无法调用World?如下

Vector3d IParam.SafeShift
{
    get => IParam.WorldX * 2.5;
}

仔细想想，IParam是什么？它是接口，接口怎么会有instance(实例化)？！？对吧？理解这个东西之后，我们需要把IParam显示指定实例化this，这样：

Vector3d IParam.SafeShift
{
    get => ((IParam)this).WorldX * 2.5;
}

📌difference between explicit and implicit implementation

首先我定义了接口

interface IParam
{
    double X { get; }
    double Y { get; }
}

explicitly implement an interface

我显示实现(explicitly)：

public class ParamP : IParam
{
    double IParam.X { get => 1; }
    double IParam.Y { get => 2; }
}

public class ParamZ : IParam
{
    double IParam.X { get => 10; }
    double IParam.Y { get => 20; }
}

那么我在用的时候，需要这样：

IParam pp = new ParamP() as IParam;
IParam pz = new ParamZ() as IParam;

或者这样：

ParamP pp = new ParamP();
ParamZ pz = new ParamZ();
IParam p;
p = flag ? pp as IParam : pz as IParam;

​ implicitly implement an interface

我隐式实现：

public class ParamP : IParam
{
    public double X { get => 1; }
    public double Y { get => 2; }
}

public class ParamZ : IParam
{
    public double X { get => 10; }
    public double Y { get => 20; }
}

那我就可以直接用了：

ParamP pp = new ParamP();
ParamZ pz = new ParamZ();

为什么这么设计呢？

你可以想象interface前面都是virtual，如果你不public，怎么override它呢？

2.Keywords

2.1Modifiers

Access Modifiers

📌private vs internal

internal is for assembly scope (i.e. only accessible from code in the same .exe or .dll)

private is for class scope (i.e. accessible only from code in the same class).

2.2.Statement

2.2.1.Iteration Statement

📌 How to loop backward in foreach?

//use `.Reverse()`
int[] arr = new int[]{1,2,3,4,5,6};
foreach(int i in arr.Reverse())
{
    Console.WriteLine(i.ToString());
}
//Reverse list return void, so first cast to array
List<int> list = new List<int>{1,2,3,4,5,6};
foreach(int i in list.ToArray().Reverse())
{
    Console.WriteLine(i.ToString());
}

2.2.2.Selection Statement

2.3Method Parameters

📌 how to use out？

public static void Run(out List<Plane> planes, out List<string> commands)
{
    //用out的话，一定要在最前面赋值！！
    planes = new List<Plane>();
    commands = new List<string>();
    //然后对这些List进行操作
    planes.Add(Plane.WorldXY);
    planes.Add(Plane.WorldYZ);
    planes.Add(Plane.WorldZX);
    commands.Add("Hello!");
    commands.Add("My name is John!");
}

在调用的时候是这样：

List<Plane> ps = new List<Plane>();
List<string> ss = new List<string>();
Run(out ps, out ss);

2.4.Namespace

2.5.Generic Type Constraint

2.6.Access

2.7.Literal

2.8.Contextual

2.9.Queries

📌Select没有返回值的item

​ 假设你有List<Curve> crvs，你想对它们整体偏移。那么你可能会：

using System.Linq;
var shiftCrvs = crvs.Select(crv => crv.Translate(100, 0, 0)).ToList();

​ 但是实际上你可能得到的是一系列boolean， 因为

public bool Translate(
	double x,
	double y,
	double z
)

​ 因此应该怎么做呢？

var shiftCrvs = crvs.Select(crv => {crv.Translate(100,0,0); return crv;}).ToList();

过程其实一目了然，

1.先选择crv为crvs里的每一row，

2.然后{}是对crv的操作。

3.该操作是，先Translate，然后return该已经位移的crv

📌use Take to split a list

​ imagine you have a list of 10 double, and you want to split the list into two chunks

List<double> doubleList = new List<double> { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
var firstFour = doubleList.Take(4).ToList();  //now you have the first four item

📌use Skip to bypass items

​ it is sort of dual method of Take which bypass certain amount of elements

List<double> doubleList = new List<double> { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
var lastSix = doubleList.Skip(4).ToList();  //now you have the last six item

📌Create a list of booleans like in Python

using System.Linq;

List<double> Z;
var allFalse = Enumerable.Repeat(false, Z.Count);

📌 Combine two list and return a new list

var z = x.Concat(y).ToList();  //the `concat` would not affect either x or y

📌 Sort a list by list value

using System.Linq;

//sorts in acending order {1,2,3,4,5}
myList.Sort();

//Sorts in decending order {5,4,3,2,1}
myList.Sort();
myList.Reverse();

📌 Set operation

Method Name	Description
Distinct	Removes duplicates
Except	Returns the set difference
Intersect	Returns the set intersection
Union	Returns the set union

using System.Linq;

List<int> LA = new List<int>() { 1, 2, 2, 3 };
List<int> LB = new List<int>() { 3, 4, 5, 6 };
var LC = LA.Distinct().ToList();  //{1, 2, 3}
var LC = LA.Except(LB).ToList();  //{1, 2}
var LC = LA.Intersect(LB).ToList();  //{3}
var LC = LA.Union(LB).ToList();  //{1, 2, 3, 4, 5, 6}

‼️ If you notice, the result are all set no matter if it .Distinct(). That is due to the Set operation literally.

📌Pro Way to Query a List

Don't use List[0], rather than FirstOrDefault()

📌Find a Single element from a List

Suppose you have an id, you have to query an object from a list which matches the id.

string plateId  = "5ca99c57-2cc2-4f65-b5db-83afca088264";
var plate = PlateDtoList.SingleOrDefault(dto => dto.Id == plateId);

3.Operators and expressions

4.Statements

Selection statements

📌 Multiple condition in ONE case

case [data_type] [data_name] when [condition]

switch (this.diameter)
{
    case 6:
        return 35.0;
    case 8:
        return 20.0;
    case int n when (n >= 16 && n <= 28):
        return 20;
    default:
        throw new Exception($"No data for {this.diameter}mm steel.");
}

5.Special Character

6.Attributes

7.Unsafe code and pointers

8.Preprocessor directives

9.Compiler Error

10.XML documentations comments

11.Specifications

12.C# Nuget Package

Name	Link	Function
Super Simple TCP	https://github.com/jchristn/SuperSimpleTcp	TCP client and server
MVVM Tooklit	https://docs.microsoft.com/en-us/windows/communitytoolkit/mvvm/introduction	Successor of MVVM Light

13.未整理

C++

C++ header template

/************************************************************************\


Copyright (c) 2021 HE, Xingxin


Module Name:

	xxx.


Notes:
	[Optional]


History:
	Created on mm/dd/yyyy xingxin.he@mail.polimi.it


\************************************************************************/

1.C4996, fopen: This function or variable may be unsafe.

当你用Visual Studio IDE去编译C++的时候，经常会遇到

'fopen': This function or variable may be unsafe. 
'fscanf': This function or variable may be unsafe. 

✔️ 解决方案是：

Select your project and click Properties in the context menu.

In the dialog, chose Configuration Properties -> C/C++ -> Preprocessor

In the field PreprocessorDefinitions add ;_CRT_SECURE_NO_WARNINGS to turn those warnings off.

然后，

Configuration Properties > C/C++ > Advanced property page.

Edit the Disable Specific Warnings property to add 4996

⚠️ 需要注意的是，上面更改Properties的地方是x64，但是你Build的时候有可能是x86！！！因此，最好properties两个platform都改。

2.C2440, cannot convert from char * to LPWSTR

Project=>Properties=>Configuration Properties=>Advanced=>Character Set=>Use Multi-Byte Character Set✔️

3.E0135 namespace "std" has no member "filesystem"

Because std::filesystem is a feature of C++17.

Project=>Properties=>Configuration Properties=>C/C++=>Language=>C++ Language Standard=>ISO C++17 Standard (/std:c++17)✔️

4.How do you know Project Property Inheritance?

On a property page, a property in bold is defined in the current context. A property in normal font is inherited.

e.g. The following bold label like C++ Language Standard is defined by myself.

5.Use Property Manager to apply your setting across different projects

If you are tired of setting Property manually in VS IDE, you can set up a property sheet(.props) to apply the setting.

View > Property Manager or View > Other Windows > Property Manager

Compile

CMake

📌Install from Ubuntu

1. First you have to make sure you have already gcc which contains make command for C binaries.
2. Make sure you installed OpenSSL package

$ sudo apt-get install libssl-dev
$ wget http://www.cmake.org/files/v3.21/cmake-3.21.2.tar.gz
$ tar xzf cmake-3.21.2.tar.gz
$ cd cmake-3.21.2
$ ./configure --prefix=/opt/cmake
$ make  #for compilation
$ sudo make install  
$ /opt/cmake/bin/cmake -version  #check if installed

📌CMake 101

​ Setup⚙

Suppose you have the following file structure:

├── CMake101
│   ├── CMakeLists.txt
│   ├── main.cpp

The CMake file looks like this:

#specify the minimum CMake version
cmake_minimum_required(VERSION 3.10)

#get the directory name as the project name, stored in the variable ProjectId
get_filename_component(ProjectId ${CMAKE_CURRENT_SOURCE_DIR} NAME)
#    this REPLACE means replacing the space to "_" in case the project including white space
string(REPLACE " " "_" ProjectId ${ProjectId})

#set the project
project(${ProjectId})

# add the executable, cmake will automatically find the header files
add_executable(${ProjectId} main.cpp)

​ Build🔨 - option1 (recommend✌)

When you are in the folder CMake101, open the shell. Do the following commands:

$ cmake -B build -G "Visual Studio 17 2022"

The -G stands for generator. You can search cmake generator on explorer to find out more options.

After this command, CMake will create a subfolder called build in this folder with makefile. The structure becomes:

.
└── 📁CMake101
    ├── 📃CMakeLists.txt
    ├── 📁build
    │   ├── ALL_BUILD.vcxproj
    │   ├── ALL_BUILD.vcxproj.filters
    │   ├── CMake101.dir
    │   ├── CMake101.sln
    │   ├── CMake101.vcxproj
    │   ├── CMake101.vcxproj.filters
    │   ├── CMakeCache.txt
    │   ├── CMakeFiles
    │   ├── Debug
    │   ├── ZERO_CHECK.vcxproj
    │   ├── ZERO_CHECK.vcxproj.filters
    │   └── cmake_install.cmake
    └── 📃main.cpp

After you modify the codes, and you really want to build it(this is equivalent to press the Build button in VS IDE). You can:

$ cmake --build build

The structure becomes:

.
└── 📁CMake101
    ├── 📃CMakeLists.txt
    ├── 📁build
    │   ├── ALL_BUILD.vcxproj
    │   ├── ALL_BUILD.vcxproj.filters
    │   ├── CMake101.dir
    │   ├── CMake101.sln
    │   ├── CMake101.vcxproj
    │   ├── CMake101.vcxproj.filters
    │   ├── CMakeCache.txt
    │   ├── CMakeFiles
    │   ├── Debug
    │   │   ├── CMake101.exe  👈
    │   │   └── CMake101.pdb  👈
    │   ├── ZERO_CHECK.vcxproj
    │   ├── ZERO_CHECK.vcxproj.filters
    │   └── cmake_install.cmake
    └── 📃main.cpp

​ Build🔨 - option2

When you are in the CMake101 folder, you type:

$ mkdir build
$ cd build
$ cmake ..

The preceding operation is equivalent to the previous one. This is common but in-and-out a folder seems complicate to me.

.
└── 📁CMake101
    ├── 📃CMakeLists.txt
    ├── 📁build
    │   └── ... 👈(you are here)
    └── 📃main.cpp

📌CMake102

In this section, we will demonstrate how to add a library in CMake. (code)

​ Setup⚙

Suppose you have the following files:

.
└── 📁CMake102
    ├── 📃CMakeLists.txt
    ├── 📃main.cpp      👈#include "math.h" in the top
    ├── 📃math.cpp      👈suppose this is the library
    └── 📃math.h        👈suppose this is the library

​ CMake

cmake_minimum_required(VERSION 3.10)
project(cmake102)

#🤚 add library here.
#                     1st argument is the library name
#                     The STATIC means static library, in Windows is `.lib`
#                     The SHARED means dynamic link library, in Windows is `.dll`
#                     3rd argument is the compile source of the library
add_library(math_lib STATIC math.cpp)

add_executable(cmake102 main.cpp)

#add linker to the project
#                      1st argument is the project
#                      3rd argument is the library
target_link_libraries(cmake102 PUBLIC math_lib)

​ Build🔨 - (MinGW and make)

$ cmake -B build -G "MinGW Makefiles"

After that, you would have

.
└── 📁CMake102
    ├── 📃CMakeLists.txt
    ├── 📁build
    │   ├── CMakeCache.txt
    │   ├── CMakeFiles
    │   ├── Makefile             👈this is the Makefile
    │   └── cmake_install.cmake
    ├── 📃main.cpp
    ├── 📃math.cpp
    └── 📃math.h

Change to build folder then simply type make:

$ cd build
$ make

​ Build with STATIC in CMake

You will have:

.
└── 📁CMake102
    ├── 📃CMakeLists.txt
    ├── 📁build
    │   ├── CMakeCache.txt
    │   ├── CMakeFiles
    │   ├── Makefile             
    │   ├── cmake102.exe          👈this is the executable
    │   ├── cmake_install.cmake   
    │   └── libmath_lib.a         👈this is the static library, since I use gnu, so it is `.a`
    ├── 📃main.cpp
    ├── 📃math.cpp
    └── 📃math.h

​ Build with SHARED in CMake

You will have:

.
└── 📁CMake102
    ├── 📃CMakeLists.txt
    ├── 📁build
    │   ├── CMakeCache.txt
    │   ├── CMakeFiles
    │   ├── Makefile
    │   ├── cmake102.exe         👈this is the executable
    │   ├── cmake_install.cmake
    │   ├── libmath_lib.dll      👈this is the shared library
    │   └── libmath_lib.dll.a  
    ├── 📃main.cpp
    ├── 📃math.cpp
    └── 📃math.h

📌Key Difference between SHARED and STATIC library🌟

For SHARED library, the library should be with executable.

✔

│   ├── cmake102.exe
│   ├── libmath_lib.dll  

❌

│   ├── cmake102.exe
│   ├── libmath_lib.dll  

For STATIC library, you can delete the .lib or .a file. Since everything in the library has been written into the executable.

✔

│   ├── cmake102.exe

✔

│   ├── cmake102.exe
│   ├── libmath_lib.a   

📌Choose SHARED or STATIC when you press CMake

You can modify the following line in CMakeLists.txt:

add_library(math_lib STATIC math.cpp)

to

add_library(math_lib math.cpp)

You can rather do it like this:

​ SHARED

$ cmake -B build -G "MinGW Makefiles" -DBUILD_SHARED_LIBS=ON

​ STATIC

$ cmake -B build -G "MinGW Makefiles" -DBUILD_SHARED_LIBS=OFF

GCC

📌Install from Ubuntu

$ sudo apt update
$ sudo apt install build-essential
$ sudo apt-get install manpages-dev
$ gcc --version  #verified you have already installed

📌What is GCC and G++?

gcc: GNU Compiler Collections(for compile C and C++)

g++: GNU c++ compiler(for compile C++)

📌Simplest Compile

g++ <fileName.cpp> . The default build is a.out in Linux and a.exe in Windows

$ g++ helloworld.cpp
$ ./a.exe
Hello, world!

📌Use -o to specify output name

g++ helloworld.cpp -o helloworld

📌Compile in C++11 standard

Use -std=c++11

g++ -g -Wall -std=c++11 main.cpp

📌Frequently Used Commands

https://www.runoob.com/cplusplus/cpp-environment-setup.html

Flag or Option	Notes
-Wall	Generate all the errors

📌****

📌****

📌****

📌****

📌****

📌****

📌****

📌****

C++ Libraries:books:

Eigen

📌 Initialization before operation is a must

#include <Eigen/Dense>

Eigen::Vector3d rhs;
{
    //operations on `rhs`
}

In this situation, the rhs will be super super big like $e^{23} $. Therefore, a good practice is to initialize before using it.

#include <Eigen/Dense>
#include <Eigen/Core>

//you may have some specific size
int numRow = 10;
int numColumn = 10;

//declare
Eigen::MatrixXd matrix;
Eigen::VectorXd rhs;
matrix.resize(numRow, numColumn);
rhs.resize(numRow);

//init with zero
matrix.setZero();
rhs.setZero();

Matrix Class

📌 Class Overview

The concept of vector and matrix are all in this class since vector can be see as a $n\cross 1$ matrix.

📌 Tricks to Know the Type

​ MatrixXd , X=dynamic the dimension can be define later, d=double this is a double type matrix.

​ Vector3f , 3=3rows, f=float.

📌 Mandatory Variable

There are 3 mandatory variables in Matrix class.

Matrix<typename Scalar, int RowsAtCompileTime, int ColsAtCompileTime>

​ Scalar is the scalar type, i.e. the type of the coefficients.

​ RowsAtCompileTime and ColsAtCompileTime are the dimension of matrix.

Therefore you can define as follow:

typedef Matrix<float, 4, 4> Matrix4f;
typedef Matrix<double, 3, 1> Vector3d;

📌 Vector as a matrix

In Eigen, vector can be seen as a special case of matrix, with either 1 row or 1 column.

​ column vector (default)

typedef Matrix<float, 3, 1> Vector3f;

​ row vector

typedef Matrix<int, 1, 2> RowVector2i;

📌 Dynamic as X

Use Dynamic to indicate that the size is unknown at compile time.

​ Fixed size

typedef Matrix<double, 3, 3> Matrix3d;

​ Dynamic size

typedef Matrix<double, Dynamic, Dynamic> MatrixXd;
typedef Matrix<int, Dynamic, 1> VectorXi;

📌Constructor

​ Matrix Init:heavy_check_mark: , Size Init:heavy_check_mark:, Coefficients Init:x:

Matrix3f a;

​ Matrix Init:heavy_check_mark: , Size Init:x:, Coefficients Init:x:

MatrixXf b;

​ Matrix Init:heavy_check_mark: , Size Init:heavy_check_mark:, Coefficients Init:x:

MatrixXf a(10,15);
VectorXf b(30);

​ Matrix Init:heavy_check_mark: , Size Init:heavy_check_mark:, Coefficients Init:heavy_check_mark:

//Vector
Vector2i a(1, 2);
Vector3d b(5.0, 6.0, 7.0);
Vector4d c(5.0, 6.0, 7.0, 8.0);
Matrix<int, 5, 1> b {1, 2, 3, 4, 5};  //column vector
Matrix<int, 1, 5> c = {1, 2, 3, 4, 5};  //row vector

//Matrix
MatrixXi a {      // construct a 2x2 matrix
      {1, 2},     // first row
      {3, 4}      // second row
};
Matrix<double, 2, 3> b {
      {2, 3, 4},
      {5, 6, 7},
};

📌Coefficient accessors

The accessor and constructor are the same... which is the overloaded bracket ( ).

#include <iostream>
#include <Eigen/Dense>
 
using namespace Eigen;
 
int main()
{
  MatrixXd m(2,2);
  m(0,0) = 3;
  m(1,0) = 2.5;
  m(0,1) = -1;
  m(1,1) = m(1,0) + m(0,1);
  std::cout << "Here is the matrix m:\n" << m << std::endl;
  VectorXd v(2);
  v(0) = 4;
  v(1) = v(0) - 1;
  std::cout << "Here is the vector v:\n" << v << std::endl;
}

m(1,0) = 2.5 is $M_{1,0}=2.5$

📌 Comma-initialization

//Fixed size
Matrix3f m;
m << 1, 2, 3,
     4, 5, 6,
     7, 8, 9;
cout << m;

//Dynamic size
MatrixXd M(3,3);
M << 1, 0, 0,
     0, 1, 0,
     0, 0, 1;
cout << M << endl;

📌 Size, Resize, and conservativeResize

rows() gets numbers of rows in this matrix.

cols() gets numbers of columns in this matrix.

size() gets numbers of coefficients in this matrix.

int main()
{
  Matrix4d m;
  std::cout << "The matrix m is of size "
            << m.rows() << "x" << m.cols() << std::endl;
  std::cout << "It has "
      		<< m.size() << "elements." << std::endl;
}

​ :warning: Only dynamic-size matrix can use resize() to change its dimension.

int main()
{
  //Matrix resize
  MatrixXd m(2,5);
  m.resize(4,3);
  //Vector resize
  VectorXd v(2);
  v.resize(5);
}

​ :warning:resize() may change the coefficients but conservativeResize() preserve its original value

int main()
{
	MatrixXd m = MatrixXd::Random(3, 3);
	cout << "m =" << endl << m << endl;

	m.conservativeResize(4, 4);  //the value of coefficients in the row 4 and col4 are not init
	cout << m << endl;
}

📌 Assignment are reference type

MatrixXf a(2,2);
MatrixXf b(3,3);
a = b;
std::cout << "a is now of size " << a.rows() << "x" << a.cols() << std::endl;

📌 When should use Fixed or Dynamic?

	Fixed	Dynamic
Choice	for very small sizes where you can	for larger sizes or where you have to
Memory	Stack	Heap

🏁 The fixed-sized matrix is nothing but a plain array.

//the followings are equivalent
MatrixXf mymatrix(rows,columns); 
float *mymatrix = new float[rows*columns]; 

⚠️ When the dimension is extremely big, performance has no difference between two choice.

‼️ When you try to use fixed-size for an extremely big data like a Mesh(e.g. $V=1500;F=600$), it could result a stackoverflow.

📌 Complete Variables

​ The first 3 has been discussed above. The following will focus on the last 3.

Matrix<typename Scalar,
       int RowsAtCompileTime,
       int ColsAtCompileTime,
       int Options = 0,
       int MaxRowsAtCompileTime = RowsAtCompileTime,
       int MaxColsAtCompileTime = ColsAtCompileTime>

​ Options indicate how we iterate the element in a matrix.

For example a $3\cross 4$ Matrix. $$ \begin{bmatrix} 8& 2& 2& 9\ 9& 1& 4& 4\ 3& 5& 4& 5\ \end{bmatrix} $$

​ Iterate in column-major(default):

Matrix<int, 3, 4, ColMajor> A;
A << 8, 2, 2, 9,
	 9, 1, 4, 4,
	 3, 5, 4, 5;
cout << "In memory (column-major):" << endl;
for (int i = 0; i < A.size(); i++)
  cout << *(A.data() + i) << "  ";

It will print: 8 9 3 2 1 5 2 4 4 9 4 5.

​ Iterate in row-major:

Matrix<int, 3, 4, RowMajor> A;
A << 8, 2, 2, 9,
	 9, 1, 4, 4,
	 3, 5, 4, 5;
cout << "In memory (row-major):" << endl;
for (int i = 0; i < A.size(); i++)
  cout << *(A.data() + i) << "  ";

It will print: 8 2 2 9 9 1 4 4 3 5 4 5.

​ MaxRowsAtCompileTime and MaxColsAtCompileTime are fixed upper bound known at compile time. Which is very handy to avoid dynamic memory allocation

Matrix<float, Dynamic, Dynamic, 0, 3, 4>

Matrix and vector arithmetic

📌 Addition, subtraction, multiplication, and division

Nothing special, since the +,-,*,/ have been overloaded by Eigen.

The compound version is also fine: +=, -=, *=, /=

📌Transposition and conjugation , (possible aliasing effect)

The transpose $A^T$, conjugate $\bar{A}$, and adjoint (i.e., conjugate transpose) $A^*$ of a matrix $A$ or vector $A$ are obtained by the member functions

• transpose()
• conjugate()
• adjoint()

⚠️WARNING!! transpose() and adjoint() simply return a proxy object without doing the actual transposition. If you want to assign its transpose, you need to use transposeInPlace().

​ :x: WRONG EXAMPLE, so called aliasing effect

Matrix2i a; a << 1, 2, 3, 4;
cout << "Here is the matrix a:\n" << a << endl;
 
a = a.transpose(); // !!! do NOT do this !!!
cout << "and the result of the aliasing effect:\n" << a << endl;

​ :heavy_check_mark:CORRECT EXAMPLE

MatrixXf a(2,3); a << 1, 2, 3, 4, 5, 6;
cout << "Here is the initial matrix a:\n" << a << endl;

a.transposeInPlace();
cout << "and after being transposed:\n" << a << endl;

📌 Matrix-matrix and matrix-vector multiplication , (no aliasing effect)

​ nothing special, just use * and *=.

There are matrix $M=\begin{bmatrix}1&2\3&4\end{bmatrix}$ , and vector $u=\begin{bmatrix}-1\1\end{bmatrix}, v=\begin{bmatrix}2\0\end{bmatrix}$

Matrix2d M;
M << 1, 2,
3, 4;
Vector2d u(-1,1), v(2,0);

​ $MM$, matrix * matrix

std::cout << "matrix * matrix:\n" << M*M << std::endl;

​ $Mu$ , matrix * vector

std::cout << "matrix * vector:\n" << M*u << std::endl;

​ $u^TM$ , column-vector * matrix

std::cout << "column-vector * matrix:\n" << u.transpose()*M << std::endl;

​ $uv^T$ , column-vector * row-vector

std::cout << "column-vector * row-vector:\n" << u*v.transpose() << std::endl;

📌 dot product and cross product

​ just use .dot() , .cross() , and .adjoint()

#include <iostream>
#include <Eigen/Dense>
 
using namespace Eigen;
using namespace std;
int main()
{
  Vector3d v(1,2,3);
  Vector3d w(0,1,2);
 
  cout << "Dot product: " << v.dot(w) << endl;
  double dp = v.adjoint()*w; // automatic conversion of the inner product to a scalar
  cout << "Dot product via a matrix product: " << dp << endl;
  cout << "Cross product:\n" << v.cross(w) << endl;
}

📌 arithmetic reduction operations

.sum() , the sum of all coefficients

.prod() , the product of matrix

.maxCoeff() , the maximum of all coefficients

.minCoeff() , the minimum of all coefficients

.trace() , the sum of diagonal , equivalent to .diagonal().sum()

#include <iostream>
#include <Eigen/Dense>
 
using namespace std;
int main()
{
  Eigen::Matrix2d M;
  M << 1, 2,
         3, 4;
  cout << "M.sum():       " << mat.sum()       << endl;
  cout << "M.prod():      " << mat.prod()      << endl;
  cout << "M.mean():      " << mat.mean()      << endl;
  cout << "M.minCoeff():  " << mat.minCoeff()  << endl;
  cout << "M.maxCoeff():  " << mat.maxCoeff()  << endl;
  cout << "M.trace():     " << mat.trace()     << endl;
}

And we have 10, 24, 2.5, 1, 4, 5

⭐ You can also get back the index of row and column by using &<index>

#include <iostream>
#include <Eigen/Dense>
 
using namespace std;
int main()
{
  Eigen::Matrix2d M;
  M << 1, 2,
         3, 4;
  Eigen::RowVector3i V;
  V << 9, 8, 7;
    
  std::ptrdiff_t i, j;
  
  //Get the min and its index of a matrix
  float minOfM = M.minCoeff(&i,&j);
  cout << minOfM << i << j << endl;

  //Get the max index of a vector
  int maxOfv = V.maxCoeff(&i);
  cout << maxOfv << i << endl;
}

Array class and coefficient-wise operations

⭐ Big picture before diving in. The design of Array class aims to conduct coefficient-wise(element-wise) operation.

📌 Array types

The types of Array class is very similar to Matrix in addition to its element-wise operation. You can define it in the way of Matrix class.

Array<typename Scalar, int RowsAtCompileTime, int ColsAtCompileTime>

There are also some common types already defined in Eigen.

Type	Typedef
Array<float,Dynamic,1>	ArrayXf
Array<float,3,1>	Array3f
Array<double,Dynamic,Dynamic>	ArrayXXd
Array<double,3,3>	Array33d

📌 Accessing values inside an array

access and assign values are no difference with Matrix class.

#include <Eigen/Dense>
#include <iostream>
 
using namespace Eigen;
using namespace std;
 
int main()
{
  ArrayXXf  m(2,2);
  
  // assign values by row-col
  m(0,0) = 1.0; m(0,1) = 2.0;
  m(1,0) = 3.0; m(1,1) = m(0,1) + m(1,0);
  
  // print
  cout << m << endl << endl;
 
  // using the comma-initializer
  m << 1.0,2.0,
       3.0,4.0;
     
  // print
  cout << m << endl;
}

📌 Element-wise arithmetic operation

​ Array-scalar operation

#include <Eigen/Dense>
#include <iostream>
 
using namespace Eigen;
using namespace std;
 
int main()
{
  ArrayXXf a(3,3);
  a << 1,2,3,
       4,5,6,
       7,8,9;

  // Subtracting a scalar from an array
  cout << "a - 2 = " << endl << a - 2 << endl;
  //
  //  -1  0  1
  //   2  3  4
  //   5  6  7
}

​ Array-Array operation

#include <Eigen/Dense>
#include <iostream>
 
using namespace Eigen;
using namespace std;
 
int main()
{
  ArrayXXf a(2,2);
  ArrayXXf b(2,2);
  a << 1,2,
       3,4;
  b << 5,6,
       7,8;
  cout << "a * b = " << endl << a * b << endl;
  // 5 12
  // 21 32
}

📌 Other coefficient-wise operations

.abs()

.sqrt()

.min()

📌 Converting between array and matrix expressions

.array() , Matrix => Array

.matrix() , Array => Matrix

📌 Type-safe function in Eigen

✔️ These are legal.

​ temporarily convert to matrix before matrix-operation , store in matrix.

Eigen::Matrix = Eigen::Array.matrix() * Eigen::Array.matrix()

​ temporarily convert to array before array-operation(element-wise) , store in matrix.

Eigen::Matrix = Eigen::Matrix.array() * Eigen::Matrix.array()

​ temporarily convert to array before array-operation(element-wise), store in array

Eigen:Array = Eigen::Matrix.array() * Eigen::Matrix.array()

❌ This is illegal.

Eigen::Matrix = Eigen::Matrix * Eigen::Array

📌 TWO Example to explain Array-Matrix operation

There is const .cwiseProduct(.) method for element-wise operation in matrix directly, since the matrix may always require element-wise operation but store its result in matrix. The following two are equivalent.

Eigen::Matrix = Eigen::Matrix.array() * Eigen::Matrix.array()

Eigen::Matrix = Eigen::Matrix.cwiseProduct(Eigen::Matrix)

int main()
{
  MatrixXf m(2,2);
  MatrixXf n(2,2);
  MatrixXf result(2,2);
 
  m << 1,2,
       3,4;
  n << 5,6,
       7,8;
  //M-M
  result = m * n;
  cout << "-- Matrix m*n: --" << endl << result << endl << endl;
    
  //(M->A) - (M->A)
  result = m.array() * n.array();
  cout << "-- Array m*n: --" << endl << result << endl << endl;
  
  //no explicitly cast, direct element-wise operation
  result = m.cwiseProduct(n);
  cout << "-- With cwiseProduct: --" << endl << result << endl << endl;

  //(M->A), then do element-wise operation
  result = m.array() + 4;
  cout << "-- Array m + 4: --" << endl << result << endl << endl;
}

int main()
{
  MatrixXf m(2,2);
  MatrixXf n(2,2);
  MatrixXf result(2,2);
 
  m << 1,2,
       3,4;
  n << 5,6,
       7,8;
  //(M->A), do element-wise, (A->M), do matrix product; case 1
  result = (m.array() + 4).matrix() * m;
  cout << "-- Combination 1: --" << endl << result << endl << endl;
  //(M->A), do element-wise, (A->M), do matrix product; case 2
  result = (m.array() * n.array()).matrix() * m;
  cout << "-- Combination 2: --" << endl << result << endl << endl;
}

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

📌 ****

Python

3.IDE & Text Editor 📝💻

Visual Studio IDE

Property Tab

Post-build Event

​ Use this relative path instead which can run on anyone's PC

​ 1st line means: copy anything in your TargetDir(usually \bin) to the path you defined

​ 2nd line means: copy the ***.dll file to the location and rename it to ***.gha

XCOPY "$(TargetDir)*" "$(USERPROFILE)\AppData\Roaming\Grasshopper\Libraries\$(ProjectName)\" /S /Y
Copy "$(TargetPath)" "$(USERPROFILE)\AppData\Roaming\Grasshopper\Libraries\$(ProjectName)\$(ProjectName).gha"

Common Tricks

📌Info on Lines and Cursor

如上图所示，我光标在2.5;后面，这里显示Ln: 697 Ch: 61代表第697行，第61个字符的位置

📌How to modify output folder of Build?

在C#里，可以在Properties的Build tab修改，那C++呢？在 General tab.

📌Shortcut of Bookmarks

Ctrl+K,K,double click K. Then the bookmark will be opened.

📌Code Snippets

I think this can be very handy! Code snippets is sort of alias of the code. Similar to the alias of Grasshopper component. So, how to invoke code snippets?

Ctrl+K,B open the Code Snippets Manager. You can find all the code snippets here!! For example, if you left

tryf

in the code editor, the cursor is right after tryf, you double tab!! IDE would generate the following code for you.

try 
{	        

}
finally
{

}

📌Ctrl+Tab to change tab

Ctrl+Tab You can loop over the files you opened.

Ctrl+Shift+Tab You can loop over the files you open reversely.

📌Close all the tabs

Managing many opening tabs is bit distraction for me. Sometimes I want to close all the tabs by:

Alt+W,L

📌 Ctrl+, for quick search ⭐

This is very handy!! When I was working, I only left one file opened. And the fastest way to find the method in which file is to

Ctrl+, It will navigate you to the where the code is exactly!!

See above animation, I don't have any files opened until I press Ctrl+, and navigate to where I am interested it.

📌Full screen mode

Shift+Alt+Enter You can access/quit full screen mode by this shortcut.

📌Collapse code

Ctrl+M,M

📌Separate Windows on a same file

If on Class.cs, click Window > New Window to split a file into 2 windows.

📌Relative Path when compile

The syntax is following:

@"pack://application:,,,/{PROJECT_NAME};component This is the root path.

public Uri imagePath = new Uri(@"D:\slowbrow\pocket_slowbro\src\UiVersion_1\_UCComponents\Icons\Vector.png");
public Uri imagePath = new Uri(@"pack://application:,,,/UiVersion_1;component\_UCComponents\Icons\Vector.png",UriKind.Absolute);

Visual Studio Code

1.Shortcut

Function	Keyboard	Notes
Auto-format	Alt+Shift+F

Vim

4.Database

SQLite

📌SQLite C# 101

​ Solution Overview

Suppose you have 2 projects in Visual Studio.

• DemoLibrary, the project with data structure
• WinFormUI, the project with UI

​ Define Data Structure in DemoLibrary

The following is a simple data structure for a person.

public class PersonModel
{
    public int Id { get; set; }
    public string FirstName { get; set; }
    public string LastName { get; set; }

    public string FullName
    {
        get
        {
            return $"{ FirstName } { LastName }";
        }
    }
}

​ Setup in DB Browser

1️⃣ Init .db file for the project

In DB Browser for SQLite, click File > New Database > {Location of Executable Project} > Name.db

Normally, the .db file should be in the executable project. e.g. The root path of WinForm / WPF / UWP, etc.

2️⃣Include .db file in your porject

In Visual Studio > Solution Explorer > Show All Files > Right Click the .db File > Include in Project

3️⃣Create table matching data structure

In DB Browser for SQLite, click Database Structure > Create Table

The field is the Property you defined in C# class. The preceding image is the table matching the data structure you defined.

The options of Type are:

• INTEGER
• TEXT
• BLOB, files
• REAL, floating number
• NUMERICAL

The flags are:

• NN = Not Null
• PK = Primary Key
• AI = Auto Increment
• U = Unique

​ Add References

1️⃣Nuget Packages

• System.Data.SQLite.Core , installed in project with data structure and UI executable, e.g. DemoLibrary&WinFormUI
• Dapper , installed in project with data structure, e.g. DemoLibrary

2️⃣Search in Assemblies

• System.Configuration , project with data structure installed in DemoLibrary

​ Set-up Configuration in UI Executable

In WinFormUI, locate at App.Config. Add connectionStrings tab into the configuration. Data Source points to where .db locates.

<configuration>
	<connectionStrings>
		<add name="Default" 
			 connectionString="Data Source=./DemoDB.db;Version=3;" 
			 providerName="System.Data.SqlClient"/>
	</connectionStrings>
</configuration>

​ Create methods load and save data

public class SqliteDataAccess
{
    //this is the method loading data from database
    public static List<PersonModel> LoadPeople()
    {
        using(IDbConnection cnn = new SQLiteConnection(LoadConnectionString()))
        {
            //query database
            var output = cnn.Query<PersonModel>(
                "select * from Person",
                new DynamicParameters());
            return output.ToList();
        }
    }

    //this is the method saving data from database
    public static void SavePerson(PersonModel person)
    {
        using (IDbConnection cnn = new SQLiteConnection(LoadConnectionString()))
        {
            //match the input with the data structure
            cnn.Execute(
                "insert into Person (FirstName, LastName) values (@FirstName, @LastName)", 
                person);
        }
    }

    //private function for init connection to database
    private static string LoadConnectionString(string id = "Default")//the default here is in `App.Config`
    {
        return ConfigurationManager.ConnectionStrings[id].ConnectionString;
    }
}

4.Devops

Documentation Tool:books:

Sandcastle

• Configure Visual Studio:
  • Build > Output tab of your project's properties.
  • XML documentation file ✔

1.如何设置docfx的template？

普通：

{
  ...
  {
    "build" :
    {
      ...
      "template": "custom",
      ...
    }
  ...
}

带文件路径的：

{
  ...
  {
    "build" :
    {
      ...
      "template": ["default", "X:/template/custom"],
      ...
    }
  ...
}

Docker

5.Geometry Kernel:triangular_ruler:

RhinoCommon&Grasshopper:rhinoceros:🦗

1.Point3d和Plane

它们都是struct，因此是value type，所以不用Duplicate或者Clone，当它们是double那种值类型就好了。a.k.a. RhinoCommon的doc里面写着structure的，都是struct!!

Point3d pt1 = new Point3d(0, 0, 0);
Point3d pt2 = new Point3d(1, 0, 0);
pt2 = pt1;
pt1.X = 10;
pt2.Transform(Transform.Translation(new Vector3d(100, 0, 0)));

A = pt1;
B = pt2;

因为是值类型，那么可以想象得到，A是{10, 0, 0}，B是{100, 0, 0}。

2.如何在SolveInstance外调用RuntimeMessage

之前我加RuntimeMessage都是这样，在

protected override void SolveInstance(IGH_DataAccess DA)
{
    AddRuntimeMessage(GH_RuntimeMessageLevel.Warning, "Warning!");
}

但如果我写了一个Class, 那个class里面有方法想要AddRuntimeMessage怎么办？？

可以这样写那个Class

public class ClassName
{
    public ClassName()
    {
        
    }
    public void Run(GH_ActiveObject obj, ref List<Plane> planes, ref List<string> commands)
    {

        obj.AddRuntimeMessage(GH_RuntimeMessageLevel.Warning, "xxx");
    }
}

那到时候在SolveInstance就可以

protected override void SolveInstance(IGH_DataAccess DA)
{
    var instance = new ClassName();
    instance.Run(this, planes, commands);
}

❓为啥要用this? 其实AddRuntimeMessage是该类的方法，因此obj代指当前的实例，然后在SolveInstance那里，我会把this是当前实例

3. Grasshopper的RuntimeError

AddRuntimeMessage(GH_RuntimeMessageLevel.Warning, "No intersection is found from two edges.");

这个RuntimeMessageLevel有4层，Blank(任何信息都没有)，Remark(有信息pop-out，但是没颜色)，Warning(橙色warning)，Error(红色error警告)

4.:thumbsup: Resources for Rhino/Grasshopper developer

📌 Template to share a common library with both a C++ and a C# plug-in

​ https://github.com/dalefugier/Moose

5.Frequent Used Methods

📌 Get .3dm file path

string path = Rhino.RhinoDoc.ActiveDoc.Path;

6.Algorithm

Optimization

1.RANSAC？

综述

It can be interpreted as an outlier detection method!! Inlier和Outlier很重要，这个算法还有把inlier看作是一个很重要的指标。如下图的Line Fitting问题，如果是用 simple least squares method的话就很容易出问题。因为后者把outlier也当作关键信息，而RANSAC也会process outliers，但是它却会exclude它们。

怎么运作？

大致的思路是

1. 随机筛选一些subset of data，然后fit model出来
2. 找剩余的data来测试这个model，如果能通过，那就称为inlier data
3. 不断重复，如果fit的data太少了，这个model就会被reject
4. 若良好，那么这个dataset就是consensus set
5. 这个model不断完善，这个consensus set越来越大

⭐Pros and Cons

这非常重要，一项技术总有trade-off

Pros

若含超过50%的inliers，那么RANSAC会非常robust

Cons

反之，若只有例如30%的inlier，那么RANSAC没法找出好东西

RANSAC是model-dataset一一对应的关系，即一个dataset不可能有两个model。比如是一个折线的scatter plot，那就需要两个model，RANSAC不适合这种情况

no upper bound计算时间，因此需要根据数据集的复杂程度决定iteration大小

⭐RANSAC的threshold是problem-specific thresholds，例如在我找折线的那个电池，是算的threshold是data point的Y值的差异。那么，当我的模型从1mm变成1000mm，threshold就要改变了

7.Robotics🦿

1.法兰盘的XYZ和Target的XYZ

关系应该如下图：

X轴-红色

Y轴-绿色

Z轴-蓝色

记住，法兰盘的Z轴是怼出来的。因此右手法则应该向着法兰盘

⚠️但是这个XY轴并不一定如此！！！！！下面举例，三大机器人品牌，ABB, KUKA, FANUC它们末端法兰的区别。

从左到右，KUKA, FANUC, ABB

2021-06-10

1.KUKA界面

注释；修改指令

密码：KUKA

一般使用模块（module）

module-更改属性-会出现.DAT， .SRC

STOP

forward and backward

custom menu，装好之后有按钮；e.g. 点焊和弧焊

Menu

文件-存档(备份backup)

USB(KCP)示教器上面的USB接口

配置

1.外部启动条件

2.专家

3.其它 - 干涉区域

4.机器配置 - 轴值的范围；软限位(轴值的限制)

5.Submit解释器，SPS是KUKA开机之后就在运行的，一般是运行的逻辑，可以理解为C#的field，监控global variables，防止电流过大，或者拿工具头出错

显示

1.输入/输出

2.实际位置(相当于fanuc的POS)

3.变量(专家模式)

计数器：这个测节拍经常要用到

相当于退出该src文件

global variable = 在R1-system-config里面设置好的,所有的机器人程序都能调用

投入运行

1.测量-工具-XYZ4点法（这是我们之前做针尖TCP的时候做的）

2.测量-基座标-3点

3.调整-EMD（就是重新校准KRC，也就是说机械臂撞了，然后轴和轴之间的相对位置改变了，需要用EMD来重新恢复default设置）

4.售后服务-长文本

5.机器人数据 - 机器人的型号和名称

这个类似于FANUC的step，可以一直运行到底，或者运行step by step

TO-DO: 机器人基础编程

2.FAUNC

1.创建程序

Menu

软件工程师培训（机器人基本操作、机器人在线编程基础、KUKA机器人编程四书）\培训资料\发那科机器人编程手册\B-83284CM_04机器人基础编程.PDF

I/O里面的ITEM相当于KUKA的至，相当于跳到哪个变量

DATA-[类型]-位置寄存器(相当于Robim里面的PR，KUKA的E6个axis)

FCTN - (function)

GROUP - 打组，相当于把外部轴和机械臂打组，然后你挪外部轴的J7，外部轴会带着你的机械臂挪动。如果你不GROUP，你挪J7，机械臂就相当于只是基座在动。

J P[1] 是关节逼近

L P[1]是线性逼近

C P[1]是arc移动

后台逻辑≈KUKA的SPS

.ls 变成 .tp程序文件是ASCII转码的

MENU-6系统-2变量-找到MASTER_ENB-把0变成1-就可以改标定零点

3.ABB

2021-04-27

1.FANUC标定针尖TCP

MENU - 6.设置 - 4.坐标系 - F3[编号]找一个空的工具TCP - F2[方法] - 1三点法

然后移动机器人，三个姿态戳点，每个点在接近点那里选择SHIFT+F5记录，弄个三组就自动算好了

2.怎么切换TCP跑FANUC？

按住SHIFT+COORD, 右上角出现Tool, Jog, User，选择Tool然后输入刚刚设好的几号工具就可以了

​

MENU-下页-系统-MasterEnable-0变成1

系统-简易