Sample .devcontainer/devcontainer.json:

json { "name": "Global", "build": { "context": "..", "dockerfile": "Dockerfile" }, "containerEnv": { "PYTHONPATH": "." }, "customizations": { "vscode": { "settings": { "extensions.verifySignature": false }, "extensions": [ "GitHub.copilot", "ms-python.vscode-pylance", "ms-python.python", "eamodio.gitlens" ] } }, "initializeCommand": "/bin/bash -c '[[ -d ${HOME}/.aws ]] || { echo \"Error: ${HOME}/.aws directory not found.\"; exit 1; }; [[ -f ${HOME}/.netrc ]] || { echo \"Error: ${HOME}/.netrc file not found.\"; exit 1; }; [[ -d ${HOME}/.ssh ]] || { echo \"Error: ${HOME}/.ssh directory not found.\"; exit 1; }; echo \"\n> All required mounts found on the host machine.\"'", "onCreateCommand": { "hadolint": "apt-get update && apt-get install wget -y && wget -O /bin/hadolint https://github.com/hadolint/hadolint/releases/download/v2.12.0/hadolint-Linux-x86_64 && chmod u+x /usr/bin/hadolint", "precommit": "pip install pre-commit" }, "updateContentCommand": "/bin/bash -c 'if grep -A 2 \"machine gitlab.com\" ~/.netrc | grep -q \"password\" && GITLAB_TOKEN=$(grep -A 2 \"machine gitlab.com\" ~/.netrc | grep -oP \"(?<=password ).*\" | tr -d \"\\n\") && [ -n \"$GITLAB_TOKEN\" ]; then echo \"\n> Token found in ~/.netrc\"; else read -sp \"\n> Enter your GitLab token: \" GITLAB_TOKEN && echo; fi; echo \"export GITLAB_TOKEN=$GITLAB_TOKEN\" >> ~/.bashrc && . ~/.bashrc && poetry config http-basic.abc __token__ $GITLAB_TOKEN'", "postCreateCommand": ". ~/.bashrc && curl -s --location 'https://gitlab.com/api/v4/projects/12345/repository/files/.pre-commit-config.yaml/raw?ref=main' --header \"PRIVATE-TOKEN: $GITLAB_TOKEN\" -o .pre-commit-config.yaml", "postAttachCommand": "/bin/bash -c '. ~/.bashrc && read -p \"\n> Do you want to update the content of devcontainer.json? (y/n): \" response; if [[ \"$response\" == \"y\" ]]; then curl -s --location \"https://gitlab.com/api/v4/projects/12345/repository/files/devcontainer.json/raw?ref=main\" --header \"PRIVATE-TOKEN: $GITLAB_TOKEN\" -o .devcontainer/devcontainer.json; else echo \"\n> Skipping update of devcontainer.json\"; fi'", "mounts": [ "source=${localEnv:HOME}/.aws/,target=/root/.aws/,type=bind,readonly", "source=${localEnv:HOME}/.netrc,target=/root/.netrc,type=bind,readonly", "source=${localEnv:HOME}/.ssh/,target=/root/.ssh/,type=bind,readonly" ] }

Some more of my recent learning with devcontainer.json (its Dev Container metadata):

• Interactive commands (those waiting for user input like read) do not display the input request in (at least onCreateCommand and postCreateCommand sections), so it is better to keep them in updateContentCommand or postAttachCommand.
• If there are 2 read commands in a single section, like updateContentCommand, only the 1st one is displayed to the user, and the 2nd one is ignored.
• When I put a read command within a dictionary (with at lest 2 key/values) of postAttachCommand, the interactive command wasn't being displayed.
• We need to use /bin/bash -c to be able to use read -s (the -s flag) which allows for securely passing the password so that it does not stay in the VS Code console. Also, I had trouble with interactive commands and if statements without it.
• Using "GITLAB_TOKEN": "${localEnv:GITLAB_TOKEN}" does not easily work as it is looking for GITLAB_TOKEN env variable set locally on our host computers, and I believe no one does it.
• The dictionary seems to be executing its scripts in parallel; therefore, it is not easily possible to break down long lines which have to execute in a chronological sequence.
• JSON does not allow for human-readable line breaks; therefore, indeed, it seems impossible to improve the long one-liners.
• The files/folders mentioned within mounts need to exist locally (otherwise, Docker container build fails). They are mounted before any other section. Technically, we can protect ourselves with the following command to find an extra message in VS Code container logs:

json "initializeCommand": "/bin/bash -c '[[ -d ${HOME}/.aws ]] || { echo \"Error: ${HOME}/.aws directory not found.\"; exit 1; }; [[ -f ${HOME}/.netrc ]] || { echo \"Error: ${HOME}/.netrc file not found.\"; exit 1; }; [[ -d ${HOME}/.ssh ]] || { echo \"Error: ${HOME}/.ssh directory not found.\"; exit 1; }'",

Other option is to get rid of the error completely, but this creates files on the host machine; therefore, it is not an ideal solution:

json "initializeCommand": "mkdir -p ~/.ssh ~/.aws && touch ~/.netrc",

I would also simply put the following:

bash /bin/bash -c 'read -p 'Type a message: ' -t 10 && echo Attach $REPLY'

I also found that updateContentCommand allows for the user interaction (it displays interactive command in the VS Code console).

Explanation of 6 available devcontainer.json (Dev Container in VS Code) hooks.

This is in the contributes part of the package.json

json "contributes": { "configuration": { "type": "object", "title": "Example configuration", "properties": { "languageServerExample.maxNumberOfProblems": { "scope": "resource", "type": "number", "default": 100, "description": "Controls the maximum number of problems produced by the server." }, "languageServerExample.trace.server": { "scope": "window", "type": "string", "enum": [ "off", "messages", "verbose" ], "default": "off", "description": "Traces the communication between VS Code and the language server." } } } },

输入 code-runner

It seems like VS Code offers a better way to manage Docker containers

定位文档路由和位置

Simple difference between two core debugging modes: Launch and Attach available in VS Code.

Depending on the request (attach or launch), different attributes are required, and VS Code's launch.json validation and suggestions should help with that.

Logpoints - log messages to the console when breakpoint is hit.

Can include expressions to be evaluated with {}, e.g.:

fib({num}): {result}

(animation)

Optional arguments for launch.json:

• presentation ("order", "group" or "hidden")
• preLaunchTask
• postDebugTask
• internalConsoleOptions
• debugServer
• serverReadyAction

In the launch.json file you've to define at least those 3 variables:

• type (e.g. "node", "php", "go")
• request ("launch" or "attach")
• name (name to appear in the Debug launch configuration drop-down)

Some of the possibly supported attributes in launch.json:

• program
• args
• env
• cwd
• port
• stopOnEntry
• console (e.g. "internalConsole", "integratedTerminal", "externalTerminal")

To easily see all the refactoring options, use the "Refactor" command

Customising settings for debugging tests while running

Python: Debug All Tests

or

Python: Debug Test Method

Debugging tests themselves

1. Set a breakpoint on the first line of the failing function (e.g. test_decrement)
2. Click the "Debug Test" option above the function
3. Open Debug Console and type: inc_dec.decrement(3) to see what is the actual output when we use x=3
4. Stop the debugger and correct the tests
5. Save the test file and run the tests again to look for a positive result

pytest displays failed tests also in PROBLEMS

Another way to view the test results:

View > Output > Python Test Log

Test Explorer is activated after discovering tests in Python:

After using python.testing.autoTestDiscoverOnSaveEnabled, it'll be set to true and discovering tests whenever a test file is saved.

If discovery succeeds, the status bar shows Run Tests instead:

Start testing in VS Code by using Python: Configure Tests (it automatically chooses one testing framework and disables the rest).

Otherwise, you can configure tests manually by setting only one of the following to True:

• python.testing.unittestEnabled
• python.testing.pytestEnabled
• python.testing.nosetestsEnabled

To run cells inside a Python script in VSCode, all you need to is to define Jupyter-like code cells within Python code using a # %% comment:

# %%
msg = "Hello World"
print(msg)

# %%
msg = "Hello again"
print(msg)

If you wish to use Java inside VSCode, try downloading the Installer of Visual Studio Code for Java developers

After creating a new project related environment, it shall be specified as a default for this specific project

Try to use logpoints instead of print statements.

More info: https://code.visualstudio.com/docs/editor/debugging#_logpoints

VS Code consumes much more memory in comparison to Sublime while working on more significant projects