I went to godbolt in the middle of the night and found it extremely bright and started wondering if it could respect the system setting.

There is a CSS media feature prefers-color-scheme which indicates whether the user prefers light or dark mode. It might be nice to make the default dark mode if the user hasn't explicitly picked one yet be "System" and using the prefers-color-scheme settin

In page https://gnunn1.github.io/tilix-web/manual/vteconfig/,
at paragraph "Source vte.sh in bashrc", the following example is given for when user already having a PROMPT_COMMAND:

function custom_prompt() {
  __git_ps1 "\[\033[0;31m\]\u \[\033[0;36m\]\h:\w\[\033[00m\]" " \n\[\033[0;31m\]>\[\033[00m\] " " %s"
  VTE_PWD_THING="$(__vte_osc7)"
  PS1="$PS1$VTE_PWD_THING"
}
PROMPT_COMMA

• init32.S has some unclear error message that could be written better.
• Add generation of the full screen error text, instead of hard coding it for each error.
• Maybe reference PowerD instead of PowerNex in the errors.

I noticed that there's a variety of differences in loop iterations across languages. Kotlin has 1..1000000 which means going 1_000_000 times, from 1 through 1_000_000 inclusive. Some of the other implementations seem to be off from that by 1 in some direction or other, e.g., the Java version actually executes from 1 to 999_999. Obviously missing one iteration doesn't make any difference to any b

For some time I haven't been looking into project structure. Finally I did:

The .idea/libraries directory contained whopping 14,257 items!!! Over time, DUB's dependencies are added over and over to the libraries even if they already exist. The old one are not d

Is your feature request related to a problem? Please describe.
Yes. GUI is for children and CLI doesn't give me the launching capabilities.

Describe the solution you'd like
A non-gui/non-cli frontend

Describe alternatives you've considered
Ncurses

Additional context
None

The examples don't run on Travis right now because they fail - the older version of libclang there, even when installing the most recent one from apt, doesn't work the same way as libclang 6.0. The CI script should manually download the right libclang version, just like dstep does.

I tried to find responseHeaders on your documentation: http://htmlpreview.github.io/?https://github.com/ikod/dlang-requests/blob/master/docs/http.html

but couldn't find them. A quick look at the code shows that you do have them - is the documentation out of date or is this a problem with the documentation generator?

To prevent it accepting too many types.

It should only accept types that have a handle that decays to a JsHandle.

According to the SFML 2.4 documentation, the TcpSocket class should have a send method for raw data that returns the number of bytes sent as an out parameter.

In non-blocking mode, it may not be the case that a TcpSocket sent everything, which a developer would need to know.