Documentation: Add note about DNS lookups to browser architecture docs

This commit is contained in:
Andreas Kling 2020-07-06 18:40:12 +02:00
parent ecd4c6718e
commit 1037a24076
Notes: sideshowbarker 2024-07-19 05:06:27 +09:00

View file

@ -22,6 +22,8 @@ This process hosts the main HTML/CSS engine (**LibWeb**.) It also runs JavaScrip
This process can speak networking protocols (like HTTP, HTTPS, and Gemini) to the outside world. Each **WebContent** process gets its own **ProtocolServer** to do networking on its behalf.
For DNS lookups, **ProtocolServer** asks for help from the system's global **LookupServer** service, which handles all outgoing DNS requests.
### Process: ImageDecoder
This process can decode images (PNG, JPEG, BMP, ICO, PBM, etc.) into bitmaps. Each image is decoded in a fresh **ImageDecoder** process. These are strongly sandboxed and can't do much except receive encoded bitmap data and return a bitmap to **WebContent** if decoding successful.
@ -44,4 +46,4 @@ The `WebContentClient` speaks to a `WebContent::ClientConnection` in the **WebCo
Inside **LibWeb**, a `Web::Page` has a main `Web::Frame`, which may have subframes corresponding to `<frame>` or `<iframe>` HTML elements. Each `Web::Frame` has a `Web::Document`, which is the root node of the DOM tree.
Describing the **LibWeb** object model in detail is outside the scope of this document.
Describing the **LibWeb** object model in detail is outside the scope of this document.