apunkt/SurfSense
1
0
Fork 0
mirror of https://github.com/MODSetter/SurfSense.git synced 2026-05-23 19:05:16 +02:00
Code Issues Projects Releases Packages Wiki Activity

refactor: replace button elements with Button component for improved consistency and styling across multiple UI components

Browse source
This commit is contained in:
Anish Sarkar 2026-05-14 14:17:44 +05:30
parent 23e05acc7c
commit 3d42712b3f
27 changed files with 401 additions and 424 deletions
Download patch file Download diff file Expand all files Collapse all files

55
surfsense_web/components/editor/plugins/citation-kit.tsx
View file

@ -11,20 +11,16 @@ import {
} from "@/lib/citations/citation-parser";
/**
* Plate inline-void node modeling a single `[citation:...]` reference.
*
* Modeled after the existing `MentionPlugin` pattern in
* `inline-mention-editor.tsx` — the only confirmed pattern in this repo
* for non-text inline UI. Inline-void elements satisfy Slate's invariant
* that the editor renders both atomic widgets and surrounding text
* cleanly without breaking selection / caret semantics.
* Plate inline-void node for one `[citation:...]` reference.
* Inline voids keep the citation chip atomic while preserving caret behavior
* around the surrounding text.
*/
export type CitationElementNode = {
type: "citation";
kind: "chunk" | "doc" | "url";
chunkId?: number;
url?: string;
/** Original `[citation:...]` substring for traceability/debugging. */
/** Original literal token that produced this citation node. */
rawText: string;
children: [{ text: "" }];
};
@ -62,17 +58,14 @@ const CitationPlugin = createPlatePlugin({
},
});
/** Plugin kit shape used elsewhere in the editor. */
export const CitationKit = [CitationPlugin];
// ---------------------------------------------------------------------------
// Slate value transform — runs after MarkdownPlugin.deserialize
// Slate value transform
// ---------------------------------------------------------------------------
// Structural shapes used by the value transform. We cannot use Plate's
// generic Element / Text type predicates directly because `Descendant` is a
// constrained union and our predicates would over-narrow. Casting through
// these row types keeps the walker readable without fighting the types.
// Local structural shapes keep the recursive walker readable without forcing
// Plate's broad Descendant union into narrower generic predicates.
type SlateText = { text: string } & Record<string, unknown>;
type SlateElement = { type?: string; children: Descendant[] } & Record<string, unknown>;
@ -89,19 +82,15 @@ function asElement(node: Descendant): SlateElement {
}
/**
* Element types whose subtrees we MUST NOT inject citation void elements
* into. Each rationale documented in the citation plan:
* - `KEYS.codeBlock` / `code_line` — Plate's schema rejects inline elements
* inside code containers; the user expects literal text inside code.
* - `KEYS.link` — `<button>` inside `<a>` is invalid HTML and the link
* swallows the citation click. Mirrors the `<a>` skip in
* `MarkdownViewer`.
* Subtrees that should keep citation tokens as text:
* - Code nodes preserve source text and reject inline void children.
* - Link nodes already render as anchors; citation chips are interactive
* shadcn Button-based controls, so injecting them would nest interactions.
*/
const SKIP_SUBTREE_TYPES = new Set<string>([KEYS.codeBlock, "code_line", KEYS.link]);
/**
* Build the marks portion of a Slate text node so we can preserve formatting
* (bold/italic/etc.) on the surrounding text fragments after we split.
* Preserve text marks such as bold and italic when splitting around citations.
*/
function copyMarks(textNode: SlateText): Record<string, unknown> {
const { text: _text, ...marks } = textNode;
@ -131,9 +120,7 @@ function makeCitationElement(
}
/**
* Re-extract the raw `[citation:...]` substrings that produced each parsed
* segment, in source order. Lets us preserve the original literal for
* `rawText` on the inline-void element.
* Keep each original citation token on the generated node for diagnostics.
*/
function extractRawCitationMatches(text: string): string[] {
const matches: string[] = [];
@ -159,9 +146,7 @@ function transformTextNode(node: SlateText, urlMap: CitationUrlMap): Descendant[
let pendingText: string | null = null;
const flushText = () => {
// Slate inline-void adjacency: emit an empty text node (with copied
// marks) when the citation appears at the very start/end of the text
// node so neighbours of the void always have a text sibling.
// Inline voids need text siblings, even at text boundaries.
out.push({ ...marks, text: pendingText ?? "" } as unknown as Descendant);
pendingText = null;
};
@ -174,8 +159,7 @@ function transformTextNode(node: SlateText, urlMap: CitationUrlMap): Descendant[
const raw = rawMatches[citationIdx] ?? "";
out.push(makeCitationElement(raw, segment) as unknown as Descendant);
citationIdx += 1;
// Always reset pendingText so the next loop iteration emits a
// trailing empty text node if no further plain text follows.
// Ensure a trailing text sibling if the citation ends the node.
pendingText = "";
}
}
@ -206,12 +190,9 @@ function transformChildren(children: Descendant[], urlMap: CitationUrlMap): Desc
}
/**
* Walk a deserialized Slate value and replace every `[citation:...]`
* substring with a `citation` inline-void element. URL placeholders
* created by `preprocessCitationMarkdown` are resolved through `urlMap`.
*
* Subtrees of `code_block`, `code_line`, and `link` are returned as-is —
* see `SKIP_SUBTREE_TYPES` above.
* Replace citation tokens in a deserialized Slate tree with citation inline
* void nodes. URL placeholders from `preprocessCitationMarkdown` are resolved
* through `urlMap`; skipped subtrees are returned unchanged.
*/
export function injectCitationNodes(value: Descendant[], urlMap: CitationUrlMap): Descendant[] {
return transformChildren(value, urlMap);