mirror of
https://github.com/bytedance/deer-flow.git
synced 2026-11-20 11:20:35 +00:00
* feat(frontend): improve file attachment recovery and artifact presentation * feat(extensions): add Office artifact previews with search and navigation * fix: preserve public viewers and HTTP Office previews * fix: address file preview and upload review findings * fix: use typed upload errors and linear bookmark allocation * fix(office-preview): isolate parsing and bound preview resources * fix(ci): install Office preview fixture dependencies for E2E * fix(ci): provision matching Playwright browsers before E2E tests * fix(ci): isolate Python libraries and include Office documentation * test(frontend): wait for download status to finish rendering
141 lines
5.0 KiB
JavaScript
141 lines
5.0 KiB
JavaScript
import { createUniqueIdAllocator } from "./identifiers.js";
|
|
|
|
/** Literal matches across styled runs, with no HTML insertion. */
|
|
export const MAX_SEARCH_MATCHES = 2000;
|
|
|
|
export function highlightText(root, query, limit = MAX_SEARCH_MATCHES) {
|
|
root
|
|
.querySelectorAll("mark[data-office-match]")
|
|
.forEach((mark) => mark.replaceWith(...mark.childNodes));
|
|
root.normalize();
|
|
if (!query.trim() || limit <= 0) return [];
|
|
const groups = [];
|
|
const blocks = [
|
|
...root.querySelectorAll("p,td,th,li,h1,h2,h3,h4,h5,h6"),
|
|
].filter((node) => !node.querySelector("p,td,th,li,h1,h2,h3,h4,h5,h6"));
|
|
if (!blocks.length) blocks.push(root);
|
|
const pattern = query.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
|
|
for (const block of blocks) {
|
|
if (groups.length >= limit) break;
|
|
const walker = document.createTreeWalker(block, NodeFilter.SHOW_TEXT, {
|
|
acceptNode: (node) =>
|
|
node.parentElement.closest("style,script")
|
|
? NodeFilter.FILTER_REJECT
|
|
: NodeFilter.FILTER_ACCEPT,
|
|
});
|
|
const nodes = [];
|
|
let text = "",
|
|
node;
|
|
while ((node = walker.nextNode())) {
|
|
nodes.push({ node, start: text.length });
|
|
text += node.textContent;
|
|
}
|
|
const matches = [];
|
|
for (const match of text.matchAll(new RegExp(pattern, "giu"))) {
|
|
matches.push(match);
|
|
if (matches.length >= limit - groups.length) break;
|
|
}
|
|
const local = matches.map(() => []);
|
|
for (const { node: textNode, start } of nodes) {
|
|
const content = textNode.textContent,
|
|
fragment = document.createDocumentFragment();
|
|
let cursor = 0;
|
|
matches.forEach((match, index) => {
|
|
const from = Math.max(0, match.index - start),
|
|
to = Math.min(content.length, match.index + match[0].length - start);
|
|
if (from >= to) return;
|
|
fragment.append(document.createTextNode(content.slice(cursor, from)));
|
|
const mark = document.createElement("mark");
|
|
mark.dataset.officeMatch = "";
|
|
mark.textContent = content.slice(from, to);
|
|
fragment.append(mark);
|
|
local[index].push(mark);
|
|
cursor = to;
|
|
});
|
|
if (cursor) {
|
|
fragment.append(document.createTextNode(content.slice(cursor)));
|
|
textNode.replaceWith(fragment);
|
|
}
|
|
}
|
|
groups.push(...local.filter((group) => group.length));
|
|
}
|
|
return groups;
|
|
}
|
|
|
|
/** Add private bookmarks only to the preview copy, so outline targets survive page layout. */
|
|
export async function prepareWordOutline(zip) {
|
|
const parse = (xml) =>
|
|
new DOMParser().parseFromString(xml, "application/xml");
|
|
const attr = (node, name) =>
|
|
node?.getAttributeNS(
|
|
"http://schemas.openxmlformats.org/wordprocessingml/2006/main",
|
|
name,
|
|
);
|
|
const child = (node, name) =>
|
|
[...(node?.children ?? [])].find((entry) => entry.localName === name);
|
|
const styles = new Map();
|
|
const source = zip.file("word/styles.xml");
|
|
if (source)
|
|
for (const style of parse(
|
|
await source.async("string"),
|
|
).getElementsByTagNameNS("*", "style")) {
|
|
styles.set(attr(style, "styleId"), {
|
|
level: attr(child(child(style, "pPr"), "outlineLvl"), "val"),
|
|
base: attr(child(style, "basedOn"), "val"),
|
|
});
|
|
}
|
|
const resolveLevel = (id, visited = new Set()) => {
|
|
if (!id || visited.has(id)) return undefined;
|
|
visited.add(id);
|
|
const style = styles.get(id);
|
|
return style?.level ?? resolveLevel(style?.base, visited);
|
|
};
|
|
const file = zip.file("word/document.xml");
|
|
if (!file) return [];
|
|
const xml = parse(await file.async("string")),
|
|
outline = [];
|
|
const bookmarks = [...xml.getElementsByTagNameNS("*", "bookmarkStart")];
|
|
const nextName = createUniqueIdAllocator(
|
|
bookmarks.map((bookmark) => attr(bookmark, "name") ?? ""),
|
|
"office-",
|
|
);
|
|
const nextId = createUniqueIdAllocator(
|
|
[...bookmarks, ...xml.getElementsByTagNameNS("*", "bookmarkEnd")].map(
|
|
(bookmark) => attr(bookmark, "id") ?? "",
|
|
),
|
|
"",
|
|
100000,
|
|
);
|
|
for (const paragraph of xml.getElementsByTagNameNS("*", "p")) {
|
|
const props = child(paragraph, "pPr");
|
|
const level = Number(
|
|
attr(child(props, "outlineLvl"), "val") ??
|
|
resolveLevel(attr(child(props, "pStyle"), "val")) ??
|
|
9,
|
|
);
|
|
const title = [...paragraph.getElementsByTagNameNS("*", "t")]
|
|
.map((node) => node.textContent)
|
|
.join("")
|
|
.trim();
|
|
if (level < 0 || level > 8 || !title) continue;
|
|
const id = nextName();
|
|
const bookmarkId = nextId();
|
|
const bookmark = xml.createElementNS(
|
|
paragraph.namespaceURI,
|
|
"w:bookmarkStart",
|
|
);
|
|
bookmark.setAttributeNS(paragraph.namespaceURI, "w:name", id);
|
|
bookmark.setAttributeNS(paragraph.namespaceURI, "w:id", bookmarkId);
|
|
paragraph.insertBefore(
|
|
bookmark,
|
|
props?.nextSibling ?? paragraph.firstChild,
|
|
);
|
|
const end = xml.createElementNS(paragraph.namespaceURI, "w:bookmarkEnd");
|
|
end.setAttributeNS(paragraph.namespaceURI, "w:id", bookmarkId);
|
|
paragraph.appendChild(end);
|
|
outline.push({ id, title, level });
|
|
}
|
|
zip.file("word/document.xml", new XMLSerializer().serializeToString(xml));
|
|
return outline;
|
|
}
|