mikkel/nexus
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix(issues): decode HTML entities in @mention tokens instead of stripping

Browse source 
Addresses Greptile review on PR #1363: numeric entities decode via
code points; named entities use a small allowlist (amp, nbsp, etc.)
so M&M resolves correctly; unknown named entities are preserved.

Adds mid-token tests for & in agent names.

Made-with: Cursor
This commit is contained in:
amit221 2026-03-24 09:40:55 +02:00
parent 47449152ac
commit 730a67bb20
2 changed files with 35 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
server/src/__tests__/normalize-agent-mention-token.test.ts
View file

@ -14,6 +14,11 @@ describe("normalizeAgentMentionToken", () => {
expect(normalizeAgentMentionToken("Baba ")).toBe("Baba");
});
it("decodes named entities mid-token so agent names can include &", () => {
expect(normalizeAgentMentionToken("Ba&ba")).toBe("Ba&ba");
expect(normalizeAgentMentionToken("M&M")).toBe("M&M");
});
it("returns plain names unchanged", () => {
expect(normalizeAgentMentionToken("Baba")).toBe("Baba");
});

33
server/src/services/issues.ts
View file

@ -196,11 +196,38 @@ function unreadForUserCondition(companyId: string, userId: string) {
`;
}
const HTML_ENTITY_IN_MENTION = /&#x[0-9a-fA-F]+;|&#[0-9]+;|&[a-z]+;/gi;
/** Named entities the rich-text editor may emit in issue bodies; unknown names are left unchanged. */
const WELL_KNOWN_NAMED_HTML_ENTITIES: Readonly<Record<string, string>> = {
amp: "&",
apos: "'",
gt: ">",
lt: "<",
nbsp: "\u00A0",
quot: '"',
ensp: "\u2002",
emsp: "\u2003",
thinsp: "\u2009",
};
/** Strips common HTML entities from a raw @mention capture so UI-encoded bodies still match agent names. */
function decodeNumericHtmlEntity(digits: string, radix: 16 | 10): string | null {
const n = Number.parseInt(digits, radix);
if (Number.isNaN(n) || n < 0 || n > 0x10ffff) return null;
try {
return String.fromCodePoint(n);
} catch {
return null;
}
}
/** Decodes HTML entities in a raw @mention capture so UI-encoded bodies still match agent names. */
export function normalizeAgentMentionToken(raw: string): string {
return raw.replace(HTML_ENTITY_IN_MENTION, "").trim();
let s = raw.replace(/&#x([0-9a-fA-F]+);/gi, (full, hex: string) => decodeNumericHtmlEntity(hex, 16) ?? full);
s = s.replace(/&#([0-9]+);/g, (full, dec: string) => decodeNumericHtmlEntity(dec, 10) ?? full);
s = s.replace(/&([a-z][a-z0-9]*);/gi, (full, name: string) => {
const decoded = WELL_KNOWN_NAMED_HTML_ENTITIES[name.toLowerCase()];
return decoded !== undefined ? decoded : full;
});
return s.trim();
}
export function deriveIssueUserContext(