Marketing/4ventures-client-portal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: notas Seguimiento Semanal con HTML completo, enlaces y formato

Browse Source 
Exporta el Google Doc en HTML, parte por fecha y guarda todo el bloque (listas, negritas, links) sin truncar.

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
4Ventures
2026-09-24 17:20:31 +02:00
parent 751673c341
commit 04ea6de08a
3 changed files with 266 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

271
src/lib/superhuman-docs.ts
View File

@@ -167,7 +167,21 @@ export function parseMetricsFromMarkdown(md: string): ParsedMetrics {
return metrics;
}
/** Split plain-text / markdown seguimiento docs into one section per day heading. */
function parseDayLabel(raw: string): { dateIso: string; label: string } | null {
const m = raw.trim().match(/^(\d{1,2})[\/.\-](\d{1,2})[\/.\-](\d{2,4})$/);
if (!m) return null;
let y = Number(m[3]);
const mo = Number(m[2]);
const d = Number(m[1]);
if (y < 100) y += 2000;
if (y < 2000 || y > 2100 || mo < 1 || mo > 12 || d < 1 || d > 31) return null;
return {
dateIso: `${y}-${String(mo).padStart(2, '0')}-${String(d).padStart(2, '0')}`,
label: `${d}/${mo}/${String(y).slice(2)}`,
};
}
/** Split plain-text seguimiento docs into one section per day heading. */
export function splitNotesByDay(text: string): DayNoteSection[] {
const clean = text.replace(/^\uFEFF/, '').replace(/\r\n/g, '\n');
const dayRe =
@@ -178,17 +192,9 @@ export function splitNotesByDay(text: string): DayNoteSection[] {
for (let i = 0; i < lines.length; i++) {
const m = lines[i].trim().match(dayRe);
if (!m) continue;
let y = Number(m[3]);
const mo = Number(m[2]);
const d = Number(m[1]);
if (y < 100) y += 2000;
if (y < 2000 || y > 2100 || mo < 1 || mo > 12 || d < 1 || d > 31) continue;
const dateIso = `${y}-${String(mo).padStart(2, '0')}-${String(d).padStart(2, '0')}`;
hits.push({
idx: i,
label: `${d}/${mo}/${String(y).slice(2)}`,
dateIso,
});
const parsed = parseDayLabel(`${m[1]}/${m[2]}/${m[3]}`);
if (!parsed) continue;
hits.push({ idx: i, ...parsed });
}
if (!hits.length) return [];
@@ -211,7 +217,151 @@ export function splitNotesByDay(text: string): DayNoteSection[] {
});
}
// Newest → oldest
sections.sort((a, b) => b.dateIso.localeCompare(a.dateIso));
return sections;
}
function unwrapGoogleRedirect(href: string): string {
try {
const u = new URL(href);
if (
u.hostname.includes('google.') &&
(u.pathname.includes('/url') || u.searchParams.has('q'))
) {
const q = u.searchParams.get('q') || u.searchParams.get('url');
if (q) return q;
}
} catch {
/* keep original */
}
return href;
}
/** Clean Google Docs HTML fragment: keep structure, bold, lists, real links. */
export function sanitizeSeguimientoHtml(raw: string): string {
let html = raw
.replace(/<style[\s\S]*?<\/style>/gi, '')
.replace(/<script[\s\S]*?<\/script>/gi, '')
.replace(/<!--[\s\S]*?-->/g, '');
// Bold spans from Docs
html = html.replace(
/<span[^>]*font-weight:\s*700[^>]*>([\s\S]*?)<\/span>/gi,
'<strong>$1</strong>'
);
html = html.replace(
/<span[^>]*font-weight:\s*bold[^>]*>([\s\S]*?)<\/span>/gi,
'<strong>$1</strong>'
);
// Normalize links + unwrap google redirects
html = html.replace(
/<a\b[^>]*href=["']([^"']+)["'][^>]*>([\s\S]*?)<\/a>/gi,
(_m, href: string, inner: string) => {
const cleanHref = unwrapGoogleRedirect(String(href).replace(/&amp;/g, '&'))
.replace(/"/g, '')
.trim();
if (!/^https?:\/\//i.test(cleanHref) && !cleanHref.startsWith('mailto:')) {
return inner;
}
const text = inner.replace(/<[^>]+>/g, '').trim() || cleanHref;
return `<a href="${cleanHref}" target="_blank" rel="noreferrer noopener">${text}</a>`;
}
);
// Drop remaining spans but keep text
html = html.replace(/<\/?span\b[^>]*>/gi, '');
// Strip attributes except href on anchors; keep semantic tags
html = html.replace(
/<\/?(ul|ol|li|p|br|strong|b|em|i|u|h[1-4])\b[^>]*>/gi,
(tag) => {
const name = tag.match(/<\/?([a-z0-9]+)/i)?.[1]?.toLowerCase();
if (!name) return '';
if (tag.startsWith('</')) return `</${name}>`;
if (name === 'br') return '<br/>';
return `<${name}>`;
}
);
// Remove any leftover tags that are not allowlisted
html = html.replace(
/<\/?(?!\/?(?:ul|ol|li|p|br|strong|b|em|i|u|h[1-4]|a)\b)[a-z][^>]*>/gi,
''
);
// Collapse empty paragraphs / excess whitespace
html = html
.replace(/<p>\s*<\/p>/gi, '')
.replace(/(?:<br\s*\/?>\s*){3,}/gi, '<br/><br/>')
.replace(/\n{3,}/g, '\n\n')
.trim();
return html;
}
function htmlToPlainPreview(html: string): string {
return html
.replace(/<br\s*\/?>/gi, '\n')
.replace(/<\/(p|li|h[1-4]|div)>/gi, '\n')
.replace(/<[^>]+>/g, '')
.replace(/&nbsp;/g, ' ')
.replace(/&amp;/g, '&')
.replace(/&lt;/g, '<')
.replace(/&gt;/g, '>')
.replace(/&quot;/g, '"')
.replace(/\n{3,}/g, '\n\n')
.trim();
}
/**
* Split Google Docs HTML export by day headings (bold date-only paragraphs).
* Body of each note = todo el contenido hasta la fecha siguiente (HTML limpio).
*/
export function splitHtmlNotesByDay(htmlDoc: string): DayNoteSection[] {
const bodyMatch = htmlDoc.match(/<body[^>]*>([\s\S]*)<\/body>/i);
const body = bodyMatch ? bodyMatch[1] : htmlDoc;
// Match every <p>…</p>; keep those whose visible text is exactly a date
const pRe = /<p\b[^>]*>[\s\S]*?<\/p>/gi;
const hits: { start: number; end: number; dateIso: string; label: string }[] =
[];
let m: RegExpExecArray | null;
while ((m = pRe.exec(body)) !== null) {
const inner = m[0].replace(/^<p\b[^>]*>/i, '').replace(/<\/p>$/i, '');
const text = inner
.replace(/<br\s*\/?>/gi, ' ')
.replace(/<[^>]+>/g, '')
.replace(/&nbsp;/gi, ' ')
.replace(/\s+/g, ' ')
.trim();
const parsed = parseDayLabel(text);
if (!parsed) continue;
hits.push({
start: m.index,
end: m.index + m[0].length,
dateIso: parsed.dateIso,
label: parsed.label,
});
}
if (!hits.length) return [];
const sections: DayNoteSection[] = [];
for (let i = 0; i < hits.length; i++) {
const contentStart = hits[i].end;
const contentEnd = i + 1 < hits.length ? hits[i + 1].start : body.length;
const fragment = body.slice(contentStart, contentEnd).trim();
if (!fragment) continue;
const cleaned = sanitizeSeguimientoHtml(fragment);
if (!cleaned || htmlToPlainPreview(cleaned).length < 2) continue;
sections.push({
dateIso: hits[i].dateIso,
label: hits[i].label,
body: cleaned,
});
}
sections.sort((a, b) => b.dateIso.localeCompare(a.dateIso));
return sections;
}
@@ -393,7 +543,7 @@ type DriveDocHit = {
webViewLink?: string | null;
};
/** Find Google Doc «Seguimiento …» matching page/tenant name. */
/** Prefer docs literally named «Seguimiento Semanal …». */
export async function findSeguimientoGoogleDoc(
hints: string[]
): Promise<DriveDocHit | null> {
@@ -402,6 +552,7 @@ export async function findSeguimientoGoogleDoc(
.map((h) => h.replace(/^\s*\d+\.\s*/, '').trim())
.filter(Boolean);
// Exact name first (incl. «Seguimiento Semanal X»)
for (const name of cleaned) {
const safe = name.replace(/'/g, "\\'");
const params = new URLSearchParams({
@@ -422,9 +573,9 @@ export async function findSeguimientoGoogleDoc(
}
const params = new URLSearchParams({
q: "name contains 'Seguimiento' and mimeType='application/vnd.google-apps.document' and trashed=false",
q: "name contains 'Seguimiento Semanal' and mimeType='application/vnd.google-apps.document' and trashed=false",
fields: 'files(id,name,webViewLink,modifiedTime)',
pageSize: '40',
pageSize: '50',
corpora: 'allDrives',
includeItemsFromAllDrives: 'true',
supportsAllDrives: 'true',
@@ -433,12 +584,36 @@ export async function findSeguimientoGoogleDoc(
headers: { Authorization: `Bearer ${token}` },
cache: 'no-store',
});
if (!r.ok) return null;
if (!r.ok) {
// fallback broader
const p2 = new URLSearchParams({
q: "name contains 'Seguimiento' and mimeType='application/vnd.google-apps.document' and trashed=false",
fields: 'files(id,name,webViewLink,modifiedTime)',
pageSize: '50',
corpora: 'allDrives',
includeItemsFromAllDrives: 'true',
supportsAllDrives: 'true',
});
const r2 = await fetch(`https://www.googleapis.com/drive/v3/files?${p2}`, {
headers: { Authorization: `Bearer ${token}` },
cache: 'no-store',
});
if (!r2.ok) return null;
const files2 = ((await r2.json()) as { files?: DriveDocHit[] }).files || [];
return scoreDriveHits(files2, cleaned);
}
const files = ((await r.json()) as { files?: DriveDocHit[] }).files || [];
return scoreDriveHits(files, cleaned);
}
function scoreDriveHits(
files: DriveDocHit[],
cleaned: string[]
): DriveDocHit | null {
let best: { file: DriveDocHit; score: number } | null = null;
for (const f of files) {
const fn = normalizeName(f.name || '');
let score = 0;
let score = /seguimiento\s+semanal/.test(fn) ? 20 : 0;
for (const h of cleaned) {
const hn = normalizeName(h);
if (!hn) continue;
@@ -447,7 +622,10 @@ export async function findSeguimientoGoogleDoc(
else {
const tokens = hn
.split(' ')
.filter((t) => t.length > 3 && t !== 'seguimiento' && t !== 'semanal');
.filter(
(t) =>
t.length > 3 && t !== 'seguimiento' && t !== 'semanal'
);
const hits = tokens.filter((t) => fn.includes(t)).length;
if (tokens.length && hits / tokens.length >= 0.6) {
score = Math.max(score, 70 + hits * 5);
@@ -459,6 +637,22 @@ export async function findSeguimientoGoogleDoc(
return best && best.score >= 70 ? best.file : null;
}
export async function exportGoogleDocHtml(fileId: string): Promise<string> {
const token = await getValidAccessToken();
const r = await fetch(
`https://www.googleapis.com/drive/v3/files/${encodeURIComponent(fileId)}/export?mimeType=text%2Fhtml`,
{
headers: { Authorization: `Bearer ${token}` },
cache: 'no-store',
}
);
if (!r.ok) {
throw new Error(`Drive export HTML ${r.status}: ${(await r.text()).slice(0, 200)}`);
}
return await r.text();
}
/** @deprecated prefer exportGoogleDocHtml — plain loses links/format */
export async function exportGoogleDocPlain(fileId: string): Promise<string> {
const token = await getValidAccessToken();
const r = await fetch(
@@ -474,17 +668,33 @@ export async function exportGoogleDocPlain(fileId: string): Promise<string> {
return await r.text();
}
function dayRowsFromText(opts: {
function dayRowsFromHtml(opts: {
tenantId: string;
text: string;
html: string;
fileId: string;
browserLink: string | null;
maxEntries: number;
now: string;
}): SeguimientoSyncRow[] {
const sections = splitNotesByDay(opts.text).slice(0, opts.maxEntries);
let sections = splitHtmlNotesByDay(opts.html);
// Fallback: plain text split if HTML date markers not found
if (!sections.length) {
const plain = htmlToPlainPreview(
opts.html.replace(/<style[\s\S]*?<\/style>/gi, '')
);
sections = splitNotesByDay(plain).map((s) => ({
...s,
body: `<pre>${s.body
.replace(/&/g, '&amp;')
.replace(/</g, '&lt;')
.replace(/>/g, '&gt;')}</pre>`,
}));
}
sections = sections.slice(0, opts.maxEntries);
return sections.map((sec) => {
const metrics = parseMetricsFromMarkdown(sec.body);
const plain = htmlToPlainPreview(sec.body);
const metrics = parseMetricsFromMarkdown(plain);
const summaryBits = [
metrics.periodo ? `Periodo: ${metrics.periodo}` : null,
metrics.ventas != null ? `Ventas: ${metrics.ventas}` : null,
@@ -493,7 +703,7 @@ function dayRowsFromText(opts: {
metrics.cac != null ? `CAC≈${metrics.cac.toFixed(0)}€` : null,
].filter(Boolean);
const firstLine =
sec.body
plain
.split('\n')
.map((l) => l.replace(/^[\s*•\-]+/, '').trim())
.find((l) => l.length > 8) || '';
@@ -507,7 +717,8 @@ function dayRowsFromText(opts: {
summaryBits.length > 0
? summaryBits.join(' · ')
: firstLine.slice(0, 220) || null,
body_md: excerpt(sec.body, 6000),
// Full day block (no truncate): HTML con listas, negritas y enlaces
body_md: sec.body,
browser_link: opts.browserLink,
metrics,
occurred_at: `${sec.dateIso}T12:00:00.000Z`,
@@ -531,7 +742,7 @@ export async function buildSeguimientoRowsForDoc(opts: {
const now = new Date().toISOString();
const rows: SeguimientoSyncRow[] = [];
// 1) Google Doc embebido / emparejado por nombre (fuente real de notas por día)
// 1) Google Doc «Seguimiento Semanal …» — HTML completo partido por fecha
const nameHints = [
...roots.filter((r) => SEGUIMIENTO_RE.test(r.name)).map((r) => r.name),
opts.tenantName ? `Seguimiento Semanal ${opts.tenantName}` : '',
@@ -542,11 +753,11 @@ export async function buildSeguimientoRowsForDoc(opts: {
try {
const hit = await findSeguimientoGoogleDoc(nameHints);
if (hit) {
const text = await exportGoogleDocPlain(hit.id);
const html = await exportGoogleDocHtml(hit.id);
rows.push(
...dayRowsFromText({
...dayRowsFromHtml({
tenantId: opts.tenantId,
text,
html,
fileId: hit.id,
browserLink:
hit.webViewLink ||
@@ -625,7 +836,7 @@ export async function upsertSeguimientoRows(
if (!rows.length) return 0;
const headers = serviceHeaders();
if (!headers) throw new Error('no service role');
const chunk = 40;
const chunk = 15;
for (let i = 0; i < rows.length; i += chunk) {
const part = rows.slice(i, i + chunk);
const r = await fetch(