chore: sync from fn-registry agent

2026-06-10 11:43:45 +02:00
commit 7c475be581
9 changed files with 519 additions and 0 deletions
@@ -0,0 +1,130 @@
+#!/usr/bin/env python3
+"""Migrador de organizaciones (empresas) desde NotasDeObsidian al vault osint.
+
+A diferencia de las personas (que parten de una ficha con wikilinks), las organizaciones se
+detectan recogiendo todas las notas cuyo titulo menciona el nombre de la empresa. La nota cuyo
+titulo ES el nombre de la empresa actua como ficha; el resto son documentos.
+
+Estructura resultante (estandar projects/osint/CONVENTIONS.md):
+  organizaciones/<slug>.md
+  organizaciones/<slug>/<doc-slug>.md
+  attachments/organizaciones/<slug>/<doc-slug>-N.ext
+
+Modo MOVER (borra originales). Idempotente (overwrite por slug).
+"""
+import sys, os, re, shutil
+
+sys.path.insert(0, "/home/enmanuel/fn_registry/python/functions")
+from obsidian import (read_obsidian_note, create_obsidian_note,
+                      slugify_obsidian_name, extract_obsidian_embeds, resolve_obsidian_embed,
+                      list_obsidian_notes)
+
+NOTAS = "/home/enmanuel/Obsidian/NotasDeObsidian"
+OSINT = "/home/enmanuel/Obsidian/osint"
+STOPWORDS = {"de", "del", "la", "las", "el", "los", "y", "a", "en"}
+
+ORGS = [
+    {"display": "FenixFood SL", "slug": "fenixfood-sl",
+     "match": re.compile(r'fenix\s*food', re.I),
+     "nametok": {"fenixfood", "fenix", "food", "sl"}},
+    {"display": "BiorganicFood SL", "slug": "biorganicfood-sl",
+     "match": re.compile(r'biorganic', re.I),
+     "nametok": {"biorganicfood", "biorganic", "food", "sl"}},
+]
+
+
+def doc_slug(title, ntok):
+    parts = [p for p in slugify_obsidian_name(title).split("-")
+             if p and p not in ntok and p not in STOPWORDS]
+    return "-".join(parts) or "documento"
+
+
+def migrate_org(org, all_notes, att_del, docs_del):
+    matched = [n for n in all_notes if org["match"].search(os.path.basename(n)[:-3])]
+    ficha_path = None
+    for n in matched:
+        t = os.path.basename(n)[:-3]
+        if slugify_obsidian_name(t) == org["slug"] or t.lower() == org["display"].lower():
+            ficha_path = n
+            break
+
+    slug, ntok = org["slug"], org["nametok"]
+    att_rel = f"attachments/organizaciones/{slug}"
+    att_abs = f"{OSINT}/{att_rel}"
+    os.makedirs(att_abs, exist_ok=True)
+
+    docs_done, ficha_extra = [], ""
+    for n in matched:
+        t = os.path.basename(n)[:-3]
+        if n == ficha_path:
+            ficha_extra = read_obsidian_note(n)["body"]
+            continue
+        dn = read_obsidian_note(n)
+        ds = doc_slug(t, ntok)
+        # evitar colision de doc-slug dentro de la misma org
+        base_ds = ds
+        k = 2
+        while os.path.exists(f"{OSINT}/organizaciones/{slug}/{ds}.md"):
+            ds = f"{base_ds}-{k}"; k += 1
+        embeds = extract_obsidian_embeds(dn["body"])
+        new = []
+        for i, emb in enumerate(embeds, 1):
+            ap = resolve_obsidian_embed(NOTAS, emb)
+            if not ap:
+                new.append(f"<!-- attachment no encontrado: {emb} -->"); continue
+            ext = os.path.splitext(ap)[1].lower()
+            nn = f"{ds}-{i}{ext}"
+            shutil.copy2(ap, f"{att_abs}/{nn}"); att_del.add(ap)
+            new.append(f"![[{att_rel}/{nn}]]")
+        # conservar texto del doc si lo tiene, ademas de los embeds reescritos
+        text = dn["body"]
+        text_wo_embeds = re.sub(r'!\[\[[^\]]+\]\]', '', text).strip()
+        parts = []
+        if text_wo_embeds:
+            parts.append(text_wo_embeds)
+        if new:
+            parts.append("\n".join(new))
+        create_obsidian_note(OSINT, f"organizaciones/{slug}/{ds}",
+                             body="\n\n".join(parts) if parts else "(sin contenido)",
+                             frontmatter={"tipo": "documento", "entidad": f"[[{slug}]]",
+                                          "fuente": "NotasDeObsidian/" + os.path.relpath(n, NOTAS)},
+                             overwrite=True)
+        docs_done.append((t, ds)); docs_del.add(n)
+
+    bl = [f"Ficha de la organizacion {org['display']}.", ""]
+    if ficha_extra.strip():
+        bl += [ficha_extra.strip(), ""]
+    bl += ["## Documentos", ""] + [f"- [[organizaciones/{slug}/{ds}|{t}]]" for t, ds in docs_done]
+    bl += ["", "## Notas", ""]
+    create_obsidian_note(OSINT, f"organizaciones/{slug}", body="\n".join(bl),
+                         frontmatter={"tipo": "organizacion", "nombre": org["display"], "slug": slug,
+                                      "tags": ["organizacion", "osint"],
+                                      "fuente": "NotasDeObsidian/" + (os.path.relpath(ficha_path, NOTAS) if ficha_path else "")},
+                         overwrite=True)
+    if ficha_path:
+        docs_del.add(ficha_path)
+    return slug, len(docs_done), ficha_path is not None
+
+
+def main():
+    all_notes = [n for n in list_obsidian_notes(NOTAS) if "/.git/" not in n and "/dist/" not in n]
+    att_del, docs_del = set(), set()
+    results = []
+    for org in ORGS:
+        results.append((org["display"],) + migrate_org(org, all_notes, att_del, docs_del))
+    for p in docs_del:
+        try: os.remove(p)
+        except FileNotFoundError: pass
+    moved_att = 0
+    for ap in att_del:
+        try:
+            os.remove(ap); moved_att += 1
+        except FileNotFoundError:
+            pass
+    print(f"docs .md movidos: {len(docs_del)} | attachments movidos: {moved_att}\n")
+    for disp, slug, ndocs, had_ficha in results:
+        print(f"  {disp} -> organizaciones/{slug} | docs={ndocs} | ficha_origen={'si' if had_ficha else 'no'}")
+
+
+if __name__ == "__main__":
+    main()