2026-11-18 15:06:43 +01:00
|
|
|
"""Le transcript lisible d'une conversation pompée (transcript.md) : du texte seul, jamais le HTML d'un mail ; les images
|
|
|
|
|
en ligne, les autres pièces jointes en lien, vers attachments/. Les messages sont déjà sous leur forme locale (pomper.py)."""
|
|
|
|
|
import re
|
|
|
|
|
import urllib.parse
|
|
|
|
|
from datetime import datetime, timezone
|
|
|
|
|
from html.parser import HTMLParser
|
|
|
|
|
|
|
|
|
|
BLOCS = {"p", "div", "li", "tr", "blockquote", "h1", "h2", "h3", "h4", "h5", "h6"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
class _Texte(HTMLParser):
|
|
|
|
|
def __init__(self):
|
|
|
|
|
super().__init__(convert_charrefs=True)
|
|
|
|
|
self.morceaux, self._cache = [], 0
|
|
|
|
|
|
|
|
|
|
def handle_starttag(self, tag, attrs):
|
|
|
|
|
if tag in ("script", "style"):
|
|
|
|
|
self._cache += 1
|
|
|
|
|
elif tag == "br":
|
|
|
|
|
self.morceaux.append("\n")
|
|
|
|
|
elif tag in BLOCS:
|
|
|
|
|
self.morceaux.append("\n\n")
|
|
|
|
|
|
|
|
|
|
def handle_endtag(self, tag):
|
|
|
|
|
if tag in ("script", "style"):
|
|
|
|
|
self._cache = max(0, self._cache - 1)
|
|
|
|
|
elif tag in BLOCS:
|
|
|
|
|
self.morceaux.append("\n\n")
|
|
|
|
|
|
|
|
|
|
def handle_data(self, data):
|
|
|
|
|
if not self._cache:
|
|
|
|
|
self.morceaux.append(data)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def texte(contenu_html: str) -> str:
|
|
|
|
|
"""Le HTML d'un message en texte : balises retirées, entités décodées, paragraphes gardés."""
|
|
|
|
|
p = _Texte()
|
|
|
|
|
p.feed(contenu_html or "")
|
|
|
|
|
p.close()
|
|
|
|
|
t = re.sub(r"[ \t]*\n[ \t]*", "\n", "".join(p.morceaux))
|
|
|
|
|
return re.sub(r"\n{3,}", "\n\n", t).strip()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _heure(t: str) -> str:
|
|
|
|
|
return datetime.fromisoformat(t).astimezone(timezone.utc).strftime("%Y-%m-%d %H:%M") + " UTC"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _nom(personne: dict | None) -> str:
|
|
|
|
|
return " ".join(x for x in ((personne or {}).get("first_name"), (personne or {}).get("last_name")) if x)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def _pieces(pieces: list[dict]) -> list[str]:
|
|
|
|
|
lignes = []
|
|
|
|
|
for a in pieces:
|
2026-11-18 15:23:21 +01:00
|
|
|
if not a.get("url"): # pas téléchargée : perdue par LibreDesk
|
|
|
|
|
lignes.append(f"{a.get('name')} (absente)")
|
|
|
|
|
continue
|
2026-11-18 15:06:43 +01:00
|
|
|
cible = urllib.parse.quote(a.get("url") or "")
|
|
|
|
|
image = (a.get("content_type") or "").startswith("image/") and a.get("content_type") != "image/svg+xml"
|
|
|
|
|
lignes.append(f"{'!' * image}[{a.get('name')}]({cible})")
|
|
|
|
|
return lignes
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def transcript(conversation: dict, messages: list[dict]) -> str:
|
|
|
|
|
c = conversation.get("contact") or {}
|
|
|
|
|
contact = " ".join(x for x in (_nom(c), c.get("email")) if x)
|
|
|
|
|
morceaux = [f"# {conversation.get('subject') or '(sans sujet)'} (#{conversation.get('reference_number')})",
|
|
|
|
|
f"Contact : {contact} · Boîte : {conversation.get('inbox_name')} · Statut : {conversation.get('status')}"
|
|
|
|
|
f" · Créée : {_heure(conversation['created_at'])}"]
|
|
|
|
|
for m in messages:
|
|
|
|
|
corps = (m.get("text_content") or "").strip() or texte(m.get("content"))
|
|
|
|
|
if m.get("type") == "activity":
|
|
|
|
|
morceaux.append(f"*{_heure(m['created_at'])} — {corps}*")
|
|
|
|
|
continue
|
|
|
|
|
client = m.get("sender_type") == "contact"
|
|
|
|
|
role = "note privée" if m.get("private") else "client" if client else "agent"
|
|
|
|
|
auteur = _nom(m.get("author")) or ("Client" if client else "Agent")
|
|
|
|
|
morceaux.append(f"## {_heure(m['created_at'])} — {auteur} ({role})")
|
|
|
|
|
if corps:
|
|
|
|
|
morceaux.append(corps)
|
|
|
|
|
if lignes := _pieces(m.get("attachments") or []):
|
|
|
|
|
morceaux.append("\n".join(lignes))
|
|
|
|
|
return "\n\n".join(morceaux) + "\n"
|