commcoach: remove silent fallbacks in PDF export, fix markdown-to-XML conversion with proper regex

Made-with: Cursor
This commit is contained in:
patrick-motsch 2026-03-07 02:38:42 +01:00
parent cdb974d658
commit 1e678a8897
2 changed files with 66 additions and 63 deletions

View file

@ -937,10 +937,8 @@ async def exportDossier(
if format == "pdf":
pdfBytes = await renderDossierPdf(ctx, sessions, tasks, scores)
if pdfBytes:
return Response(content=pdfBytes, media_type="application/pdf",
headers={"Content-Disposition": f'attachment; filename="dossier_{contextId[:8]}.pdf"'})
format = "md"
return Response(content=pdfBytes, media_type="application/pdf",
headers={"Content-Disposition": f'attachment; filename="dossier_{contextId[:8]}.pdf"'})
md = buildDossierMarkdown(ctx, sessions, tasks, scores)
return Response(content=md, media_type="text/markdown",
@ -976,10 +974,8 @@ async def exportSession(
if format == "pdf":
pdfBytes = await renderSessionPdf(session, messages, tasks, scores)
if pdfBytes:
return Response(content=pdfBytes, media_type="application/pdf",
headers={"Content-Disposition": f'attachment; filename="session_{sessionId[:8]}.pdf"'})
format = "md"
return Response(content=pdfBytes, media_type="application/pdf",
headers={"Content-Disposition": f'attachment; filename="session_{sessionId[:8]}.pdf"'})
md = buildSessionMarkdown(session, messages, tasks, scores)
return Response(content=md, media_type="text/markdown",

View file

@ -143,7 +143,7 @@ def buildSessionMarkdown(session: Dict[str, Any], messages: List[Dict[str, Any]]
async def renderDossierPdf(context: Dict[str, Any], sessions: List[Dict[str, Any]],
tasks: List[Dict[str, Any]], scores: List[Dict[str, Any]],
aiService=None) -> Optional[bytes]:
aiService=None) -> bytes:
"""Render a dossier as PDF directly from markdown content via reportlab."""
md = buildDossierMarkdown(context, sessions, tasks, scores)
title = context.get("title", "Coaching Dossier")
@ -152,71 +152,78 @@ async def renderDossierPdf(context: Dict[str, Any], sessions: List[Dict[str, Any
async def renderSessionPdf(session: Dict[str, Any], messages: List[Dict[str, Any]],
tasks: List[Dict[str, Any]], scores: List[Dict[str, Any]],
aiService=None) -> Optional[bytes]:
aiService=None) -> bytes:
"""Render a session as PDF directly from markdown content via reportlab."""
md = buildSessionMarkdown(session, messages, tasks, scores)
title = f"Coaching Session — {_formatDate(session.get('startedAt'))}"
return _markdownToPdf(md, title)
def _markdownToPdf(markdownText: str, title: str) -> Optional[bytes]:
"""Convert markdown text to a styled PDF using reportlab."""
try:
from reportlab.lib.pagesizes import A4
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.enums import TA_LEFT
from reportlab.lib import colors
except ImportError:
logger.warning("reportlab not available for PDF export")
return None
def _markdownToPdf(markdownText: str, title: str) -> bytes:
"""Convert markdown text to a styled PDF using reportlab. Raises on failure."""
import re as _re
import io
from reportlab.lib.pagesizes import A4
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.enums import TA_LEFT
from reportlab.lib import colors
try:
import io
buf = io.BytesIO()
doc = SimpleDocTemplate(buf, pagesize=A4, rightMargin=60, leftMargin=60, topMargin=50, bottomMargin=40)
buf = io.BytesIO()
doc = SimpleDocTemplate(buf, pagesize=A4, rightMargin=60, leftMargin=60, topMargin=50, bottomMargin=40)
baseStyles = getSampleStyleSheet()
sTitle = ParagraphStyle("DocTitle", parent=baseStyles["Title"], fontSize=18, spaceAfter=20, textColor=colors.HexColor("#1e40af"))
sH2 = ParagraphStyle("DocH2", parent=baseStyles["Heading2"], fontSize=13, spaceBefore=16, spaceAfter=6, textColor=colors.HexColor("#1f2937"))
sH3 = ParagraphStyle("DocH3", parent=baseStyles["Heading3"], fontSize=11, spaceBefore=12, spaceAfter=4, textColor=colors.HexColor("#374151"))
sBody = ParagraphStyle("DocBody", parent=baseStyles["Normal"], fontSize=10, leading=14, spaceAfter=4, alignment=TA_LEFT)
sBullet = ParagraphStyle("DocBullet", parent=sBody, leftIndent=18, bulletIndent=6, spaceBefore=2, spaceAfter=2)
sMeta = ParagraphStyle("DocMeta", parent=sBody, fontSize=8, textColor=colors.grey)
baseStyles = getSampleStyleSheet()
sTitle = ParagraphStyle("DocTitle", parent=baseStyles["Title"], fontSize=18, spaceAfter=20, textColor=colors.HexColor("#1e40af"))
sH2 = ParagraphStyle("DocH2", parent=baseStyles["Heading2"], fontSize=13, spaceBefore=16, spaceAfter=6, textColor=colors.HexColor("#1f2937"))
sH3 = ParagraphStyle("DocH3", parent=baseStyles["Heading3"], fontSize=11, spaceBefore=12, spaceAfter=4, textColor=colors.HexColor("#374151"))
sBody = ParagraphStyle("DocBody", parent=baseStyles["Normal"], fontSize=10, leading=14, spaceAfter=4, alignment=TA_LEFT)
sBullet = ParagraphStyle("DocBullet", parent=sBody, leftIndent=18, bulletIndent=6, spaceBefore=2, spaceAfter=2)
sMeta = ParagraphStyle("DocMeta", parent=sBody, fontSize=8, textColor=colors.grey)
story = [Paragraph(title, sTitle), Spacer(1, 10)]
story = [Paragraph(_escXml(title), sTitle), Spacer(1, 10)]
for line in markdownText.split("\n"):
stripped = line.strip()
if not stripped:
story.append(Spacer(1, 4))
elif stripped.startswith("# "):
continue
elif stripped.startswith("### "):
story.append(Paragraph(stripped[4:], sH3))
elif stripped.startswith("## "):
story.append(Paragraph(stripped[3:], sH2))
elif stripped.startswith("---"):
story.append(Spacer(1, 8))
elif stripped.startswith("*") and stripped.endswith("*"):
story.append(Paragraph(f"<i>{stripped.strip('*')}</i>", sMeta))
elif stripped.startswith("- [x] "):
story.append(Paragraph(f"\u2713 {stripped[6:]}", sBullet))
elif stripped.startswith("- [ ] "):
story.append(Paragraph(f"\u25CB {stripped[6:]}", sBullet))
elif stripped.startswith("- "):
story.append(Paragraph(f"\u2022 {stripped[2:]}", sBullet))
else:
text = stripped.replace("**", "<b>").replace("__", "<b>")
text = text.replace("*", "<i>").replace("_", "<i>")
story.append(Paragraph(text, sBody))
for line in markdownText.split("\n"):
stripped = line.strip()
if not stripped:
story.append(Spacer(1, 4))
elif stripped.startswith("# "):
continue
elif stripped.startswith("### "):
story.append(Paragraph(_escXml(stripped[4:]), sH3))
elif stripped.startswith("## "):
story.append(Paragraph(_escXml(stripped[3:]), sH2))
elif stripped.startswith("---"):
story.append(Spacer(1, 8))
elif stripped.startswith("*") and stripped.endswith("*") and not stripped.startswith("**"):
story.append(Paragraph(f"<i>{_escXml(stripped.strip('*'))}</i>", sMeta))
elif stripped.startswith("- [x] "):
story.append(Paragraph(f"\u2713 {_mdToXml(stripped[6:])}", sBullet))
elif stripped.startswith("- [ ] "):
story.append(Paragraph(f"\u25CB {_mdToXml(stripped[6:])}", sBullet))
elif stripped.startswith("- "):
story.append(Paragraph(f"\u2022 {_mdToXml(stripped[2:])}", sBullet))
else:
story.append(Paragraph(_mdToXml(stripped), sBody))
doc.build(story)
buf.seek(0)
return buf.getvalue()
except Exception as e:
logger.warning(f"PDF generation failed: {e}")
return None
doc.build(story)
buf.seek(0)
return buf.getvalue()
def _escXml(text: str) -> str:
"""Escape text for reportlab XML paragraphs."""
return text.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
def _mdToXml(text: str) -> str:
"""Convert markdown inline formatting to reportlab XML. Bold, italic, escape the rest."""
import re as _re
text = text.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
text = _re.sub(r'\*\*(.+?)\*\*', r'<b>\1</b>', text)
text = _re.sub(r'__(.+?)__', r'<b>\1</b>', text)
text = _re.sub(r'\*(.+?)\*', r'<i>\1</i>', text)
text = _re.sub(r'_(.+?)_', r'<i>\1</i>', text)
return text