feat: fase 3 — export PDF con reportlab + /export command

2026-05-04 12:57:21 +00:00
parent c33bb5337d
commit 4c7f5b521b
5 changed files with 198 additions and 0 deletions
@@ -445,3 +445,109 @@ Iterations: {session.get('iterations', 0)}
 Total words researched: {session.get('total_words', 0):,}
 ---
 """
+
+
+def generate_pdf(content: str, title: str = "ResearchOwl Output") -> bytes:
+    try:
+        from reportlab.lib.pagesizes import A4
+        from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
+        from reportlab.lib.units import cm
+        from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, HRFlowable
+        from reportlab.lib.enums import TA_LEFT
+        from reportlab.lib import colors
+        import io
+        import re
+    except ImportError:
+        raise ImportError("reportlab is required for PDF export — pip install reportlab")
+
+    buf = io.BytesIO()
+    doc = SimpleDocTemplate(
+        buf,
+        pagesize=A4,
+        rightMargin=2 * cm,
+        leftMargin=2 * cm,
+        topMargin=2.5 * cm,
+        bottomMargin=2 * cm,
+        title=title,
+    )
+
+    base = getSampleStyleSheet()
+    normal = ParagraphStyle("RO_Normal", parent=base["Normal"],
+                            fontSize=10, leading=14, spaceAfter=4)
+    h1 = ParagraphStyle("RO_H1", parent=base["Heading1"],
+                        fontSize=18, spaceBefore=12, spaceAfter=6,
+                        textColor=colors.HexColor("#1a1a2e"))
+    h2 = ParagraphStyle("RO_H2", parent=base["Heading2"],
+                        fontSize=14, spaceBefore=10, spaceAfter=4,
+                        textColor=colors.HexColor("#16213e"))
+    h3 = ParagraphStyle("RO_H3", parent=base["Heading3"],
+                        fontSize=12, spaceBefore=8, spaceAfter=4)
+    code_style = ParagraphStyle("RO_Code", parent=base["Code"],
+                                fontSize=9, leading=12, fontName="Courier",
+                                backColor=colors.HexColor("#f4f4f4"), spaceAfter=4)
+    bullet_style = ParagraphStyle("RO_Bullet", parent=normal,
+                                  leftIndent=20, bulletIndent=10, spaceAfter=2)
+
+    def md_to_para(text: str) -> str:
+        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)
+        text = re.sub(r'`(.+?)`', r'<font name="Courier">\1</font>', text)
+        return text
+
+    story = []
+    lines = content.split("\n")
+    in_code = False
+    code_buf = []
+
+    for line in lines:
+        if line.startswith("```"):
+            if not in_code:
+                in_code = True
+                code_buf = []
+            else:
+                in_code = False
+                try:
+                    story.append(Paragraph(
+                        "<br/>".join(l.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
+                                     for l in code_buf),
+                        code_style
+                    ))
+                except Exception:
+                    pass
+            continue
+
+        if in_code:
+            code_buf.append(line)
+            continue
+
+        if re.match(r'^[-*_]{3,}$', line.strip()):
+            story.append(HRFlowable(width="100%", thickness=0.5,
+                                    color=colors.grey, spaceAfter=6))
+            continue
+
+        try:
+            if line.startswith("### "):
+                story.append(Paragraph(md_to_para(line[4:]), h3))
+            elif line.startswith("## "):
+                story.append(Paragraph(md_to_para(line[3:]), h2))
+            elif line.startswith("# "):
+                story.append(Paragraph(md_to_para(line[2:]), h1))
+            elif re.match(r'^[-*+] ', line):
+                story.append(Paragraph("• " + md_to_para(line[2:]), bullet_style))
+            elif re.match(r'^\d+\. ', line):
+                story.append(Paragraph(md_to_para(line), bullet_style))
+            elif line.strip() == "":
+                story.append(Spacer(1, 6))
+            else:
+                story.append(Paragraph(md_to_para(line), normal))
+        except Exception:
+            try:
+                story.append(Paragraph(line[:300], normal))
+            except Exception:
+                pass
+
+    doc.build(story)
+    return buf.getvalue()