Merge pull request #38 from codefrydev/RichPdf

Rich pdf

Author: PrashantUnity

docs/MCP.md

@@ -214,7 +214,7 @@ Export tools write artifact files with a 24-hour TTL; in-app chat renders downlo
 
 ### Export and deliverables
 
-`export_audit_report`, `export_compare_csv`, `export_list_as_csv`, `export_sitemap_xml`, `validate_rich_results`, `compose_custom_report`, `export_custom_report`, `list_export_formats`
+`export_audit_report`, `export_compare_csv`, `export_list_as_csv`, `export_sitemap_xml`, `validate_rich_results`, `list_export_formats`
 
 Full audit exports use the same generators as the Export view. PDF export requires `reportlab`.
 

src/website_profiling/llm/agent.py

@@ -79,7 +79,6 @@ def _max_tool_rounds(cfg: dict[str, str]) -> int:
 - Full audit PDF/HTML/CSV/JSON: export_audit_report with format pdf|html|csv|json
 - Compare issue diff CSV: export_compare_csv with baseline_report_id
 - Export a list as CSV: export_list_as_csv with tool_name and tool_args (e.g. list_broken_links)
-- Custom client report: compose_custom_report with title and sections (executive_summary, category_scores, tool, notes), then export_custom_report format=pdf or html
 - After export tools succeed, tell the user their download is ready; the UI renders file buttons automatically
 
 Visualization playbook (chat UI renders charts and tables from tool JSON automatically):

src/website_profiling/reporting/pdf/__init__.py

@@ -0,0 +1,15 @@
+"""PDF document model and export pipeline."""
+from __future__ import annotations
+
+from .builder import build_pdf_document
+from .document import PdfDocument
+from .options import PdfBuildOptions, PdfLimits
+from .render import render_pdf_document
+
+__all__ = [
+    "build_pdf_document",
+    "render_pdf_document",
+    "PdfDocument",
+    "PdfBuildOptions",
+    "PdfLimits",
+]

src/website_profiling/reporting/pdf/adapters/__init__.py

@@ -0,0 +1,29 @@
+"""Section adapter registry.
+
+Each adapter maps a section key to a function that accepts the raw payload
+dict + PdfBuildOptions and returns a list of PdfSection objects.  Adapters
+that find no relevant data return an empty list.
+"""
+from __future__ import annotations
+
+from typing import Any, Callable
+
+from ..document import PdfSection
+from ..options import PdfBuildOptions
+
+SectionAdapterFn = Callable[[dict[str, Any], PdfBuildOptions], list[PdfSection]]
+
+# Populated by each sub-module at import time
+SECTION_ADAPTERS: dict[str, SectionAdapterFn] = {}
+
+
+def register(key: str) -> Callable[[SectionAdapterFn], SectionAdapterFn]:
+    """Decorator: @register("lighthouse") marks a function as a section adapter."""
+    def _wrap(fn: SectionAdapterFn) -> SectionAdapterFn:
+        SECTION_ADAPTERS[key] = fn
+        return fn
+    return _wrap
+
+
+# Import adapters so they self-register
+from . import core, findings, appendix  # noqa: E402, F401

src/website_profiling/reporting/pdf/adapters/appendix.py

@@ -0,0 +1,68 @@
+"""Appendix adapter — crawled URL sample and data-source glossary."""
+from __future__ import annotations
+
+from typing import Any
+
+from ....tools.export_audit_data import _GLOSSARY_ROWS
+from ..document import (
+    KeyValueBlock,
+    PdfSection,
+    PdfTruncation,
+    SpacerBlock,
+    UrlListBlock,
+)
+from ..options import PdfBuildOptions
+from . import register
+
+
+@register("appendix")
+def adapt_appendix(payload: dict[str, Any], opts: PdfBuildOptions) -> list[PdfSection]:
+    if not opts.include_appendix:
+        return []
+
+    sections: list[PdfSection] = []
+
+    # --- Crawled URLs sample ---
+    links = [l for l in (payload.get("links") or []) if isinstance(l, dict)]
+    if links:
+        limit = opts.limits.urls_sample
+        sample = links[:limit]
+        rows = [
+            {
+                "url": str(lnk.get("url") or ""),
+                "status": str(lnk.get("status") or ""),
+                "title": str(lnk.get("title") or "").strip(),
+            }
+            for lnk in sample
+        ]
+        has_titles = any(r["title"] for r in rows)
+        trunc = PdfTruncation(shown=len(rows), total=len(links)) if len(links) > limit else None
+        sections.append(PdfSection(
+            id="appendix.urls",
+            section_key="links",
+            title="Crawled URLs (sample)",
+            priority=80,
+            page_break_before=False,
+            blocks=[
+                UrlListBlock(
+                    id="appendix.url_list",
+                    rows=rows,
+                    show_title=has_titles,
+                    truncation=trunc,
+                ),
+                SpacerBlock(id="appendix.url_spacer", height_pt=6),
+            ],
+        ))
+
+    # --- Glossary ---
+    if opts.include_glossary:
+        gloss_rows = [(term, desc) for term, desc in _GLOSSARY_ROWS]
+        sections.append(PdfSection(
+            id="appendix.glossary",
+            section_key="core",
+            title="Data source glossary",
+            priority=90,
+            blocks=[KeyValueBlock(id="appendix.glossary_kv", rows=gloss_rows, layout="glossary")],
+        ))
+
+    return sections

src/website_profiling/reporting/pdf/adapters/core.py

@@ -0,0 +1,38 @@
+"""Core adapter — audit-details section (category scores live on cover)."""
+from __future__ import annotations
+
+from typing import Any
+
+from ....tools.export_audit_data import _format_report_date, _summary_lines
+from ..document import KeyValueBlock, PdfSection, SpacerBlock
+from ..options import PdfBuildOptions
+from . import register
+
+
+@register("core")
+def adapt_core(payload: dict[str, Any], opts: PdfBuildOptions) -> list[PdfSection]:
+    sections: list[PdfSection] = []
+
+    # Category scores are rendered on the cover page — not duplicated here.
+
+    # --- Audit details section ---
+    summary_rows = _summary_lines(payload)
+    if summary_rows:
+        formatted_rows: list[tuple[str, str]] = []
+        for key, val in summary_rows:
+            if key == "Report generated":
+                formatted_rows.append((key, _format_report_date(val)))
+            else:
+                formatted_rows.append((key, val))
+        sections.append(PdfSection(
+            id="core.audit_details",
+            section_key="core",
+            title="Audit details",
+            priority=70,
+            blocks=[
+                KeyValueBlock(id="core.audit_kv", rows=formatted_rows, layout="audit"),
+                SpacerBlock(id="core.audit_spacer", height_pt=6),
+            ],
+        ))
+
+    return sections

src/website_profiling/reporting/pdf/adapters/findings.py

@@ -0,0 +1,54 @@
+"""Findings adapter — normalizes and groups all audit issues."""
+from __future__ import annotations
+
+from typing import Any
+
+from ....tools.export_audit_data import _issues_rows, _priority_sort_key
+from ..document import PdfSection, PdfTruncation
+from ..normalize import group_issues_for_pdf, normalize_issue_for_pdf
+from ..options import PdfBuildOptions
+from . import register
+
+
+@register("findings")
+def adapt_findings(payload: dict[str, Any], opts: PdfBuildOptions) -> list[PdfSection]:
+    raw_rows = _issues_rows(payload)
+    if not raw_rows:
+        return []
+
+    raw_rows = sorted(raw_rows, key=_priority_sort_key)
+    total = len(raw_rows)
+    capped = raw_rows[: opts.limits.issues_total]
+
+    pdf_issues = [
+        normalize_issue_for_pdf(row, include_recommendation=opts.include_recommendations)
+        for row in capped
+    ]
+
+    groups = group_issues_for_pdf(
+        pdf_issues,
+        issues_per_group=opts.limits.issues_per_group,
+        issues_total=opts.limits.issues_total,
+    )
+
+    if not groups:
+        return []
+
+    section_trunc: PdfTruncation | None = None
+    if total > opts.limits.issues_total:
+        section_trunc = PdfTruncation(
+            shown=opts.limits.issues_total,
+            total=total,
+            reason="limit",
+            continue_in=["CSV", "workbook"],
+        )
+
+    return [PdfSection(
+        id="findings",
+        section_key="findings",
+        title="Findings",
+        priority=20,
+        page_break_before=False,
+        blocks=list(groups),  # type: ignore[arg-type]
+        truncation=section_trunc,
+    )]