Initial commit. monorepo deps are copied in a dirty way to the vendor folder

report_generator/layout.py

@@ -0,0 +1,90 @@
+"""Renders the SAST/SCA report template.
+
+The template and its stylesheet are the design; this module is only what turns them into a
+document. It takes a context that is already shaped for the page -- every number formatted, every
+bar already a percentage -- so that the template places strings and never computes.
+
+Nothing here reads an artifact. Filling a context from a scan is :mod:`context`, which works from
+what :mod:`sources` collected; :func:`sample.sample_context` builds one by hand, which is what the
+template tests render.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Any
+
+from jinja2 import Environment, FileSystemLoader, select_autoescape
+
+TEMPLATE_DIR = Path(__file__).resolve().parent / "template"
+
+TEMPLATE_NAME = "report.html.j2"
+
+STYLESHEET = TEMPLATE_DIR / "report.css"
+
+# What the document calls itself, in the PDF's own metadata rather than only on its first page.
+TITLE = "SAST/SCA Report"
+
+AUTHOR = "CipherCell"
+
+# The shape :func:`render_html` expects. Written down because the template guards on it: a key that
+# is absent leaves its section out, which is how a scan that ran half the pipeline produces a
+# shorter report rather than one full of empty tables.
+CONTEXT = {
+    "cover": "scan_id, footer_date, logo, title, identity[], severity_ledger{title, rows[]}",
+    "scan": "title, overview[], stages[{title, rows[], table?}]",
+    "project": "title, overview[], languages{title, rows[]}",
+    "summary": "title, overview[], distributions[{title, rows[], legend}]",
+    "critical": "title, table{columns[], rows[]}",
+    "sections": "[{title, new_page?, groups[{title, note?, cards[], table?}]}]",
+}
+
+
+def render_html(context: dict[str, Any]) -> str:
+    """Render the report as an HTML document.
+
+    Args:
+        context: The page-ready context; see :data:`CONTEXT`.
+
+    Returns:
+        The rendered HTML.
+    """
+    environment = Environment(
+        loader=FileSystemLoader(TEMPLATE_DIR),
+        autoescape=select_autoescape(default_for_string=True, default=True),
+        trim_blocks=True,
+        lstrip_blocks=True,
+    )
+
+    return environment.get_template(TEMPLATE_NAME).render(**context)
+
+
+def render_pdf(context: dict[str, Any]) -> bytes:
+    """Render the report as a PDF.
+
+    Args:
+        context: The page-ready context; see :data:`CONTEXT`.
+
+    Returns:
+        The PDF.
+    """
+    # WeasyPrint pulls in native cairo and pango bindings. Imported here so that importing this
+    # module -- to render the HTML, or merely to read CONTEXT -- does not pay for them.
+    from weasyprint import CSS, HTML
+    from weasyprint.text.fonts import FontConfiguration
+
+    # Without this the @font-face rules are ignored and Space Grotesk is silently replaced by
+    # whatever the host happens to have, which is how the design sample came to be set in Carlito.
+    fonts = FontConfiguration()
+
+    # Relative URLs in the template and the stylesheet -- the mark, the font files -- resolve
+    # against the template directory.
+    document = HTML(string=render_html(context), base_url=str(TEMPLATE_DIR)).render(
+        stylesheets=[CSS(filename=str(STYLESHEET), font_config=fonts)],
+        font_config=fonts,
+    )
+
+    document.metadata.title = f"{TITLE} - {context.get('cover', {}).get('subject', '')}".strip(" -")
+    document.metadata.authors = [AUTHOR]
+
+    return bytes(document.write_pdf())