any-file to PDF (Phase 1): converters/ dispatcher, images/text/md/csv/json/xml/svg/pdf-passthrough, 112 tests green

converters/xml_converter.py

@@ -0,0 +1,51 @@
+"""XML -> PDF (pretty text dump, monospace).
+
+Validation is explicit: lxml must parse the document, otherwise the job
+fails with "Invalid XML: <reason>" (never a render failure). A valid
+document is rendered as its text content — every text node and tail in
+document order, one per line, whitespace-only nodes dropped — inside
+<pre> with PRE_CSS. Attribute values are intentionally not dumped: the
+contract is that element text survives extraction.
+"""
+from __future__ import annotations
+
+from pathlib import Path
+
+from lxml import etree
+
+from . import common
+from .text_converter import decode_bytes
+
+
+def convert(path: Path, job_id: str, progress_cb=None, output_dir=None,
+            source_name=None) -> str:
+    """Render an XML file to a monospace text-dump PDF."""
+    raw = path.read_bytes()
+    if not raw.strip():
+        raise ValueError("empty file: nothing to convert")
+    text = decode_bytes(raw)
+    try:
+        root = etree.fromstring(text.encode("utf-8"))
+    except etree.XMLSyntaxError as exc:
+        raise ValueError(f"Invalid XML: {exc}") from exc
+
+    lines: list[str] = []
+    for el in root.iter():
+        for chunk in (el.text, el.tail):
+            if not chunk:
+                continue
+            piece = " ".join(chunk.split())
+            if piece:
+                lines.append(piece)
+    if not lines:
+        raise ValueError("empty file: XML has no text content")
+    body = "\n".join(lines)
+
+    def _build():
+        return f"<html><body><pre>{common.escape(body)}</pre></body></html>", path.name
+
+    return common.convert_single(
+        path, job_id, build_html=_build, css=common.PRE_CSS,
+        progress_cb=progress_cb, output_dir=output_dir,
+        name_src=source_name or path.name,
+    )