fix(materialize): emit frontmatter and index values verbatim

a7e1bc8 unified title rendering at both sites and routed the open question
to commons, the spec owner: does §5's whitespace collapse apply to `title`,
or only to `source_query`? The ruling is source_query only. §5 mandates the
collapse for that one field — where a multi-line SQL SELECT must render on
one line — while every other value is validated single-line at manifest load
and emitted as-is: validation, not repair.

_single_line is renamed _collapse_whitespace and now runs for source_query
alone. Frontmatter values, the index link label, and the labels_by_target
map all pass through verbatim. A title carrying an internal whitespace run
(or any other operator-supplied bytes) now survives byte-for-byte at both
the frontmatter and the index-label site, instead of being silently altered.

No golden bytes change: no title in the shipped fixtures or either consumer
carries a collapsible whitespace run.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HBbjgS5A55RVavoyjJC4FX

src/llm_ingestion_okf/materialize.py

@@ -49,18 +49,22 @@ class IngestResult:
     stamp: str
 
 
-def _single_line(value: str) -> str:
-    # §5: frontmatter values are single-lined — whitespace runs, including
-    # newlines, collapse to one space — because the format is line-oriented
-    # and parsing stops at the first `---` line. §4 makes the extraction
-    # title BOTH the `title` frontmatter and the index link label, so the
-    # same rule has to reach both sites or the one value renders two ways.
+def _collapse_whitespace(value: str) -> str:
+    # §5 mandates whitespace-run collapse for ONE field only: `source_query`
+    # (ingest-spec.md:140-141), where a legitimately multi-line SQL SELECT
+    # must render on one line. Every other value is validated single-line at
+    # manifest load and emitted verbatim — validation, not repair — so
+    # operator-supplied bytes (e.g. a title's internal double space) survive.
     return " ".join(value.split())
 
 
 def _render_frontmatter(frontmatter: dict[str, str]) -> str:
-    # Line-oriented `key: value`, insertion order.
-    return "\n".join(f"{key}: {_single_line(value)}" for key, value in frontmatter.items())
+    # Line-oriented `key: value`, insertion order. Only `source_query` is
+    # collapsed; all other values pass through verbatim.
+    return "\n".join(
+        f"{key}: {_collapse_whitespace(value) if key == 'source_query' else value}"
+        for key, value in frontmatter.items()
+    )
 
 
 def _parse_frontmatter(path: Path) -> dict[str, str]:
@@ -148,7 +152,7 @@ def _link_in_index(bundle_dir: Path, target_name: str, label: str) -> None:
     if f"]({target_name})" in body:
         return
     prefix = body if body.endswith("\n") else body + "\n"
-    index_path.write_bytes(f"{prefix}- [{_single_line(label)}]({target_name})\n".encode())
+    index_path.write_bytes(f"{prefix}- [{label}]({target_name})\n".encode())
 
 
 def materialize_bundle(
@@ -272,8 +276,7 @@ def materialize_bundle(
     # bundle_summary as its body; links are appended in extraction order.
     index_path = bundle / _INDEX_NAME
     labels_by_target = {
-        generated_filename(extraction.id): _single_line(extraction.title)
-        for extraction in manifest.extractions
+        generated_filename(extraction.id): extraction.title for extraction in manifest.extractions
     }
     if not index_path.is_file():
         _write_bytes(bundle, _INDEX_NAME, manifest.bundle_summary + "\n")

tests/test_materialize.py

@@ -159,6 +159,38 @@ def test_title_renders_identically_in_frontmatter_and_index(tmp_path: Path) -> N
     assert frontmatter_title == index_label
 
 
+def test_title_emitted_verbatim_not_collapsed(tmp_path: Path) -> None:
+    # §5 mandates whitespace-run collapse ONLY for `source_query`
+    # (ingest-spec.md:140-141). `title` is validated single-line at manifest
+    # load and emitted verbatim — validation, not repair — so an internal
+    # whitespace run survives at BOTH the frontmatter and the index-label site.
+    src = tmp_path / "src"
+    manifest_path = write_manifest(
+        src,
+        file_manifest_data(
+            [
+                {
+                    "id": "orders",
+                    "title": "Energiforbruk  2024",
+                    "query": "orders.csv",
+                    "okf_type": "dataset",
+                    "max_rows": 100,
+                }
+            ]
+        ),
+    )
+    (src / "data").mkdir()
+    (src / "data" / "orders.csv").write_text("a,b\n1,x\n", encoding="utf-8", newline="")
+    bundle = tmp_path / "bundle"
+
+    materialize_bundle(manifest_path, bundle, ingested_at=INGESTED_AT)
+
+    concept = (bundle / "ingest-orders.md").read_text(encoding="utf-8")
+    index = (bundle / "index.md").read_text(encoding="utf-8")
+    assert "title: Energiforbruk  2024\n" in concept
+    assert "- [Energiforbruk  2024](ingest-orders.md)" in index
+
+
 def test_result_lists_written_concept_files(file_setup: tuple[Path, Path]) -> None:
     manifest_path, bundle = file_setup
     result = materialize_bundle(manifest_path, bundle, INGESTED_AT)