Mark a truncated preview instead of cutting it silently

haiku_rag_slim/haiku/rag/inspector/widgets/search_modal.py

@@ -9,6 +9,7 @@ from textual.widgets import Input, ListItem, ListView, Static
 from haiku.rag.client import HaikuRAG
 from haiku.rag.inspector.widgets.detail_view import DetailView
 from haiku.rag.store.models import Chunk, SearchResult
+from haiku.rag.utils import truncated
 
 
 class SearchModal(Screen):
@@ -112,7 +113,7 @@ class SearchModal(Screen):
 
             await list_view.clear()
             for result in self.search_results:
-                first_line = result.content.split("\n")[0][:60]
+                first_line = truncated(result.content.split("\n")[0], 60)
                 score_str = f"{result.score:.2f}"
                 # Add page info if available
                 page_info = ""

haiku_rag_slim/haiku/rag/utils.py

@@ -362,6 +362,17 @@ def format_citations(citations: "list[Citation]") -> str:
     return "\n".join(lines)
 
 
+def truncated(text: str, limit: int) -> str:
+    """`text` cut to `limit` characters, marked where anything was dropped.
+
+    Without the mark a clipped value reads as the value: a sentence ending
+    "commissioned in 1991" becomes "commissioned in 1".
+    """
+    if len(text) <= limit:
+        return text
+    return text[:limit].rstrip() + "…"
+
+
 async def format_citations_rich(
     citations: "list[Citation]",
     client: "HaikuRAG | None" = None,
@@ -415,10 +426,7 @@ async def format_citations_rich(
                 else Text(f"[Figure: {ref}]", style="italic dim")
             )
 
-        preview = c.content
-        if len(preview) > CITATION_PREVIEW_CHARS:
-            preview = preview[:CITATION_PREVIEW_CHARS].rstrip() + "…"
-        body.append(Text(preview))
+        body.append(Text(truncated(c.content, CITATION_PREVIEW_CHARS)))
 
         footer = Text()
         footer.append("doc: ", style="dim")

tests/test_utils.py

@@ -793,6 +793,20 @@ async def test_format_citations_rich_names_the_database_when_federating():
     assert "medic" in output
 
 
+def test_truncated_marks_what_it_dropped():
+    """An unmarked cut reads as the value: a sentence ending "in 1991" becomes
+    one ending "in 1"."""
+    from haiku.rag.utils import truncated
+
+    sentence = "Station Kestrel sits at 980 metres and was commissioned in 1991."
+
+    assert truncated(sentence, 60) == sentence[:60] + "…"
+    assert truncated(sentence, len(sentence)) == sentence
+    assert truncated("short", 60) == "short"
+    # Trailing space before the mark reads as a gap in the text.
+    assert truncated("a bc", 2) == "a…"
+
+
 async def test_format_citations_rich_omits_the_database_for_one_database():
     """A single database is not worth naming on every citation."""
     from unittest.mock import AsyncMock