fix: web search content blocks numbered by fetch completion order break citations (#1672)

services/search/core.py

@@ -356,6 +356,12 @@ def comprehensive_web_search(
         for r in search_results if r.get("url")
     ]
 
+    # Map each URL to its [i] number in the sources list so fetched content
+    # blocks can be labeled with the SAME index the model cites.
+    _url_index = {
+        r["url"]: i for i, r in enumerate(search_results, 1) if r.get("url")
+    }
+
     # Fetch content in parallel
     fetched_content = []
     with ThreadPoolExecutor(max_workers=max_workers) as executor:
@@ -368,6 +374,10 @@ def comprehensive_web_search(
             try:
                 result = future.result()
                 if result["success"] and result["content"] and len(result["content"]) >= min_content_length:
+                    # Remember which source this fetch belongs to: redirects
+                    # can change result["url"] and completion order is
+                    # arbitrary, so the block label cannot be recomputed later.
+                    result["source_index"] = _url_index.get(url)
                     fetched_content.append(result)
             except Exception as e:
                 logger.error(f"Exception while fetching {url}: {str(e)}")
@@ -408,8 +418,15 @@ def comprehensive_web_search(
         output_parts.append("FETCHED PAGE CONTENT:")
         output_parts.append("-" * 50)
 
-        for i, content in enumerate(fetched_content, 1):
-            output_parts.append(f"\n[CONTENT {i}] From: {content['url']}")
+        # Emit blocks in source order, numbered with the same [i] as the
+        # sources list, so [CONTENT 2] really is content from source [2].
+        # Before this, blocks were numbered 1..N in fetch COMPLETION order,
+        # which matched neither the sources list nor each other run to run.
+        fetched_content.sort(key=lambda c: c.get("source_index") or len(search_results) + 1)
+        for content in fetched_content:
+            _idx = content.get("source_index")
+            _label = f"[CONTENT {_idx}]" if _idx else "[CONTENT]"
+            output_parts.append(f"\n{_label} From: {content['url']}")
             output_parts.append(f"Title: {content['title']}")
             output_parts.append("-" * 30)
 

tests/test_search_content_block_source_index.py

@@ -0,0 +1,61 @@
+"""[CONTENT i] blocks must map to the [i] sources list.
+
+comprehensive_web_search numbers its sources list by search-result order,
+but the fetched-content blocks were numbered 1..N in fetch COMPLETION
+order (as_completed). With parallel fetching the two numberings disagree,
+so the model cites "[2]" for content that actually came from source [3].
+"""
+
+import importlib
+import time
+
+import pytest
+
+
+@pytest.fixture
+def core(monkeypatch):
+    mod = importlib.import_module("services.search.core")
+    results = [
+        {"url": "http://one.example/a", "title": "One", "snippet": "s1"},
+        {"url": "http://two.example/b", "title": "Two", "snippet": "s2"},
+    ]
+    monkeypatch.setattr(mod, "_get_search_settings", lambda: {"search_provider": "searxng"})
+    monkeypatch.setattr(mod, "_get_result_count", lambda: 2)
+    monkeypatch.setattr(mod, "_call_provider", lambda *a, **k: [dict(r) for r in results])
+    monkeypatch.setattr(mod, "rank_search_results", lambda q, r: r)
+    return mod
+
+
+def _fake_fetch_delaying_first(url, timeout=8, retry_attempt=0):
+    if "one.example" in url:
+        # Force the FIRST source to finish fetching LAST
+        time.sleep(0.4)
+    return {
+        "success": True,
+        "url": url,
+        "title": "Title for " + url,
+        "content": "Content for " + url + " " + "filler " * 20,
+    }
+
+
+def test_content_blocks_numbered_by_source_not_completion_order(core, monkeypatch):
+    monkeypatch.setattr(core, "fetch_webpage_content", _fake_fetch_delaying_first)
+    out = core.comprehensive_web_search("test query", max_pages=2, max_workers=2)
+    assert "[CONTENT 1] From: http://one.example/a" in out
+    assert "[CONTENT 2] From: http://two.example/b" in out
+    assert out.index("[CONTENT 1]") < out.index("[CONTENT 2]")
+
+
+def test_redirected_fetch_keeps_its_source_index(core, monkeypatch):
+    def fetch(url, timeout=8, retry_attempt=0):
+        final = "http://final.example/landing" if "two.example" in url else url
+        return {
+            "success": True,
+            "url": final,
+            "title": "Title",
+            "content": "Content for " + final + " " + "filler " * 20,
+        }
+
+    monkeypatch.setattr(core, "fetch_webpage_content", fetch)
+    out = core.comprehensive_web_search("test query", max_pages=2, max_workers=2)
+    assert "[CONTENT 2] From: http://final.example/landing" in out