Update SearchClient to use webPageUrl instead of static fileName for …

…webpages (fixes #491) (#521)

## Motivation and Context (Why the change? What's the scenario?)

When providing webpages as facts, the "filename" currently is a static
"content.url" - this provides no value when asking the LLM to include
sources directly in the response (e.g. to have per paragraph sources).

Update SearchClient to use webPageUrl instead of static fileName for
webpages.

## High level description (Approach, Design)

When creating the facts, instead of "content.url" the webpage url is
added

Co-authored-by: Michael Keller <[email protected]>

service/Core/Search/SearchClient.cs

@@ -238,6 +238,8 @@ public async Task<MemoryAnswer> AskAsync(
 
             string fileName = memory.GetFileName(this._log);
 
+            string webPageUrl = memory.GetWebPageUrl(index);
+
             var partitionText = memory.GetPartitionText(this._log).Trim();
             if (string.IsNullOrEmpty(partitionText))
             {
@@ -248,7 +250,7 @@ public async Task<MemoryAnswer> AskAsync(
             factsAvailableCount++;
 
             // TODO: add file age in days, to push relevance of newer documents
-            var fact = $"==== [File:{fileName};Relevance:{relevance:P1}]:\n{partitionText}\n";
+            var fact = $"==== [File:{(fileName == "content.url" ? webPageUrl : fileName)};Relevance:{relevance:P1}]:\n{partitionText}\n";
 
             // Use the partition/chunk only if there's room for it
             var size = this._textGenerator.CountTokens(fact);