From 271d75270166ef0d596ab434fe0d6e8a7733b4f7 Mon Sep 17 00:00:00 2001
From: Jared Van Bortel <jared@nomic.ai>
Date: Thu, 18 Apr 2024 14:51:13 -0400
Subject: [PATCH] localdocs: small but important fixes to local docs (#2236)

* chat: use .rmodel extension for Nomic Embed

Signed-off-by: Jared Van Bortel <jared@nomic.ai>

* database: fix order of SQL arguments in updateDocument

Signed-off-by: Jared Van Bortel <jared@nomic.ai>

---------

Signed-off-by: Jared Van Bortel <jared@nomic.ai>
---
 gpt4all-chat/chatllm.cpp   |  8 ++++----
 gpt4all-chat/database.cpp  |  2 +-
 gpt4all-chat/embllm.cpp    | 13 +++++++++----
 gpt4all-chat/modellist.cpp |  6 +++---
 4 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/gpt4all-chat/chatllm.cpp b/gpt4all-chat/chatllm.cpp
index 37f494c1..b062ef44 100644
--- a/gpt4all-chat/chatllm.cpp
+++ b/gpt4all-chat/chatllm.cpp
@@ -283,10 +283,10 @@ bool ChatLLM::loadModel(const ModelInfo &modelInfo)
             QString modelName;
             {
                 QFile file(filePath);
-                file.open(QIODeviceBase::ReadOnly | QIODeviceBase::Text);
-                QTextStream stream(&file);
-                QString text = stream.readAll();
-                QJsonDocument doc = QJsonDocument::fromJson(text.toUtf8());
+                bool success = file.open(QIODeviceBase::ReadOnly);
+                (void)success;
+                Q_ASSERT(success);
+                QJsonDocument doc = QJsonDocument::fromJson(file.readAll());
                 QJsonObject obj = doc.object();
                 apiKey = obj["apiKey"].toString();
                 modelName = obj["modelName"].toString();
diff --git a/gpt4all-chat/database.cpp b/gpt4all-chat/database.cpp
index f572e048..087ab324 100644
--- a/gpt4all-chat/database.cpp
+++ b/gpt4all-chat/database.cpp
@@ -410,8 +410,8 @@ bool updateDocument(QSqlQuery &q, int id, qint64 document_time)
 {
     if (!q.prepare(UPDATE_DOCUMENT_TIME_SQL))
         return false;
-    q.addBindValue(id);
     q.addBindValue(document_time);
+    q.addBindValue(id);
     return q.exec();
 }
 
diff --git a/gpt4all-chat/embllm.cpp b/gpt4all-chat/embllm.cpp
index 95c61834..82826172 100644
--- a/gpt4all-chat/embllm.cpp
+++ b/gpt4all-chat/embllm.cpp
@@ -42,12 +42,17 @@ bool EmbeddingLLMWorker::loadModel()
     }
 
     auto filename = fileInfo.fileName();
-    bool isNomic = filename.startsWith("nomic-") && filename.endsWith(".txt");
+    bool isNomic = filename.startsWith("gpt4all-nomic-") && filename.endsWith(".rmodel");
     if (isNomic) {
         QFile file(filePath);
-        file.open(QIODeviceBase::ReadOnly | QIODeviceBase::Text);
-        QTextStream stream(&file);
-        m_nomicAPIKey = stream.readAll();
+        if (!file.open(QIODeviceBase::ReadOnly)) {
+            qWarning() << "failed to open" << filePath << ":" << file.errorString();
+            m_model = nullptr;
+            return false;
+        }
+        QJsonDocument doc = QJsonDocument::fromJson(file.readAll());
+        QJsonObject obj = doc.object();
+        m_nomicAPIKey = obj["apiKey"].toString();
         file.close();
         return true;
     }
diff --git a/gpt4all-chat/modellist.cpp b/gpt4all-chat/modellist.cpp
index 4cfd097a..6d0f7c73 100644
--- a/gpt4all-chat/modellist.cpp
+++ b/gpt4all-chat/modellist.cpp
@@ -12,7 +12,7 @@
 
 const char * const KNOWN_EMBEDDING_MODELS[] {
     "all-MiniLM-L6-v2.gguf2.f16.gguf",
-    "nomic-embed-text-v1.txt",
+    "gpt4all-nomic-embed-text-v1.rmodel",
 };
 
 QString ModelInfo::id() const
@@ -1178,7 +1178,7 @@ void ModelList::updateModelsFromDirectory()
             it.next();
             if (!it.fileInfo().isDir()) {
                 QString filename = it.fileName();
-                if (filename.startsWith("chatgpt-") && filename.endsWith(".txt")) {
+                if (filename.endsWith(".txt") && (filename.startsWith("chatgpt-") || filename.startsWith("nomic-"))) {
                     QString apikey;
                     QString modelname(filename);
                     modelname.chop(4); // strip ".txt" extension
@@ -1648,7 +1648,7 @@ void ModelList::parseModelsJsonFile(const QByteArray &jsonData, bool save)
             "<li>You can apply for an API key <a href=\"https://atlas.nomic.ai/\">with Nomic Atlas.</a></li>");
         const QString modelName = "Nomic Embed";
         const QString id = modelName;
-        const QString modelFilename = "nomic-embed-text-v1.txt"; // FIXME: This should be made to use '.rmodel' as well
+        const QString modelFilename = "gpt4all-nomic-embed-text-v1.rmodel";
         if (contains(modelFilename))
             changeId(modelFilename, id);
         if (!contains(id))