Revert "keycloak integration, mysql enablement, megaservice refactor and import sample button"

.github/workflows/_e2e-test.yml

@@ -53,7 +53,7 @@ jobs:
           fi
           sleep 5
           sudo apt install ansible -y
-          ansible-playbook genai-studio.yml -e "container_registry=${OPEA_IMAGE_REPO}opea" -e "container_tag=${{ inputs.tag }}" -e "mysql_host=$(hostname -I | awk '{print $1}')"
+          ansible-playbook genai-studio.yml -e "container_registry=${OPEA_IMAGE_REPO}opea" -e "container_tag=${{ inputs.tag }}"
           sleep 5
           kubectl wait --for=condition=ready pod --all --namespace=studio --timeout=300s --field-selector=status.phase!=Succeeded
           kubectl wait --for=condition=ready pod --all --namespace=monitoring --timeout=300s --field-selector=status.phase!=Succeeded
@@ -74,7 +74,7 @@ jobs:
       - name: Update Playwright Config
         run: |
           NODE_IP=$(kubectl get nodes -o jsonpath='{.items[0].status.addresses[?(@.type=="InternalIP")].address}')
-          sed -i "s|baseURL:.*|baseURL: \"https://$NODE_IP:30007\",|" playwright.config.js
+          sed -i "s|baseURL:.*|baseURL: \"http://$NODE_IP:30007\",|" playwright.config.js
         working-directory: ${{ github.workspace }}/tests/playwright
 
       - name: Run Playwright Tests

app-backend/Dockerfile

@@ -12,12 +12,14 @@ RUN useradd -m -s /bin/bash user && \
     chown -R user /home/user/
 
 WORKDIR /home/user/
-RUN git clone --depth 1 https://github.com/opea-project/GenAIComps.git
+# temporary pointing to v1.1 GenAIComps for Gateway dependency
+RUN git clone https://github.com/opea-project/GenAIComps.git -b v1.1rc
 
 WORKDIR /home/user/GenAIComps
 RUN pip install --no-cache-dir --upgrade pip==24.3.1 setuptools==75.3.0 && \
     pip install --no-cache-dir -r /home/user/GenAIComps/requirements.txt
 
+COPY ./app_gateway.py /home/user/app_gateway.py
 COPY ./templates/microservices/* /home/user/templates/microservices/
 COPY ./megaservice.py /home/user/megaservice.py
 COPY config/* /home/user/config/
@@ -30,4 +32,4 @@ WORKDIR /home/user
 
 RUN echo 'ulimit -S -n 999999' >> ~/.bashrc
 
-ENTRYPOINT ["python", "megaservice.py"]
+ENTRYPOINT ["python", "megaservice.py"]

app-backend/app_gateway.py

@@ -0,0 +1,89 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+import os
+import json
+import logging
+
+# library import
+from fastapi import Request
+from fastapi.responses import StreamingResponse
+
+# comps import
+from comps import Gateway, MicroService, ServiceOrchestrator, ServiceType
+from comps.cores.proto.api_protocol import (
+    AudioChatCompletionRequest,
+    ChatCompletionRequest,
+    ChatCompletionResponse,
+    ChatCompletionResponseChoice,
+    ChatMessage,
+    EmbeddingRequest,
+    UsageInfo
+)
+from comps.cores.proto.docarray import LLMParams, LLMParamsDoc, RerankedDoc, RerankerParms, RetrieverParms, TextDoc
+
+category_params_map = {
+    'LLM': LLMParams,
+    'Reranking': RerankerParms,
+    'Retreiver': RetrieverParms,
+}
+
+class AppGateway(Gateway):
+    def __init__(self, megaservice, host='0.0.0.0', port=8888):
+        try: 
+            with open('config/project-info.json', 'r') as f:
+                self.project_info = json.load(f)
+        except:
+            logging.error('Failed to load project-info.json')
+        super().__init__(
+            megaservice, host, port, '/v1/app-backend', ChatCompletionRequest, ChatCompletionResponse
+        )
+
+    async def handle_request(self, request: Request):
+        data = await request.json()
+        print('\n'*5, '====== handle_request ======\n', data)
+        if 'chat_completion_ids' in self.project_info:
+            prompt = self._handle_message(data['messages'])
+            params = {}
+            llm_parameters = None
+            for id, node in self.project_info['nodes'].items():
+                if node['category'] in category_params_map:
+                    param_class = category_params_map[node['category']]()
+                    param_keys = [key for key in dir(param_class) if not key.startswith('__') and not callable(getattr(param_class, key))]
+                    print('param_keys', param_keys)
+                    params_dict = {}
+                    for key in param_keys:
+                        if key in data:
+                            params_dict[key] = data[key]
+                            # hadle special case for stream and streaming
+                            if key in ['stream', 'streaming']:
+                                params_dict[key] = data.get('stream', True) and data.get('streaming', True)
+                        elif key in node['inference_params']:
+                            params_dict[key] = node['inference_params'][key]
+                    params[id] = params_dict
+                if node['category'] == 'LLM':
+                    params[id]['max_new_tokens'] = params[id]['max_tokens']
+                    llm_parameters = LLMParams(**params[id])
+            result_dict, runtime_graph = await self.megaservice.schedule(
+                initial_inputs={'query':prompt, 'text': prompt},
+                llm_parameters=llm_parameters,
+                params=params,
+            )
+            print('runtime_graph', runtime_graph.graph)
+            for node, response in result_dict.items():
+                if isinstance(response, StreamingResponse):
+                    return response
+            last_node = runtime_graph.all_leaves()[-1]
+            print('result_dict:', result_dict)
+            print('last_node:',last_node)
+            response = result_dict[last_node]['text']
+            choices = []
+            usage = UsageInfo()
+            choices.append(
+                ChatCompletionResponseChoice(
+                    index=0,
+                    message=ChatMessage(role='assistant', content=response),
+                    finish_reason='stop',
+                )
+            )
+            return ChatCompletionResponse(model='custom_app', choices=choices, usage=usage)