feat: removendo arquivos e pastas antigas

_antigos/pdfs/admin.py

@@ -1,7 +0,0 @@
-from django.contrib import admin
-
-# from pdfs.models import PDFsModel
-
-# Register your models here.
-
-# admin.site.register(PDFsModel)

_antigos/pdfs/apps.py

@@ -1,6 +0,0 @@
-from django.apps import AppConfig
-
-
-class PdfsConfig(AppConfig):
-    default_auto_field = "django.db.models.BigAutoField"
-    name = "pdfs"

_antigos/pdfs/migrations/0001_initial.py

@@ -1,21 +0,0 @@
-# Generated by Django 4.1 on 2024-11-09 22:42
-
-from django.db import migrations, models
-
-
-class Migration(migrations.Migration):
-
-    initial = True
-
-    dependencies = [
-    ]
-
-    operations = [
-        migrations.CreateModel(
-            name='EndpointTesteModel',
-            fields=[
-                ('id', models.BigAutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')),
-                ('teste', models.CharField(max_length=300)),
-            ],
-        ),
-    ]

_antigos/pdfs/migrations/0002_delete_endpointtestemodel.py

@@ -1,16 +0,0 @@
-# Generated by Django 4.1 on 2024-11-16 00:46
-
-from django.db import migrations
-
-
-class Migration(migrations.Migration):
-
-    dependencies = [
-        ('pdfs', '0001_initial'),
-    ]
-
-    operations = [
-        migrations.DeleteModel(
-            name='EndpointTesteModel',
-        ),
-    ]

_antigos/pdfs/models.py

@@ -1,4 +0,0 @@
-from django.db import models
-
-# Create your models here.
-# class PDFsModel(models.Model):

_antigos/pdfs/serializer.py

@@ -1,8 +0,0 @@
-from rest_framework import serializers
-
-class PDFUploadSerializer(serializers.Serializer):
-    files = serializers.ListField(child=serializers.FileField(), required=True)
-    system_prompt = serializers.CharField(required=True)
-    user_message = serializers.CharField(required=True)
-    model = serializers.CharField(required=False)
-    embedding = serializers.CharField(required=False)

_antigos/pdfs/tests.py

@@ -1,3 +0,0 @@
-from django.test import TestCase
-
-# Create your tests here.

_antigos/pdfs/views.py

@@ -1,52 +0,0 @@
-import tempfile, os
-from pdfs.serializer import PDFUploadSerializer
-from setup.environment import default_model
-from drf_spectacular.utils import extend_schema
-
-from rest_framework.decorators import api_view, parser_classes
-from rest_framework.parsers import MultiPartParser
-from rest_framework.response import Response
-
-from _utils.main import get_llm_answer
-
-@extend_schema(
-    request=PDFUploadSerializer,
-)
-@api_view(["POST"])
-@parser_classes([MultiPartParser])
-def getPDF(request):
-    if request.method == "POST":
-        serializer = PDFUploadSerializer(data=request.data)
-        if serializer.is_valid(raise_exception=True):
-            listaPDFs = []
-            print('\n\n')
-            data = request.data
-            print('data: ', data)
-            embedding = serializer.validated_data.get("embedding", "gpt")
-            model = serializer.validated_data.get("model", default_model)
-
-            # pdf_file = serializer.validated_data['file']
-            for file in serializer.validated_data['files']:
-                print("file: ", file)
-                file.seek(0)
-                # Create a temporary file to save the uploaded PDF
-                with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as temp_file:
-                    # Write the uploaded file content to the temporary file
-                    for chunk in file.chunks():
-                        temp_file.write(chunk)
-                    temp_file_path = temp_file.name  # Get the path of the temporary file
-                    listaPDFs.append(temp_file_path)
-                # print('temp_file_path: ', temp_file_path)
-                print('listaPDFs: ', listaPDFs)
-
-            resposta_llm = None
-            # resposta_llm = get_llm_answer(data["system_prompt"], data["user_message"], temp_file_path, model=model, embedding=embedding)
-            resposta_llm = get_llm_answer(data["system_prompt"], data["user_message"], listaPDFs, model=model, embedding=embedding)
-
-            for file in listaPDFs:
-                os.remove(file)
-            # os.remove(temp_file_path)
-
-            return Response({
-                "Resposta": resposta_llm
-            })

_antigos/resumos/admin.py

@@ -1,3 +0,0 @@
-from django.contrib import admin
-
-# Register your models here.

_antigos/resumos/apps.py

@@ -1,6 +0,0 @@
-from django.apps import AppConfig
-
-
-class ResumosConfig(AppConfig):
-    default_auto_field = 'django.db.models.BigAutoField'
-    name = 'resumos'

_antigos/resumos/models.py

@@ -1,3 +0,0 @@
-from django.db import models
-
-# Create your models here.

_antigos/resumos/serializer.py

@@ -1,29 +0,0 @@
-from rest_framework import serializers
-from setup.environment import default_model
-# from _utils.utils import DEFAULT_SYSTEM_PROMPT
-
-prompt_template = """
-    Based on the following context, provide multiple key points from the document.
-    For each point, create a new paragraph.
-    Each paragraph should be a complete, self-contained insight.
-    
-    Context: {context}
-    
-    Key points:
-    """
-
-class ResumoPDFSerializer(serializers.Serializer):
-    files = serializers.ListField(child=serializers.FileField(), required=True)
-    system_prompt = serializers.CharField(required=False)
-    user_message = serializers.CharField(required=False, default="")
-    model = serializers.CharField(required=False)
-    iterative_refinement = serializers.BooleanField(required=False, default=False)  # type: ignore
-
-class ResumoCursorSerializer(serializers.Serializer):
-    files = serializers.ListField(child=serializers.FileField(), required=True)
-    system_prompt = serializers.CharField(required=False, default=prompt_template)
-    user_message = serializers.CharField(required=False, default="")
-    model = serializers.CharField(required=False, default=default_model)
-    hf_embedding = serializers.CharField(required=False, default="all-MiniLM-L6-v2")
-    chunk_size = serializers.IntegerField(required=False, default=3500)
-    chunk_overlap = serializers.IntegerField(required=False, default=800)

_antigos/resumos/tests.py

@@ -1,3 +0,0 @@
-from django.test import TestCase
-
-# Create your tests here.

_antigos/resumos/views.py

@@ -1,144 +0,0 @@
-from rest_framework.views import APIView
-import tempfile, os
-from rest_framework.response import Response
-from _utils.resumo_simples_cursor import get_llm_summary_answer_by_cursor
-from _utils.utils import DEFAULT_SYSTEM_PROMPT
-from .serializer import (
-    ResumoPDFSerializer,
-    ResumoCursorSerializer,
-)
-from _utils.main import get_llm_answer_summary, get_llm_answer_summary_with_embedding
-from setup.environment import default_model
-from rest_framework.parsers import MultiPartParser
-from drf_spectacular.utils import extend_schema
-
-
-class ResumoView(APIView):
-    parser_classes = [MultiPartParser]
-
-    @extend_schema(
-        request=ResumoPDFSerializer,
-    )
-    def post(self, request):
-        serializer = ResumoPDFSerializer(data=request.data)
-        if serializer.is_valid(raise_exception=True):
-            listaPDFs = []
-            data = serializer.validated_data
-            model = serializer.validated_data.get("model", default_model)
-            print("serializer.validated_data: ", serializer.validated_data)
-
-            for file in serializer.validated_data["files"]:
-                print("file: ", file)
-                file.seek(0)
-                with tempfile.NamedTemporaryFile(
-                    delete=False, suffix=".pdf"
-                ) as temp_file:  # Create a temporary file to save the uploaded PDF
-                    for (
-                        chunk
-                    ) in (
-                        file.chunks()
-                    ):  # Write the uploaded file content to the temporary file
-                        temp_file.write(chunk)
-                    temp_file_path = (
-                        temp_file.name
-                    )  # Get the path of the temporary file
-                    listaPDFs.append(temp_file_path)
-                # print('listaPDFs: ', listaPDFs)
-
-            system_prompt = data.get("system_prompt", DEFAULT_SYSTEM_PROMPT)
-            resposta_llm = get_llm_answer_summary(
-                system_prompt,
-                data["user_message"],
-                listaPDFs,
-                model=model,
-                isIterativeRefinement=data["iterative_refinement"],
-            )
-
-            for file in listaPDFs:
-                os.remove(file)
-
-            return Response({"resposta": resposta_llm})
-
-
-class ResumoEmbeddingView(APIView):
-    parser_classes = [MultiPartParser]
-
-    @extend_schema(
-        request=ResumoPDFSerializer,
-    )
-    def post(self, request):
-        serializer = ResumoPDFSerializer(data=request.data)
-        if serializer.is_valid(raise_exception=True):
-            listaPDFs = []
-            data = serializer.validated_data
-            model = serializer.validated_data.get("model", default_model)
-            print("serializer.validated_data: ", serializer.validated_data)
-
-            for file in serializer.validated_data["files"]:
-                file.seek(0)
-                with tempfile.NamedTemporaryFile(
-                    delete=False, suffix=".pdf"
-                ) as temp_file:  # Create a temporary file to save the uploaded PDF
-                    for (
-                        chunk
-                    ) in (
-                        file.chunks()
-                    ):  # Write the uploaded file content to the temporary file
-                        temp_file.write(chunk)
-                    temp_file_path = (
-                        temp_file.name
-                    )  # Get the path of the temporary file
-                    listaPDFs.append(temp_file_path)
-            print("listaPDFs: ", listaPDFs)
-
-            system_prompt = data.get("system_prompt", DEFAULT_SYSTEM_PROMPT)
-            resposta_llm = get_llm_answer_summary_with_embedding(
-                system_prompt,
-                data["user_message"],
-                listaPDFs,
-                model=model,
-                isIterativeRefinement=data["iterative_refinement"],
-            )
-
-            for file in listaPDFs:
-                os.remove(file)
-
-            return Response({"resposta": resposta_llm})
-
-
-class ResumoSimplesCursorView(APIView):
-    parser_classes = [MultiPartParser]
-
-    @extend_schema(
-        request=ResumoCursorSerializer,
-    )
-    def post(self, request):
-        serializer = ResumoCursorSerializer(data=request.data)
-        if serializer.is_valid(raise_exception=True):
-            listaPDFs = []
-            data = serializer.validated_data
-            print("\nserializer.validated_data: ", serializer.validated_data)
-
-            for file in serializer.validated_data["files"]:
-                file.seek(0)
-                with tempfile.NamedTemporaryFile(
-                    delete=False, suffix=".pdf"
-                ) as temp_file:  # Create a temporary file to save the uploaded PDF
-                    for (
-                        chunk
-                    ) in (
-                        file.chunks()
-                    ):  # Write the uploaded file content to the temporary file
-                        temp_file.write(chunk)
-                    temp_file_path = (
-                        temp_file.name
-                    )  # Get the path of the temporary file
-                    listaPDFs.append(temp_file_path)
-            print("listaPDFs: ", listaPDFs)
-
-            resposta_llm = get_llm_summary_answer_by_cursor(data, listaPDFs)
-
-            for file in listaPDFs:
-                os.remove(file)
-
-            return Response({"resposta": resposta_llm})

gerar_documento/serializer.py

@@ -1,5 +1,4 @@
 from rest_framework import serializers
-from _antigos.resumos.serializer import ResumoCursorSerializer
 from _utils.gerar_relatorio_modelo_usuario.prompts import (
     prompt_gerar_documento,
     prompt_auxiliar_padrao,
@@ -9,13 +8,34 @@ from setup.environment import default_model
 user_message = "What are the main points of this document?"
 
 
+prompt_template = """
+    Based on the following context, provide multiple key points from the document.
+    For each point, create a new paragraph.
+    Each paragraph should be a complete, self-contained insight.
+    
+    Context: {context}
+    
+    Key points:
+    """
+
+
+class GerarDocumentoInitialSerializer(serializers.Serializer):
+    files = serializers.ListField(child=serializers.FileField(), required=True)
+    system_prompt = serializers.CharField(required=False, default=prompt_template)
+    user_message = serializers.CharField(required=False, default="")
+    model = serializers.CharField(required=False, default=default_model)
+    hf_embedding = serializers.CharField(required=False, default="all-MiniLM-L6-v2")
+    chunk_size = serializers.IntegerField(required=False, default=3500)
+    chunk_overlap = serializers.IntegerField(required=False, default=800)
+
+
 class FileInfoSerializer(serializers.Serializer):
     unique_id = serializers.CharField(max_length=255)
     tipo_arquivo = serializers.CharField(max_length=255)
     link_arquivo = serializers.URLField()
 
 
-class GerarDocumentoSerializer(ResumoCursorSerializer):
+class GerarDocumentoSerializer(GerarDocumentoInitialSerializer):
     system_prompt = None
 
     files = serializers.ListField(child=FileInfoSerializer(), required=True)
@@ -55,7 +75,7 @@ class GerarDocumentoSerializer(ResumoCursorSerializer):
     version = serializers.CharField(required=True)
 
 
-class GerarDocumentoComPDFProprioSerializer(ResumoCursorSerializer):
+class GerarDocumentoComPDFProprioSerializer(GerarDocumentoInitialSerializer):
     system_prompt = None
     # prompt_auxiliar = serializers.CharField(
     #     required=False, default=prompt_auxiliar_padrao

setup/urls.py

@@ -2,9 +2,6 @@ from django.contrib import admin
 from django.urls import path, include
 from rest_framework import routers
 from drf_spectacular.views import SpectacularSwaggerView, SpectacularAPIView
-from _antigos.resumos.views import (
-    ResumoSimplesCursorView,
-)  # Não sei por quê, mas se remover esta importação, o endpoint de gerar_documentos para de funcionar
 
 router = routers.DefaultRouter()
 # router.register("endpoint-teste", EndpointTesteViewSet, basename="Basename do endpoint-teste")