update prompt (#1137)

* update prompt

* test and fix. Add dockerfile.dev

* up

py/Dockerfile

@@ -48,6 +48,7 @@ ARG PORT=8000
 ARG HOST=0.0.0.0
 ENV PORT=$PORT HOST=$HOST
 EXPOSE $PORT
+ENV TESSDATA_PREFIX=/usr/share/tesseract-ocr/5/tessdata
 
 # Run the application
 CMD ["sh", "-c", "uvicorn core.main.app_entry:app --host $HOST --port $PORT"]

py/Dockerfile.dev

@@ -0,0 +1,49 @@
+FROM python:3.10-slim AS builder
+# Install system dependencies (including those needed for Unstructured and OpenCV)
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    gcc g++ musl-dev curl libffi-dev gfortran libopenblas-dev \
+    tesseract-ocr libtesseract-dev libleptonica-dev pkg-config \
+    poppler-utils libmagic1 \
+    libgl1-mesa-glx libglib2.0-0 \
+    && apt-get clean && rm -rf /var/lib/apt/lists/*
+
+RUN pip install --no-cache-dir poetry
+
+RUN mkdir -p /app/py
+WORKDIR /app/py
+COPY pyproject.toml /app/py/pyproject.toml
+
+# Install the dependencies, including gunicorn, uvicorn, and unstructured
+RUN poetry config virtualenvs.create false \
+    && poetry install --extras "core" --no-dev --no-root \
+    && pip install --no-cache-dir gunicorn uvicorn
+# Create the final image
+FROM python:3.10-slim
+# Install runtime dependencies
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends \
+    curl tesseract-ocr poppler-utils libmagic1 \
+    libgl1-mesa-glx libglib2.0-0 \
+    && apt-get clean && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+COPY --from=builder /usr/local/lib/python3.10/site-packages /usr/local/lib/python3.10/site-packages
+COPY --from=builder /usr/local/bin /usr/local/bin
+
+RUN python -c "from unstructured.partition.model_init import initialize; initialize()"
+
+
+# Expose the port and set environment variables
+ARG PORT=8000 HOST=0.0.0.0
+ENV PORT=$PORT HOST=$HOST TESSDATA_PREFIX=/usr/share/tesseract-ocr/5/tessdata
+EXPOSE $PORT
+
+COPY . /app
+# Copy the application and config
+COPY core /app/core
+COPY r2r /app/r2r
+COPY r2r.toml /app/r2r.toml
+COPY pyproject.toml /app/pyproject.toml
+
+CMD ["sh", "-c", "uvicorn core.main.app_entry:app --host $HOST --port $PORT"]

py/cli/commands/server.py

@@ -214,6 +214,12 @@ def generate_report():
     default=False,
     help="Use the unstructured Docker image",
 )
+@click.option(
+    "--dev",
+    is_flag=True,
+    default=False,
+    help="Run in development mode",
+)
 def serve(
     host,
     port,
@@ -228,6 +234,7 @@ def serve(
     config_path,
     build,
     unstructured,
+    dev,
 ):
     """Start the R2R server."""
     load_dotenv()
@@ -280,7 +287,7 @@ def serve(
                 "-t",
                 image,
                 "-f",
-                "Dockerfile",
+                f"Dockerfile{'.dev' if dev else ''}",
                 ".",
             ],
             check=True,
@@ -308,6 +315,7 @@ def serve(
             image,
             config_name,
             config_path,
+
         )
         if (
             "pytest" in sys.modules

py/core/base/abstractions/prompt.py

@@ -1,7 +1,7 @@
 """Abstraction for a prompt that can be formatted with inputs."""
 
 from typing import Any
-
+from datetime import datetime
 from pydantic import BaseModel
 
 
@@ -10,6 +10,8 @@ class Prompt(BaseModel):
 
     name: str
     template: str
+    created_at: datetime = datetime.now()
+    updated_at: datetime = datetime.now()
     input_types: dict[str, str]
 
     def format_prompt(self, inputs: dict[str, Any]) -> str:

py/core/base/api/models/management/responses.py

@@ -14,6 +14,8 @@ class UpdatePromptResponse(BaseModel):
 class PromptResponse(BaseModel):
     name: str
     template: str
+    created_at: datetime
+    updated_at: datetime
     input_types: Dict[str, str]
 
 

py/core/providers/prompts/r2r_prompts.py

@@ -3,7 +3,7 @@
 import os
 from pathlib import Path
 from typing import Any, Optional
-from uuid import uuid4
+from uuid import uuid5, NAMESPACE_DNS
 
 import yaml
 from sqlalchemy import text
@@ -55,15 +55,19 @@ def create_table(self):
     def _load_prompts_from_database(self):
         query = text(
             f"""
-            SELECT prompt_id, name, template, input_types
+            SELECT prompt_id, name, template, input_types, created_at, updated_at
             FROM {self._get_table_name('prompts')}
             """
         )
         results = self.execute_query(query).fetchall()
         for row in results:
-            prompt_id, name, template, input_types = row
+            prompt_id, name, template, input_types, created_at, updated_at = row
             self.prompts[name] = Prompt(
-                name=name, template=template, input_types=input_types
+                name=name,
+                template=template,
+                input_types=input_types,
+                created_at=created_at,
+                updated_at=updated_at,
             )
 
     def _load_prompts_from_yaml_directory(
@@ -88,11 +92,18 @@ def _load_prompts_from_yaml_directory(
                     data = yaml.safe_load(file)
                     for name, prompt_data in data.items():
                         if name not in self.prompts:
+                            modify_prompt = True
+                        else:
+                            modify_prompt = self.prompts[name].created_at == self.prompts[name].updated_at
+
+                        if modify_prompt:
                             self.add_prompt(
                                 name,
                                 prompt_data["template"],
                                 prompt_data.get("input_types", {}),
+                                modify_created_at = True,
                             )
+
             except yaml.YAMLError as e:
                 error_msg = (
                     f"Error loading prompts from YAML file {yaml_file}: {e}"
@@ -105,13 +116,11 @@ def _load_prompts_from_yaml_directory(
                 raise ValueError(error_msg)
 
     def add_prompt(
-        self, name: str, template: str, input_types: dict[str, str]
+        self, name: str, template: str, input_types: dict[str, str], modify_created_at: bool = False
     ) -> None:
-        if name in self.prompts:
-            raise ValueError(f"Prompt '{name}' already exists.")
         prompt = Prompt(name=name, template=template, input_types=input_types)
         self.prompts[name] = prompt
-        self._save_prompt_to_database(prompt)
+        self._save_prompt_to_database(prompt, modify_created_at=modify_created_at)
 
     def get_prompt(
         self,
@@ -165,7 +174,13 @@ def delete_prompt(self, name: str) -> None:
         )
         self.execute_query(query, {"name": name})
 
-    def _save_prompt_to_database(self, prompt: Prompt):
+    def _save_prompt_to_database(self, prompt: Prompt, modify_created_at: bool = False):
+
+        if modify_created_at:
+            modify_created_at_clause = "created_at = NOW(),"
+        else:
+            modify_created_at_clause = ""
+
         query = text(
             f"""
             INSERT INTO {self._get_table_name('prompts')}
@@ -174,13 +189,14 @@ def _save_prompt_to_database(self, prompt: Prompt):
             ON CONFLICT (name) DO UPDATE SET
                 template = EXCLUDED.template,
                 input_types = EXCLUDED.input_types,
+                {modify_created_at_clause}
                 updated_at = NOW();
             """
         )
         result = self.execute_query(
             query,
             {
-                "prompt_id": uuid4(),
+                "prompt_id": uuid5(NAMESPACE_DNS, prompt.name),
                 "name": prompt.name,
                 "template": prompt.template,
                 "input_types": json.dumps(prompt.input_types),