fix docker builds (#963)

py/Dockerfile

@@ -17,11 +17,15 @@ COPY . /app
 
 # Ensure that the working directory is set to /app/py
 WORKDIR /app/py
+COPY scripts /app/py/scripts
 
 # Install dependencies
 RUN poetry config virtualenvs.create false \
     && poetry install --extras "core" --no-dev --no-root \
-    && pip install --no-cache-dir gunicorn uvicorn
+    && pip install --no-cache-dir gunicorn uvicorn \
+    && poetry run python scripts/download_nltk_data.py
+
+
 
 # Final stage to keep the image small
 FROM python:3.10-slim
@@ -36,6 +40,7 @@ WORKDIR /app
 # Copy the installed packages from the builder
 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
+COPY --from=builder /root/nltk_data /root/nltk_data
 
 # Copy the necessary application files
 COPY core /app/core

py/Dockerfile.unstructured

@@ -12,6 +12,7 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
 ENV PATH="/root/.cargo/bin:${PATH}"
 
 WORKDIR /app
+COPY scripts /app/py/scripts
 
 RUN pip install --no-cache-dir poetry
 
@@ -23,7 +24,8 @@ RUN poetry config virtualenvs.create false \
     && poetry install --extras "core" --no-dev --no-root \
     && poetry add "unstructured[all-docs]" \
     && poetry add "unstructured_client" \
-    && pip install --no-cache-dir gunicorn uvicorn
+    && pip install --no-cache-dir gunicorn uvicorn \
+    && poetry run python scripts/download_nltk_data.py
 
 # Create the final image
 FROM python:3.10-slim
@@ -40,6 +42,7 @@ WORKDIR /app/py
 # Copy the installed packages from the builder
 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
+COPY --from=builder /root/nltk_data /root/nltk_data
 
 # Copy the application and config
 COPY py/core /app/py/core

py/core/providers/database/group.py

@@ -17,7 +17,7 @@ class GroupMixin(DatabaseMixin):
     def create_table(self) -> None:
         query = f"""
         CREATE TABLE IF NOT EXISTS {self._get_table_name('groups')} (
-            group_id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
+            group_id UUID PRIMARY KEY DEFAULT public.uuid_generate_v4(),
             name TEXT NOT NULL,
             description TEXT,
             created_at TIMESTAMPTZ DEFAULT NOW(),

py/core/providers/database/relational.py

@@ -30,14 +30,12 @@ def execute_query(self, query, params=None):
     def _initialize_relational_db(self):
         with self.vx.Session() as sess:
             with sess.begin():
-                try:
-                    sess.execute(
-                        text('CREATE EXTENSION IF NOT EXISTS "uuid-ossp";')
-                    )
-                except exc.ProgrammingError as e:
-                    logger.error(f"Error enabling uuid-ossp extension: {e}")
-                    raise
+                sess.execute(
+                    text('CREATE EXTENSION IF NOT EXISTS "uuid-ossp";')
+                )
+                sess.commit()
 
+            with sess.begin():
                 # Call create_table for each mixin
                 for base_class in self.__class__.__bases__:
                     if issubclass(base_class, DatabaseMixin):

py/core/providers/database/tokens.py

@@ -8,7 +8,7 @@ class BlacklistedTokensMixin(DatabaseMixin):
     def create_table(self):
         query = f"""
         CREATE TABLE IF NOT EXISTS {self._get_table_name('blacklisted_tokens')} (
-            id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
+            id UUID PRIMARY KEY DEFAULT public.uuid_generate_v4(),
             token TEXT NOT NULL,
             blacklisted_at TIMESTAMPTZ DEFAULT NOW()
         );

py/core/providers/database/user.py

@@ -15,7 +15,7 @@ class UserMixin(DatabaseMixin):
     def create_table(self):
         query = f"""
         CREATE TABLE IF NOT EXISTS {self._get_table_name('users')} (
-            user_id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
+            user_id UUID PRIMARY KEY DEFAULT public.uuid_generate_v4(),
             email TEXT UNIQUE NOT NULL,
             hashed_password TEXT NOT NULL,
             is_superuser BOOLEAN DEFAULT FALSE,

py/r2r/__init__.py

@@ -13,4 +13,4 @@
     __all__ += core.__all__
 except ImportError:
     # Core dependencies not installed
-    pass
+    pass