From 0c0cfd3cf6dd7e2c8c94d4f9bd6b765ea51077aa Mon Sep 17 00:00:00 2001
From: Ioannis Paraskevakos <iparask@princeton.edu>
Date: Mon, 22 Jun 2026 14:39:25 -0400
Subject: [PATCH 1/3] feat: setting up the API

---
 .gitignore                |  2 +-
 .pre-commit-config.yaml   | 36 ++++++++++++++++
 pyproject.toml            | 91 ++++++++++++++++++++++++++++++++++++---
 src/archivist/__init__.py |  6 +--
 src/archivist/__main__.py | 61 ++++++++++++++++++++++++++
 src/archivist/api.py      | 16 +++++++
 src/archivist/archive.py  |  2 +-
 src/archivist/config.py   |  3 ++
 src/archivist/core.py     | 71 ++++++++++++++++++++++++++++++
 src/archivist/registry.py |  5 ++-
 src/archivist/storage.py  |  9 ++--
 tests/test_classes.py     | 12 +-----
 12 files changed, 287 insertions(+), 27 deletions(-)
 create mode 100644 .pre-commit-config.yaml
 create mode 100644 src/archivist/__main__.py
 create mode 100644 src/archivist/api.py
 create mode 100644 src/archivist/config.py
 create mode 100644 src/archivist/core.py

diff --git a/.gitignore b/.gitignore
index ecd41cb..15779c2 100644
--- a/.gitignore
+++ b/.gitignore
@@ -101,7 +101,7 @@ ipython_config.py
 #   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
 #   This is especially recommended for binary packages to ensure reproducibility, and is more
 #   commonly ignored for libraries.
-#uv.lock
+uv.lock
 
 # poetry
 #   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
new file mode 100644
index 0000000..b35a7a5
--- /dev/null
+++ b/.pre-commit-config.yaml
@@ -0,0 +1,36 @@
+repos:
+  - repo: https://github.com/pycqa/isort
+    rev: 6.0.1
+    hooks:
+      - id: isort
+
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.15.17
+    hooks:
+      - id: ruff-check
+      - id: ruff-format
+        args: [--check, --diff]
+
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v5.0.0
+    hooks:
+      - id: check-yaml
+      - id: end-of-file-fixer
+      - id: trailing-whitespace
+      - id: check-added-large-files
+      - id: check-merge-conflict
+      - id: debug-statements
+
+# Replace commitsar with conventional commit checker
+  - repo: https://github.com/compilerla/conventional-pre-commit
+    rev: v2.1.1
+    hooks:
+      - id: conventional-pre-commit
+        stages: [commit-msg]
+        args: [] # default: [feat, fix, ci, chore, docs, style, refactor, perf, test, build, revert]
+
+  - repo: https://github.com/pre-commit/pygrep-hooks
+    rev: v1.10.0
+    hooks:
+    - id: python-check-blanket-noqa
+    - id: python-no-eval
diff --git a/pyproject.toml b/pyproject.toml
index 60115e7..80d6890 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -13,11 +13,11 @@ classifiers = [
     "License :: OSI Approved :: MIT License",
     "Operating System :: OS Independent",
 ]
-dependencies = ["numpy"]
-
-[project.optional-dependencies]
-dev = [
-    "pytest"
+dependencies = [
+    "click>=8.1.8",
+    "fastapi>=0.128.8",
+    "numpy",
+    "pydantic>=2.13.4",
 ]
 
 [project.urls]
@@ -25,3 +25,84 @@ Homepage = "https://github.com/simonsobs/archivist"
 
 [tool.setuptools_scm]
 write_to = "src/archivist/_version.py"
+
+[dependency-groups]
+dev = [
+    "coverage>=7.10.7",
+    "coveralls>=4.0.1",
+    "isort>=6.1.0",
+    "pre-commit>=4.3.0",
+    "pytest>=8.4.2",
+    "pytest-cov>=7.1.0",
+    "pytest-random-order>=1.2.0",
+    "ruff>=0.15.17",
+]
+
+[project.scripts]
+archivist = "archivist.__main__:main"
+
+[tool.ruff]
+line-length = 120
+
+[tool.ruff.lint]
+ignore = ["ANN001",
+          "SIM118",
+          "T201",
+          "D100",
+          "INP001",
+          "D212",
+          "PTH119",
+          "D415",
+          "D103",
+          "D411",
+          "D410",
+          "COM812",
+          "D205",
+          "S101",
+          "D202",
+          "SLF001",
+          "D401",
+          "D400",
+          "FBT001",
+          "ANN202",
+          "PTH123",
+          "FBT002",
+          "ANN201",
+          "F401",
+          "F841"
+]
+
+[tool.ruff.format]
+docstring-code-format = true
+docstring-code-line-length = 80
+
+[tool.ruff.lint.per-file-ignores]
+"**/scripts/*" = [
+  "INP001",
+  "T201",
+]
+"**/tests/**/*" = [
+  "PLC1901",
+  "PLR2004",
+  "PLR6301",
+  "S",
+  "TID252",
+  "T201",
+  "DTZ005",
+  "N802",
+  "ARG005"
+]
+
+[tool.ruff.lint.flake8-tidy-imports]
+ban-relative-imports = "all"
+
+[tool.ruff.lint.isort]
+known-first-party = ["archivist"]
+
+[tool.ruff.lint.flake8-pytest-style]
+fixture-parentheses = false
+mark-parentheses = false
+
+[tool.isort]
+line_length = 120
+skip = ["_version.py"]
diff --git a/src/archivist/__init__.py b/src/archivist/__init__.py
index c5e352d..3fc3ad7 100644
--- a/src/archivist/__init__.py
+++ b/src/archivist/__init__.py
@@ -1,11 +1,9 @@
 # Copyright (c) 2025-2026 Simons Observatory.
 # Full license can be found in the top level "LICENSE" file.
 
-"""Archive tools for data.
-"""
+"""Archive tools for data."""
 
 from ._version import __version__
-
 from .archive import Archive
-from .registry import RegistrySqlite, RegistryLibrarian
+from .registry import RegistryLibrarian, RegistrySqlite
 from .storage import StorageDisk, StorageHPSS
diff --git a/src/archivist/__main__.py b/src/archivist/__main__.py
new file mode 100644
index 0000000..0b84386
--- /dev/null
+++ b/src/archivist/__main__.py
@@ -0,0 +1,61 @@
+import sys
+
+import click
+
+import Archivist
+
+
+@click.group()
+@click.option(
+    "--config",
+    "-c",
+    type=click.Path(exists=True),
+    required=False,
+    help="Path to the Archivist configuration file",
+)
+@click.pass_context
+def main(ctx, config):
+    if config is None:
+        click.echo("Archivist requires a configuration file", err=True)
+        click.echo("See readme for more information", err=True)
+        sys.exit(1)
+    ctx.ensure_object(dict)
+    ctx.obj["archivist"] = Archivist(config)
+
+
+@main.command()
+@click.option("--source-path", type=str, help="The source path of the file to archive")
+@click.option(
+    "--dest-path",
+    type=str,
+    help="The destination path where the file should be archived",
+)
+@click.pass_context
+def archive(ctx, source_path, dest_path):
+    """Command to archive a file.
+    For example: `archivist archive "--source-path /path/to/source --dest-path /path/to/dest"`
+    """
+    pass
+
+
+@main.command()
+@click.argument("cmd", nargs=1)
+@click.option("--archive", type=str, help="Name of the job")
+@click.option(
+    "--dest-path",
+    type=str,
+    help="The destination path where the file should be restored",
+)
+@click.pass_context
+def extract(ctx, archive, dest_path):
+    """Command to record a job.
+    For example: `slurmise record "-o 2 -i 3 -m fast"`
+    """
+    pass
+
+
+@main.command()
+@click.pass_context
+def start_server(ctx):
+    """Command to check the status of the archivist."""
+    pass
diff --git a/src/archivist/api.py b/src/archivist/api.py
new file mode 100644
index 0000000..73c42d8
--- /dev/null
+++ b/src/archivist/api.py
@@ -0,0 +1,16 @@
+class Archivist:
+    def __init__(self, config):
+        self.config = config
+        self.archivist = None
+
+    def start_server(self):
+        # Initialize the archivist with the provided configuration
+        self.archivist = self.config.get("archivist", {})
+
+    def archive(self, source_path, dest_path):
+        # Implementation for archiving a file
+        pass
+
+    def extract(self, archive, dest_path):
+        # Implementation for extracting an archive
+        pass
diff --git a/src/archivist/archive.py b/src/archivist/archive.py
index 1fe5b1e..a9abcd0 100644
--- a/src/archivist/archive.py
+++ b/src/archivist/archive.py
@@ -20,6 +20,7 @@ class Archive(object):
             for all objects.
 
     """
+
     def __init__(self, manifest=None, paths=None, root=None):
         pass
 
@@ -34,4 +35,3 @@ def root(self):
     @property
     def checksum(self):
         pass
-
diff --git a/src/archivist/config.py b/src/archivist/config.py
new file mode 100644
index 0000000..dfe602f
--- /dev/null
+++ b/src/archivist/config.py
@@ -0,0 +1,3 @@
+class ArchivistConfiguration:
+    def __init__(self, config):
+        self.config = config
diff --git a/src/archivist/core.py b/src/archivist/core.py
new file mode 100644
index 0000000..75adc63
--- /dev/null
+++ b/src/archivist/core.py
@@ -0,0 +1,71 @@
+"""
+Pydantic modems for the admin endpoints
+"""
+
+from datetime import datetime
+
+from pydantic import BaseModel
+
+
+class ManifestEntry(BaseModel):
+    name: str
+    "The name of the file."
+    create_time: datetime
+    "The time the file was created."
+    size: int
+    "The size of the file in bytes."
+    checksum: str
+    "The checksum of the file."
+    uploader: str
+    "The uploader of the file."
+    source: str
+    "The source of the file."
+
+    instance_path: str
+    "The path to the instance on the store."
+    instance_create_time: datetime
+    "The time the instance was created."
+    instance_available: bool
+    "Whether the instance is available or not. If not, no outgoing transfer is created."
+
+    outgoing_transfer_id: int
+    "The ID of the outgoing transfer, if it exists."
+
+
+class ManifestRequest(BaseModel):
+    store_name: str
+    "The name of the store to get the manifest for."
+
+    create_outgoing_transfers: bool = False
+    "Whether to create outgoing transfers for the files in the manifest."
+
+    destination_librarian: str = ""
+    "The name of the librarian to send the files to, if create_outgoing_transfers is true."
+
+    disable_store: bool = False
+    "Whether to disable the store after creating the outgoing transfers."
+
+    mark_local_instances_as_unavailable: bool = False
+    "Mark the local instances as unavailable after creating the outgoing transfers."
+
+
+class ManifestResponse(BaseModel):
+    librarian_name: str
+    "The name of the librarian that generated this manifest."
+
+    store_name: str
+    "The name of the store."
+
+    store_files: list[ManifestEntry]
+    "The files on the store."
+
+
+class Archive(BaseModel):
+    manifest: str
+    "Path to a manifest file listing the files / directories contained in the archive."
+
+    paths: list[str]
+    "Alternatively list the full set of files / directories contained in the archive."
+
+    root: str
+    "The common root filesystem location for building relative paths for all objects."
diff --git a/src/archivist/registry.py b/src/archivist/registry.py
index 52ba84c..a85569b 100644
--- a/src/archivist/registry.py
+++ b/src/archivist/registry.py
@@ -4,9 +4,8 @@
 
 
 class Registry(object):
-    """Base class representing a registry of archives.
+    """Base class representing a registry of archives."""
 
-    """
     def __init__(self):
         pass
 
@@ -44,6 +43,7 @@ def remove(self, archive_name):
 
 class RegistrySqlite(Registry):
     """Simple class which registers archives in an SQLite DB."""
+
     def __init__(self):
         super().__init__()
 
@@ -56,6 +56,7 @@ def _remove(self, archive_name):
 
 class RegistryLibrarian(Registry):
     """Store archive metadata with with the librarian."""
+
     def __init__(self):
         super().__init__()
 
diff --git a/src/archivist/storage.py b/src/archivist/storage.py
index e48bba4..c4f24e4 100644
--- a/src/archivist/storage.py
+++ b/src/archivist/storage.py
@@ -4,8 +4,8 @@
 
 
 class Storage(object):
-    """Base class representing an archive storage system.
-    """
+    """Base class representing an archive storage system."""
+
     def __init__(self):
         pass
 
@@ -61,6 +61,7 @@ class StorageDisk(Storage):
         directory (str):  The top-level location for storing archives.
 
     """
+
     def __init__(self, directory):
         self._directory = directory
         super().__init__()
@@ -75,8 +76,8 @@ def _extract(self, archive_name, storage_info, outdir, paths=None):
 
 
 class StorageHPSS(Storage):
-    """Store archives to HPSS tapes.
-    """
+    """Store archives to HPSS tapes."""
+
     def __init__(self):
         super().__init__()
 
diff --git a/tests/test_classes.py b/tests/test_classes.py
index b24d1a6..2db294e 100644
--- a/tests/test_classes.py
+++ b/tests/test_classes.py
@@ -4,15 +4,8 @@
 """Test package imports."""
 
 import archivist
-from archivist import __version__ as pkg_version
-
-from archivist import (
-    Archive,
-    RegistrySqlite,
-    RegistryLibrarian,
-    StorageDisk,
-    StorageHPSS,
-)
+from archivist import Archive, RegistryLibrarian, RegistrySqlite, StorageDisk, StorageHPSS
+from archivist import _version as pkg_version
 
 
 def test_construction():
@@ -21,4 +14,3 @@ def test_construction():
     reg_sql = RegistrySqlite()
     store_disk = StorageDisk(".")
     store_hpss = StorageHPSS()
-

From 05bec5b46a3f237fec0cb0e0132b74d5fd4dd260 Mon Sep 17 00:00:00 2001
From: Ioannis Paraskevakos <iparask@princeton.edu>
Date: Mon, 22 Jun 2026 19:24:15 -0400
Subject: [PATCH 2/3] feat: creating the archive router

---
 src/archivist/__init__.py                 |  6 +--
 src/archivist/api/__init__.py             |  3 ++
 src/archivist/api/archive.py              | 47 +++++++++++++++++++++++
 src/archivist/core/__init__.py            |  0
 src/archivist/{ => core}/archive.py       |  0
 src/archivist/{core.py => core/models.py} | 31 ++++++---------
 src/archivist/{ => core}/registry.py      |  0
 src/archivist/{ => core}/storage.py       |  0
 8 files changed, 65 insertions(+), 22 deletions(-)
 create mode 100644 src/archivist/api/__init__.py
 create mode 100644 src/archivist/api/archive.py
 create mode 100644 src/archivist/core/__init__.py
 rename src/archivist/{ => core}/archive.py (100%)
 rename src/archivist/{core.py => core/models.py} (69%)
 rename src/archivist/{ => core}/registry.py (100%)
 rename src/archivist/{ => core}/storage.py (100%)

diff --git a/src/archivist/__init__.py b/src/archivist/__init__.py
index 3fc3ad7..9e2db2a 100644
--- a/src/archivist/__init__.py
+++ b/src/archivist/__init__.py
@@ -4,6 +4,6 @@
 """Archive tools for data."""
 
 from ._version import __version__
-from .archive import Archive
-from .registry import RegistryLibrarian, RegistrySqlite
-from .storage import StorageDisk, StorageHPSS
+from .core.archive import Archive
+from .core.registry import RegistryLibrarian, RegistrySqlite
+from .core.storage import StorageDisk, StorageHPSS
diff --git a/src/archivist/api/__init__.py b/src/archivist/api/__init__.py
new file mode 100644
index 0000000..91129c5
--- /dev/null
+++ b/src/archivist/api/__init__.py
@@ -0,0 +1,3 @@
+from fastapi import APIRouter
+
+router = APIRouter(prefix="/api/v1")
diff --git a/src/archivist/api/archive.py b/src/archivist/api/archive.py
new file mode 100644
index 0000000..3ac800d
--- /dev/null
+++ b/src/archivist/api/archive.py
@@ -0,0 +1,47 @@
+import json
+import uuid
+
+from fastapi import Depends, Response
+
+from archivist.api import router
+from archivist.core.models import Archive, ManifestFailedResponse, ManifestRequest, ManifestResponse
+
+
+@router.post("/archive", response_model=ManifestResponse | ManifestFailedResponse)
+def archive(
+    manifest_request: ManifestRequest,
+    response: Response,
+    queue: "Queue" = Depends(get_queue),  # noqa: F821
+    settings: "Settings" = Depends(get_settings),  # noqa: F821
+):
+    """
+    Endpoint to archive a file.
+    """
+    # Here you would implement the logic to handle the archiving process
+    # For now, we will just return a dummy response
+
+    # TODO: Implement the actual archiving logic here.
+
+    # TODO: get the basic settings to know the type of archival store and the size of the
+    #      archive manifest
+
+    total_size = sum(entry.size for entry in manifest_request.store_files)
+    if total_size > 1000000000:  # Example size limit of 1GB
+        response.status_code = 400
+        return ManifestFailedResponse(error="The total size of the files exceeds the allowed limit.")
+
+    manifest_id = uuid.uuid4()
+
+    archive = Archive(
+        manifest_id=str(manifest_id),
+        manifest=json.dumps(manifest_request),
+        paths=[entry.instance_path for entry in manifest_request.store_files],
+        root=settings.archive_root,  # Example root path
+    )
+
+    # TODO: Add the archive to a queue for processing
+    queue.enqueue_archive(archive)
+
+    return ManifestResponse(
+        manifest_id=str(manifest_id),
+    )
diff --git a/src/archivist/core/__init__.py b/src/archivist/core/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/src/archivist/archive.py b/src/archivist/core/archive.py
similarity index 100%
rename from src/archivist/archive.py
rename to src/archivist/core/archive.py
diff --git a/src/archivist/core.py b/src/archivist/core/models.py
similarity index 69%
rename from src/archivist/core.py
rename to src/archivist/core/models.py
index 75adc63..c8c3cd9 100644
--- a/src/archivist/core.py
+++ b/src/archivist/core/models.py
@@ -33,34 +33,27 @@ class ManifestEntry(BaseModel):
 
 
 class ManifestRequest(BaseModel):
-    store_name: str
-    "The name of the store to get the manifest for."
-
-    create_outgoing_transfers: bool = False
-    "Whether to create outgoing transfers for the files in the manifest."
-
-    destination_librarian: str = ""
-    "The name of the librarian to send the files to, if create_outgoing_transfers is true."
-
-    disable_store: bool = False
-    "Whether to disable the store after creating the outgoing transfers."
+    librarian_name: str
+    "The name of the librarian that generated this manifest."
 
-    mark_local_instances_as_unavailable: bool = False
-    "Mark the local instances as unavailable after creating the outgoing transfers."
+    store_files: list[ManifestEntry]
+    "The files on the store."
 
 
 class ManifestResponse(BaseModel):
-    librarian_name: str
-    "The name of the librarian that generated this manifest."
+    manifest_id: str
+    "The ID of the manifest."
 
-    store_name: str
-    "The name of the store."
 
-    store_files: list[ManifestEntry]
-    "The files on the store."
+class ManifestFailedResponse(BaseModel):
+    error: str
+    "The error message indicating why the manifest request failed."
 
 
 class Archive(BaseModel):
+    manifest_id: str
+    "The ID of the manifest associated with this archive."
+
     manifest: str
     "Path to a manifest file listing the files / directories contained in the archive."
 
diff --git a/src/archivist/registry.py b/src/archivist/core/registry.py
similarity index 100%
rename from src/archivist/registry.py
rename to src/archivist/core/registry.py
diff --git a/src/archivist/storage.py b/src/archivist/core/storage.py
similarity index 100%
rename from src/archivist/storage.py
rename to src/archivist/core/storage.py

From 870645d9886d4557b3502dc3248a09717abf6f52 Mon Sep 17 00:00:00 2001
From: Ioannis Paraskevakos <iparask@princeton.edu>
Date: Tue, 30 Jun 2026 14:29:10 -0400
Subject: [PATCH 3/3] feat:Archivist API

---
 pyproject.toml                 |  10 ++-
 src/archivist/api/archive.py   |  15 ++--
 src/archivist/api/extract.py   |  25 ++++++
 src/archivist/core/__init__.py |   3 +
 src/archivist/core/archive.py  |  27 ++++--
 src/archivist/queue.py         |  60 +++++++++++++
 src/archivist/settings.py      | 152 +++++++++++++++++++++++++++++++++
 7 files changed, 275 insertions(+), 17 deletions(-)
 create mode 100644 src/archivist/api/extract.py
 create mode 100644 src/archivist/queue.py
 create mode 100644 src/archivist/settings.py

diff --git a/pyproject.toml b/pyproject.toml
index 80d6890..4236756 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -14,10 +14,14 @@ classifiers = [
     "Operating System :: OS Independent",
 ]
 dependencies = [
-    "click>=8.1.8",
-    "fastapi>=0.128.8",
+    "click",
+    "fastapi",
     "numpy",
-    "pydantic>=2.13.4",
+    "pydantic",
+    "pydantic-settings",
+    "sqlalchemy",
+    "sqlalchemy-utils",
+    "loguru",
 ]
 
 [project.urls]
diff --git a/src/archivist/api/archive.py b/src/archivist/api/archive.py
index 3ac800d..16d09e4 100644
--- a/src/archivist/api/archive.py
+++ b/src/archivist/api/archive.py
@@ -5,14 +5,16 @@
 
 from archivist.api import router
 from archivist.core.models import Archive, ManifestFailedResponse, ManifestRequest, ManifestResponse
+from archivist.queue import ArchiveQueue, get_queue
+from archivist.settings import Settings, get_settings
 
 
 @router.post("/archive", response_model=ManifestResponse | ManifestFailedResponse)
 def archive(
     manifest_request: ManifestRequest,
     response: Response,
-    queue: "Queue" = Depends(get_queue),  # noqa: F821
-    settings: "Settings" = Depends(get_settings),  # noqa: F821
+    queue: ArchiveQueue = Depends(get_queue),
+    settings: Settings = Depends(get_settings),
 ):
     """
     Endpoint to archive a file.
@@ -20,13 +22,8 @@ def archive(
     # Here you would implement the logic to handle the archiving process
     # For now, we will just return a dummy response
 
-    # TODO: Implement the actual archiving logic here.
-
-    # TODO: get the basic settings to know the type of archival store and the size of the
-    #      archive manifest
-
     total_size = sum(entry.size for entry in manifest_request.store_files)
-    if total_size > 1000000000:  # Example size limit of 1GB
+    if total_size > settings.maximal_size_bytes:  # Example size limit of 1GB
         response.status_code = 400
         return ManifestFailedResponse(error="The total size of the files exceeds the allowed limit.")
 
@@ -36,7 +33,7 @@ def archive(
         manifest_id=str(manifest_id),
         manifest=json.dumps(manifest_request),
         paths=[entry.instance_path for entry in manifest_request.store_files],
-        root=settings.archive_root,  # Example root path
+        root=settings.storage_root,  # Example root path
     )
 
     # TODO: Add the archive to a queue for processing
diff --git a/src/archivist/api/extract.py b/src/archivist/api/extract.py
new file mode 100644
index 0000000..a203e89
--- /dev/null
+++ b/src/archivist/api/extract.py
@@ -0,0 +1,25 @@
+import json
+import uuid
+
+from fastapi import Depends, Response
+
+from archivist.api import router
+from archivist.core.models import Archive, ManifestFailedResponse, ManifestRequest, ManifestResponse
+from archivist.queue import ExtractQueue, get_extract_queue
+from archivist.settings import Settings, get_settings
+
+
+@router.post("/extract", response_model=ManifestResponse | ManifestFailedResponse)
+def extract(
+    manifest_request: ManifestRequest,
+    response: Response,
+    queue: ExtractQueue = Depends(get_extract_queue),
+    settings: Settings = Depends(get_settings),
+):
+    """
+    Endpoint to extract a file.
+    """
+    # Here you would implement the logic to handle the extraction process
+    # For now, we will just return a dummy response
+
+    pass
diff --git a/src/archivist/core/__init__.py b/src/archivist/core/__init__.py
index e69de29..ee43863 100644
--- a/src/archivist/core/__init__.py
+++ b/src/archivist/core/__init__.py
@@ -0,0 +1,3 @@
+from .storage import StorageDisk, StorageHPSS
+
+storage_factory = {"hpss:": StorageHPSS, "posix": StorageDisk}
diff --git a/src/archivist/core/archive.py b/src/archivist/core/archive.py
index a9abcd0..b8c9198 100644
--- a/src/archivist/core/archive.py
+++ b/src/archivist/core/archive.py
@@ -2,6 +2,8 @@
 # Full license can be found in the top level "LICENSE" file.
 """Classes for working with archives."""
 
+import subprocess
+
 
 class Archive(object):
     """Class representing a single archive in a storage system.
@@ -21,17 +23,32 @@ class Archive(object):
 
     """
 
-    def __init__(self, manifest=None, paths=None, root=None):
-        pass
+    def __init__(self, manifest=None, paths=None, root=None, type=None):
+        self._manifest = manifest
+        self._paths = paths
+        self._root = root
+        self._type = type
+        self._checksum = None  # Placeholder for checksum value
 
     @property
     def paths(self):
-        pass
+        return self._paths
 
     @property
     def root(self):
-        pass
+        return self._root
 
     @property
     def checksum(self):
-        pass
+        """Return the checksum of the archive."""
+
+    def create_archive(self):
+        """Create the archive based on the manifest or paths."""
+        # Placeholder for archive creation logic
+        if self._type == "posix":
+            # Implement logic for creating a POSIX archive
+            pass
+        elif self._type == "hpss":
+            # Implement logic for creating an HPSS archive
+            # Call subprocess to create the archive using HPSS commands
+            pass
diff --git a/src/archivist/queue.py b/src/archivist/queue.py
new file mode 100644
index 0000000..f6ed0e3
--- /dev/null
+++ b/src/archivist/queue.py
@@ -0,0 +1,60 @@
+import queue
+
+from archivist.core.models import Archive
+
+_archive_queue: "ArchiveQueue | None" = None
+_extract_queue: "ExtractQueue | None" = None
+
+
+class ArchiveQueue:
+    """Thread-safe queue for Archive jobs awaiting processing."""
+
+    def __init__(self) -> None:
+        self._q: queue.Queue[Archive] = queue.Queue()
+
+    def enqueue_archive(self, archive: Archive) -> None:
+        self._q.put(archive)
+
+    def dequeue_archive(self, block: bool = True, timeout: float | None = None) -> Archive:
+        return self._q.get(block=block, timeout=timeout)
+
+    def task_done(self) -> None:
+        self._q.task_done()
+
+    @property
+    def size(self) -> int:
+        return self._q.qsize()
+
+
+class ExtractQueue:
+    """Thread-safe queue for Extract jobs awaiting processing."""
+
+    def __init__(self) -> None:
+        self._q: queue.Queue[Archive] = queue.Queue()
+
+    def enqueue_archive(self, archive: Archive) -> None:
+        self._q.put(archive)
+
+    def dequeue_archive(self, block: bool = True, timeout: float | None = None) -> Archive:
+        return self._q.get(block=block, timeout=timeout)
+
+    def task_done(self) -> None:
+        self._q.task_done()
+
+    @property
+    def size(self) -> int:
+        return self._q.qsize()
+
+
+def get_archive_queue() -> ArchiveQueue:
+    global _archive_queue
+    if _archive_queue is None:
+        _archive_queue = ArchiveQueue()
+    return _archive_queue
+
+
+def get_extract_queue() -> ExtractQueue:
+    global _extract_queue
+    if _extract_queue is None:
+        _extract_queue = ExtractQueue()
+    return _extract_queue
diff --git a/src/archivist/settings.py b/src/archivist/settings.py
new file mode 100644
index 0000000..fb2eb5f
--- /dev/null
+++ b/src/archivist/settings.py
@@ -0,0 +1,152 @@
+import datetime
+import os
+from pathlib import Path
+
+from pydantic import BaseModel, Field, ValidationError, field_validator
+from pydantic_settings import BaseSettings, SettingsConfigDict
+from sqlalchemy import URL
+
+
+class Settings(BaseSettings):
+    """
+    Settings for the archivist . Note that because this is a BaseSettings
+    object, you can overwrite the values in the config file with environment
+    variables.
+    """
+
+    # Top level name of the server. Should be unique.
+    name: str = "archivist_server"
+
+    # Whether to enable debugging features, like the API docs and OpenAPI schema.
+    debug: bool = False
+
+    # Database settings.
+    database_driver: str = "sqlite"
+    database_user: str | None = None
+    database_password: str | None = None
+    database_host: str | None = None
+    database_port: int | None = None
+    database: str | None = None
+
+    database_password_file: Path | None = None
+
+    log_level: str = "DEBUG"
+
+    # Display name and description of the site, used in UI only.
+    displayed_site_name: str = "Untitled Archivist Server"
+    displayed_site_description: str = "No description set."
+
+    # Host and port to bind to.
+    host: str = "0.0.0.0"
+    port: int = 8080
+
+    # storage options
+    storage_type: str = "posix"
+    storage_root: str = "/tmp/archivist_storage"
+
+    # Database migration settings
+    alembic_config_path: str = "."
+    alembic_path: str = "alembic"
+
+    # Client restraints
+    max_search_results: int = 64
+    maximal_size_bytes: int = 1_000_000_000_000  # 1 TB
+
+    # Globus integration; by default disable this. This contains a client ID and
+    # login secret (for authenticating with Globus as a service), whether this
+    # client is a "native app" or not, as well as the UUID for the local Globus
+    # endpoint. That endpoint ID should be duplicated in any async transfer manager
+    # definitions we want to use a source. We also save the "local root" for the Globus
+    # endpoint, which is the root directory presented to the Globus endpoint and
+    # all file transfers using Globus are relative to. For example, if the
+    # endpoint is configured so that `/mnt/globus` is the local root, then the
+    # path to `/mnt/globus/file.txt` should be given as `/~/file.txt`.  This is
+    # important for forming paths that Globus can understand.
+    globus_enable: bool = False
+    globus_client_id: str = ""
+    globus_client_native_app: bool = False
+    globus_local_endpoint_id: str = ""
+    globus_local_root: str | None = None
+    globus_encrypt_transfers: bool = True
+
+    globus_client_secret: str | None = None
+    globus_client_secret_file: Path | None = None
+
+    # Checksumming options
+    checksum_threads: int = 4
+    checksum_timeout: datetime.timedelta = datetime.timedelta(days=1)
+
+    def model_post_init(__context, *args, **kwargs):
+        """
+        Read sensitive data from their appropriate files.
+        """
+
+        if __context.globus_client_secret_file is not None:
+            with open(__context.globus_client_secret_file, "r") as handle:
+                __context.globus_client_secret = handle.read().strip()
+
+        if __context.database_password_file is not None:
+            with open(__context.database_password_file, "r") as handle:
+                __context.database_password = handle.read().strip()
+
+    @property
+    def sqlalchemy_database_uri(self) -> URL:
+        """
+        The SQLAlchemy database URI.
+        """
+
+        return URL.create(
+            self.database_driver,
+            username=self.database_user,
+            password=self.database_password,
+            host=self.database_host,
+            port=self.database_port,
+            database=self.database,
+        )
+
+    @classmethod
+    def from_file(cls, config_path: Path | str) -> "Settings":
+        """
+        Loads the settings from the given path.
+        """
+
+        with open(config_path, "r") as handle:
+            return cls.model_validate_json(handle.read())
+
+
+# Automatically create a variable, server_settings, from the environment variable
+# on _use_!
+_settings = None
+
+
+def get_settings() -> "Settings":
+    """
+    Dependency to get the application settings.
+    """
+    global _settings
+
+    try_paths = [
+        os.environ.get("LIBRARIAN_CONFIG_PATH", None),
+    ]
+
+    for path in try_paths:
+        if path is not None:
+            path = Path(path)
+        else:
+            continue
+
+        if path.exists():
+            try:
+                _settings = Settings.from_file(path)
+            except ValidationError as e:
+                print(f"Error loading settings from {path}: {e}")
+                raise e
+
+            return _settings
+    try:
+        _settings = Settings()
+    except ValidationError as e:
+        print(f"Not all settings have defaults: {e}")
+        raise e
+
+    return _settings