Get the versions of the namespaces/extensions cached in an NWB file or used by PyNWB

get_namespace_versions.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "bd63f12d-68f1-4792-9724-ebfb843bba56",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/rly/Downloads/20230622_155936 (1).nwb\n",
      "Namespaces cached in this file:\n",
      "\t core 2.5.0\n",
      "\t hdmf-common 1.6.0\n",
      "\t hdmf-experimental 0.3.0\n",
      "\t ndx-franklab-novela 0.1.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rly/Documents/NWB/hdmf/src/hdmf/spec/namespace.py:531: UserWarning: Ignoring cached namespace 'hdmf-common' version 1.6.0 because version 1.8.0 is already loaded.\n",
      "  warn(\"Ignoring cached namespace '%s' version %s because version %s is already loaded.\"\n",
      "/Users/rly/Documents/NWB/hdmf/src/hdmf/spec/namespace.py:531: UserWarning: Ignoring cached namespace 'core' version 2.5.0 because version 2.6.0-alpha is already loaded.\n",
      "  warn(\"Ignoring cached namespace '%s' version %s because version %s is already loaded.\"\n",
      "/Users/rly/Documents/NWB/hdmf/src/hdmf/spec/namespace.py:531: UserWarning: Ignoring cached namespace 'hdmf-experimental' version 0.3.0 because version 0.5.0 is already loaded.\n",
      "  warn(\"Ignoring cached namespace '%s' version %s because version %s is already loaded.\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Namespaces loaded by PyNWB when loading this file:\n",
      "\t core 2.6.0-alpha\n",
      "\t hdmf-common 1.8.0\n",
      "\t hdmf-experimental 0.5.0\n",
      "\t ndx-franklab-novela 0.1.0\n",
      "\n",
      "/Users/rly/Documents/NWB_Data/dandisets/000409/sub-CSHL047_ses-b52182e7-39f6-4914-9717-136db589706e_behavior+ecephys+image.nwb\n",
      "Namespaces cached in this file:\n",
      "\t core 2.5.0\n",
      "\t hdmf-common 1.5.1\n",
      "\t hdmf-experimental 0.2.0\n",
      "\t ndx-ibl 0.1.0\n",
      "\t ndx-pose 0.1.1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rly/Documents/NWB/hdmf/src/hdmf/spec/namespace.py:531: UserWarning: Ignoring cached namespace 'hdmf-common' version 1.5.1 because version 1.8.0 is already loaded.\n",
      "  warn(\"Ignoring cached namespace '%s' version %s because version %s is already loaded.\"\n",
      "/Users/rly/Documents/NWB/hdmf/src/hdmf/spec/namespace.py:531: UserWarning: Ignoring cached namespace 'hdmf-experimental' version 0.2.0 because version 0.5.0 is already loaded.\n",
      "  warn(\"Ignoring cached namespace '%s' version %s because version %s is already loaded.\"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Namespaces loaded by PyNWB when loading this file:\n",
      "\t core 2.6.0-alpha\n",
      "\t hdmf-common 1.8.0\n",
      "\t hdmf-experimental 0.5.0\n",
      "\t ndx-ibl 0.1.0\n",
      "\t ndx-pose 0.1.1\n",
      "\n",
      "/Users/rly/Downloads/20230622_155936 (1).nwb\n",
      "Namespaces cached in this file:\n",
      "\t core 2.5.0\n",
      "\t hdmf-common 1.6.0\n",
      "\t hdmf-experimental 0.3.0\n",
      "\t ndx-franklab-novela 0.1.0\n",
      "Namespaces loaded by PyNWB when loading this file:\n",
      "\t core 2.6.0-alpha\n",
      "\t hdmf-common 1.8.0\n",
      "\t hdmf-experimental 0.5.0\n",
      "\t ndx-franklab-novela 0.1.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from pynwb import get_type_map, NWBHDF5IO\n",
    "from hdmf.build import TypeMap\n",
    "from hdmf.spec import NamespaceCatalog\n",
    "from pynwb.spec import NWBDatasetSpec, NWBGroupSpec, NWBNamespace\n",
    "\n",
    "paths = [\n",
    "    \"/Users/rly/Downloads/20230622_155936 (1).nwb\",\n",
    "    \"/Users/rly/Documents/NWB_Data/dandisets/000409/sub-CSHL047_ses-b52182e7-39f6-4914-9717-136db589706e_behavior+ecephys+image.nwb\",\n",
    "    \"/Users/rly/Downloads/20230622_155936 (1).nwb\",\n",
    "]\n",
    "for path in paths:\n",
    "    print(path)\n",
    "\n",
    "    # 1) get namespaces that are in the file\n",
    "    \n",
    "    # get a new type map with no namespaces\n",
    "    tm = TypeMap(NamespaceCatalog(NWBGroupSpec, NWBDatasetSpec, NWBNamespace))\n",
    "    NWBHDF5IO.load_namespaces(tm, path)  #, file=file_obj, driver=driver)\n",
    "    namespaces = tm.namespace_catalog.namespaces\n",
    "    print(\"Namespaces cached in this file:\")\n",
    "    for ns in sorted(namespaces):\n",
    "        print(\"\\t\", ns, tm.namespace_catalog.get_namespace(ns)[\"version\"])\n",
    "\n",
    "    # 2) get namespaces that would be used by pynwb\n",
    "    \n",
    "    # if the namespace version loaded by pynwb is greater than the one\n",
    "    # cached in the file, then the namespace loaded by pynwb will be used\n",
    "    # (and a warning will be raised).\n",
    "    # else, the namespace cached in the file will be used.\n",
    "    \n",
    "    # get a new copy of the default type map loaded by pynwb\n",
    "    # NOTE that pynwb.load_namespaces(yaml_file) (if called) affects this \n",
    "    # loaded type map \n",
    "    tm = get_type_map()\n",
    "    NWBHDF5IO.load_namespaces(tm, path)  #, file=file_obj, driver=driver)\n",
    "    namespaces = tm.namespace_catalog.namespaces\n",
    "    print(\"Namespaces loaded by PyNWB when loading this file:\")\n",
    "    for ns in sorted(namespaces):\n",
    "        print(\"\\t\", ns, tm.namespace_catalog.get_namespace(ns)[\"version\"])\n",
    "\n",
    "    # 3) same as 2 but using NWBHDF5IO. takes about the same amount of time\n",
    "    # io = NWBHDF5IO(path, \"r\", load_namespaces=True)\n",
    "    # namespaces = io.manager.type_map.namespace_catalog.namespaces\n",
    "    # print(\"Namespaces loaded by PyNWB when loading this file:\")\n",
    "    # for ns in sorted(namespaces):\n",
    "    #     print(\"\\t\", ns, tm.namespace_catalog.get_namespace(ns)[\"version\"])\n",
    "\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0cf2f5de-c4ea-4d38-bcdc-b9c92d2c6126",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/rly/Downloads/20230622_155936 (1).nwb\n",
      "Namespaces cached in this file:\n",
      "\t core ('2.5.0',)\n",
      "\t hdmf-common ('1.6.0',)\n",
      "\t hdmf-experimental ('0.3.0',)\n",
      "\t ndx-franklab-novela ('0.1.0',)\n",
      "\n",
      "/Users/rly/Documents/NWB_Data/dandisets/000409/sub-CSHL047_ses-b52182e7-39f6-4914-9717-136db589706e_behavior+ecephys+image.nwb\n",
      "Namespaces cached in this file:\n",
      "\t core ('2.5.0',)\n",
      "\t hdmf-common ('1.5.1',)\n",
      "\t hdmf-experimental ('0.2.0',)\n",
      "\t ndx-ibl ('0.1.0',)\n",
      "\t ndx-pose ('0.1.1',)\n",
      "\n",
      "/Users/rly/Downloads/20230622_155936 (1).nwb\n",
      "Namespaces cached in this file:\n",
      "\t core ('2.5.0',)\n",
      "\t hdmf-common ('1.6.0',)\n",
      "\t hdmf-experimental ('0.3.0',)\n",
      "\t ndx-franklab-novela ('0.1.0',)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# using h5py is significantly faster to read the cached namespace versions\n",
    "# if you do not need to parse and resolve the spec\n",
    "import h5py\n",
    "\n",
    "def get_extension_versions(path: str) -> dict[str, tuple]:\n",
    "    # TODO first check whether this is an HDF5 file\n",
    "    file = h5py.File(path, \"r\")\n",
    "    ret = dict()\n",
    "    for ns_name in file[\"specifications\"].keys():\n",
    "        ret[ns_name] = tuple(file[\"specifications\"][ns_name].keys())\n",
    "    return ret\n",
    "\n",
    "for path in paths:\n",
    "    print(path)\n",
    "    ver = get_extension_versions(path)\n",
    "    \n",
    "    print(\"Namespaces cached in this file:\")\n",
    "    for ns_name, ns_versions in ver.items():\n",
    "        print(\"\\t\", ns_name, ns_versions)\n",
    "\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9b59b475-300c-413c-a0b1-53e240ff473e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}