update documentation

matthiasprobst · Dec 12, 2023 · 555dc99 · 555dc99
1 parent 51b2be2
commit 555dc99
Show file tree

Hide file tree

Showing 7 changed files with 119 additions and 640 deletions.
diff --git a/docs/conventions/examples/Provenance.ipynb b/docs/conventions/examples/Provenance.ipynb
diff --git a/docs/conventions/ontologies.ipynb b/docs/conventions/ontologies.ipynb
@@ -551,7 +551,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.0"
+   "version": "3.8.18"
   }
  },
  "nbformat": 4,

diff --git a/docs/database/index.rst b/docs/database/index.rst
@@ -1,7 +1,7 @@
 HDF5-Database
 =============
 
-HDF5 can be considered a database itself, as it allows multiple datasets and their metadata (attributes) to be stored in a single file. Most of the time you want to find records in an HDF5 file based on the attributes. However, the `h5py` package does not provide a function to do this.
+HDF5 can be considered a database itself, as it allows multiple datasets and their metadata (attributes) to be stored in a single file. Most of the time, you want to find records in an HDF5 file based on the attributes. However, the `h5py` package does not provide a function to do this.
 
 The `h5rdmtoolbox` provides an interface to perform queries on a single or even multiple HDF5 files. This is shown in one of the subchapters here. However, this may not always be the fastest way to find data in an HDF5 file. A more effective way is to map the metadata to a dedicated database. One such example is MongoDB. The query is performed on the much more efficient dedicated database, then returned to the original file to continue working.
 

diff --git a/docs/database/mongo.ipynb b/docs/database/mongo.ipynb
@@ -52,9 +52,7 @@
     "import numpy as np\n",
     "from pprint import pprint\n",
     "\n",
-    "h5tbx.use(None)\n",
-    "\n",
-    "h5tbx.__version__"
+    "h5tbx.use(None)"
    ]
   },
   {

diff --git a/docs/wrapper/DumpFile.ipynb b/docs/wrapper/DumpFile.ipynb
diff --git a/docs/wrapper/GroupCreation.ipynb b/docs/wrapper/GroupCreation.ipynb
@@ -100,26 +100,46 @@
    "id": "70a68b4a-6715-46ff-889d-fcc1a7f7ec3a",
    "metadata": {},
    "source": [
-    "The equivalent could be achieved by using the `find` query - in fact it is called in the method `get_groups()` (More on `find()` can be found [here](../database/Serverless.ipynb))"
+    "An alternative of exploring HDF5 files in general is the usage of databases. The HDF5 file can serve as a database itself. For this and more on database interfaces, see [here](../database/firstSteps.ipynb))"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 5,
-   "id": "b9edd387-e3a2-49b9-b51d-cf1da347c803",
+   "id": "50de3a0b-2aa6-4832-8d1b-2081dad74a42",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from h5rdmtoolbox.database import hdfdb"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "c90e3188-bb22-48f3-b7dc-88bc02daa5e4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "db = hdfdb.FileDB(h5.hdf_filename)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "22c37bdb-b7cd-4d58-a59a-2e478e704778",
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[<HDF5 wrapper group \"/mygrp\" (members: 0, convention: \"h5py\")>]\n"
+      "<LGroup \"/mygrp\" in \"C:\\Users\\da4323\\AppData\\Local\\h5rdmtoolbox\\h5rdmtoolbox\\tmp\\tmp_3\\tmp1.hdf\">\n"
      ]
     }
    ],
    "source": [
-    "with h5tbx.File(h5.hdf_filename) as h5:\n",
-    "    print(h5.find({'$name': {'$regex': '.*'}}, rec=False))"
+    "for r in db.find({'$name': {'$regex': '.*'}}, recursive=False):\n",
+    "    print(r)"
    ]
   }
  ],
@@ -139,7 +159,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.17"
+   "version": "3.8.18"
   }
  },
  "nbformat": 4,

diff --git a/h5rdmtoolbox/database/hdfdb/filedb.py b/h5rdmtoolbox/database/hdfdb/filedb.py
@@ -22,7 +22,9 @@ def find_one(self, *args, **kwargs) -> lazy.LHDFObject:
     def find(self, *args, **kwargs) -> Generator[lazy.LHDFObject, None, None]:
         """Please refer to the docstring of the find method of the GroupDB class"""
         with h5py.File(self.filename, 'r') as h5:
-            return GroupDB(h5).find(*args, **kwargs)
+            results = list(GroupDB(h5).find(*args, **kwargs))
+        for r in results:
+            yield r
 
 
 class FilesDB(NonInsertableDatabaseInterface, HDF5DatabaseInterface):