Fixup wording in notebook 3

Since `read_parquet` is discussed in notebook 2, we can drop the language introducing it here. I also drop the note on accessing the tables via `con.tables.<name>` since it feels out of place here IMO.
ibis-project · Jul 8, 2024 · de2d947 · de2d947
1 parent 4f54af4
commit de2d947
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 75 deletions.
diff --git a/03 - Switching Backends.ipynb b/03 - Switching Backends.ipynb
@@ -41,11 +41,11 @@
         "\n",
         "### Parquet loading\n",
         "\n",
-        "In the previous examples we used a pre-existing DuckDB database, and some\n",
-        "in-memory tables. Another common pattern is that you have a few parquet files\n",
-        "you want to work with. We can load those in to an in-memory DuckDB connection.\n",
-        "(Note that \"in-memory\" here just means ephemeral, DuckDB is still very happy to\n",
-        "operate on as much data as your hard drive can hold)."
+        "Here we use the `read_parquet` method discussed in the previous notebook to\n",
+        "load some data into DuckDB. Note that DuckDB treats this as a \"view\", meaning\n",
+        "that the data isn't copied or loaded into memory - it still only exists on disk\n",
+        "in the `parquet` files, meaning you can happily operate on parquet datasets\n",
+        "that are much larger than the RAM on your laptop."
       ]
     },
     {
@@ -91,47 +91,6 @@
       "execution_count": null,
       "outputs": []
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "The `read_parquet` method returns an Ibis table that points to the\n",
-        "to-be-ingested `parquet` file. \n",
-        "\n",
-        "`read_parquet` also registers the table with DuckDB (or another backend), so\n",
-        "you can also load the tables like we did for the `penguins` table in the\n",
-        "previous notebook."
-      ]
-    },
-    {
-      "cell_type": "code",
-      "metadata": {},
-      "source": [
-        "basics = con.tables.imdb_title_basics  # this cell is redundant, just here for demonstration"
-      ],
-      "execution_count": null,
-      "outputs": []
-    },
-    {
-      "cell_type": "code",
-      "metadata": {},
-      "source": [
-        "ratings = con.tables.imdb_title_ratings  # this cell is redundant, just here for demonstration"
-      ],
-      "execution_count": null,
-      "outputs": []
-    },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "scrolled": true
-      },
-      "source": [
-        "basics"
-      ],
-      "execution_count": null,
-      "outputs": []
-    },
     {
       "cell_type": "markdown",
       "metadata": {},
@@ -148,9 +107,7 @@
     },
     {
       "cell_type": "code",
-      "metadata": {
-        "scrolled": true
-      },
+      "metadata": {},
       "source": [
         "%load solutions/nb03_ex01.py"
       ],

diff --git a/quarto/03 - Switching Backends.qmd b/quarto/03 - Switching Backends.qmd
@@ -30,11 +30,11 @@ we'll run the same query on the full dataset.
 
 ### Parquet loading
 
-In the previous examples we used a pre-existing DuckDB database, and some
-in-memory tables. Another common pattern is that you have a few parquet files
-you want to work with. We can load those in to an in-memory DuckDB connection.
-(Note that "in-memory" here just means ephemeral, DuckDB is still very happy to
-operate on as much data as your hard drive can hold).
+Here we use the `read_parquet` method discussed in the previous notebook to
+load some data into DuckDB. Note that DuckDB treats this as a "view", meaning
+that the data isn't copied or loaded into memory - it still only exists on disk
+in the `parquet` files, meaning you can happily operate on parquet datasets
+that are much larger than the RAM on your laptop.
 
 ```{python}
 import ibis
@@ -59,26 +59,6 @@ ratings = con.read_parquet(
 )
 ```
 
-The `read_parquet` method returns an Ibis table that points to the
-to-be-ingested `parquet` file. 
-
-`read_parquet` also registers the table with DuckDB (or another backend), so
-you can also load the tables like we did for the `penguins` table in the
-previous notebook.
-
-```{python}
-basics = con.tables.imdb_title_basics  # this cell is redundant, just here for demonstration
-```
-
-```{python}
-ratings = con.tables.imdb_title_ratings  # this cell is redundant, just here for demonstration
-```
-
-```{python}
-#| scrolled: true
-basics
-```
-
 ## Exercises
 
 ### Exercise 1
@@ -89,7 +69,6 @@ Join `basics` with `ratings` on the `tconst` column.
 #### Solution
 
 ```{python}
-#| scrolled: true
 %load solutions/nb03_ex01.py
 ```