docs(blog-post): pydata performance part 2; polars and datafusion

Author: cpcloud

docs/_freeze/posts/pydata-performance-part2/index/execute-results/html.json

@@ -0,0 +1,33 @@
+from __future__ import annotations
+
+import ibis
+from ibis import _
+
+ibis.set_backend("datafusion")
+
+expr = (
+    ibis.read_parquet("/data/pypi-parquet/*.parquet")
+    .filter(
+        [
+            _.path.re_search(
+                r"\.(asm|c|cc|cpp|cxx|h|hpp|rs|[Ff][0-9]{0,2}(?:or)?|go)$"
+            ),
+            ~_.path.re_search(r"(^|/)test(|s|ing)"),
+            ~_.path.contains("/site-packages/"),
+        ]
+    )
+    .group_by(
+        month=_.uploaded_on.truncate("M"),
+        ext=_.path.re_extract(r"\.([a-z0-9]+)$", 1)
+        .re_replace(r"cxx|cpp|cc|c|hpp|h", "C/C++")
+        .re_replace("^f.*$", "Fortran")
+        .replace("rs", "Rust")
+        .replace("go", "Go")
+        .replace("asm", "Assembly")
+        .nullif(""),
+    )
+    .aggregate(project_count=_.project_name.nunique())
+    .dropna("ext")
+    .order_by([_.month.desc(), _.project_count.desc()])
+)
+df = expr.to_pandas()

docs/posts/pydata-performance-part2/datafusion_ibis.py

@@ -0,0 +1,12 @@
+from __future__ import annotations
+
+import datafusion
+
+with open("./datafusion_native.sql") as f:
+    query = f.read()
+
+ctx = datafusion.SessionContext()
+ctx.register_parquet(name="pypi", path="/data/pypi-parquet/*.parquet")
+expr = ctx.sql(query)
+
+df = expr.to_pandas()

docs/posts/pydata-performance-part2/datafusion_native.py

@@ -0,0 +1,47 @@
+SELECT
+  month,
+  ext,
+  COUNT(DISTINCT project_name) AS project_count
+FROM (
+  SELECT
+    project_name,
+    DATE_TRUNC('month', uploaded_on) AS month,
+    NULLIF(
+      REPLACE(
+        REPLACE(
+          REPLACE(
+            REGEXP_REPLACE(
+              REGEXP_REPLACE(
+                REGEXP_MATCH(path, CONCAT('(', '\.([a-z0-9]+)$', ')'))[2],
+                'cxx|cpp|cc|c|hpp|h',
+                'C/C++',
+                'g'
+              ),
+              '^f.*$',
+              'Fortran',
+              'g'
+            ),
+            'rs',
+            'Rust'
+          ),
+          'go',
+          'Go'
+        ),
+        'asm',
+        'Assembly'
+      ),
+      ''
+    ) AS ext
+  FROM pypi
+  WHERE COALESCE(
+      ARRAY_LENGTH(
+        REGEXP_MATCH(path, '\.(asm|c|cc|cpp|cxx|h|hpp|rs|[Ff][0-9]{0,2}(?:or)?|go)$')
+      ) > 0,
+      FALSE
+    )
+    AND NOT COALESCE(ARRAY_LENGTH(REGEXP_MATCH(path, '(^|/)test(|s|ing)')) > 0, FALSE)
+    AND NOT STRPOS(path, '/site-packages/') > 0
+)
+WHERE ext IS NOT NULL
+GROUP BY month, ext
+ORDER BY month DESC, project_count DESC

docs/posts/pydata-performance-part2/datafusion_native.sql

@@ -0,0 +1,31 @@
+from __future__ import annotations
+
+import ibis
+from ibis import _
+
+expr = (
+    ibis.read_parquet("/data/pypi-parquet/*.parquet")
+    .filter(
+        [
+            _.path.re_search(
+                r"\.(asm|c|cc|cpp|cxx|h|hpp|rs|[Ff][0-9]{0,2}(?:or)?|go)$"
+            ),
+            ~_.path.re_search(r"(^|/)test(|s|ing)"),
+            ~_.path.contains("/site-packages/"),
+        ]
+    )
+    .group_by(
+        month=_.uploaded_on.truncate("M"),
+        ext=_.path.re_extract(r"\.([a-z0-9]+)$", 1)
+        .re_replace(r"cxx|cpp|cc|c|hpp|h", "C/C++")
+        .re_replace("^f.*$", "Fortran")
+        .replace("rs", "Rust")
+        .replace("go", "Go")
+        .replace("asm", "Assembly")
+        .nullif(""),
+    )
+    .aggregate(project_count=_.project_name.nunique())
+    .dropna("ext")
+    .order_by([_.month.desc(), _.project_count.desc()])  # <1>
+)
+df = expr.to_pandas()