From d4b8d9e7a5357b8a417a744c0d83792ab86010ce Mon Sep 17 00:00:00 2001
From: Simon Willison <swillison@gmail.com>
Date: Fri, 11 Dec 2020 21:41:07 -0800
Subject: [PATCH] Improvements to most/least common values

* Record total_rows for each column
* Record (value, count) if there is just a single distinct value
* Do not calculate most/least common if all values are distinct
* Calculate table count once per table, not once per column
---
 sqlite_utils/cli.py |  6 ++++--
 sqlite_utils/db.py  | 10 +++++++---
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/sqlite_utils/cli.py b/sqlite_utils/cli.py
index e779bef..6cd23f7 100644
--- a/sqlite_utils/cli.py
+++ b/sqlite_utils/cli.py
@@ -1357,7 +1357,7 @@ def insert_files(
 @cli.command(name="analyze-tables")
 @click.argument(
     "path",
-    type=click.Path(file_okay=True, dir_okay=False, allow_dash=False),
+    type=click.Path(file_okay=True, dir_okay=False, allow_dash=False, exists=True),
     required=True,
 )
 @click.argument("tables", nargs=-1)
@@ -1375,12 +1375,14 @@ def analyze_tables(
     if not tables:
         tables = db.table_names()
     todo = []
+    table_counts = {}
     for table in tables:
+        table_counts[table] = db[table].count
         for column in db[table].columns:
             todo.append((table, column.name))
     # Now we now how many we need to do
     for i, (table, column) in enumerate(todo):
-        column_details = db[table].analyze_column(column)
+        column_details = db[table].analyze_column(column, total_rows=table_counts[table])
         if save:
             db["_analyze_tables"].insert(
                 column_details._asdict(), pk=("table", "column"), replace=True
diff --git a/sqlite_utils/db.py b/sqlite_utils/db.py
index 460cb9a..ae77f6a 100644
--- a/sqlite_utils/db.py
+++ b/sqlite_utils/db.py
@@ -56,6 +56,7 @@ ColumnDetails = namedtuple(
     (
         "table",
         "column",
+        "total_rows",
         "num_null",
         "num_blank",
         "num_distinct",
@@ -1942,9 +1943,11 @@ class Table(Queryable):
             )
         return self
 
-    def analyze_column(self, column, common_limit=10):
+    def analyze_column(self, column, common_limit=10, total_rows=None):
         db = self.db
         table = self.name
+        if total_rows is None:
+            total_rows = db[table].count
         num_null = db.execute(
             "select count(*) from [{}] where [{}] is null".format(table, column)
         ).fetchone()[0]
@@ -1960,8 +1963,8 @@ class Table(Queryable):
             value = db.execute(
                 "select [{}] from [{}] limit 1".format(column, table)
             ).fetchone()[0]
-            most_common = [value]
-        else:
+            most_common = [(value, total_rows)]
+        elif num_distinct != total_rows:
             most_common = [
                 (r[0], r[1])
                 for r in db.execute(
@@ -1985,6 +1988,7 @@ class Table(Queryable):
         return ColumnDetails(
             self.name,
             column,
+            total_rows,
             num_null,
             num_blank,
             num_distinct,