From 9b18b6bec51fa0c475a46e50d81f9bd53a6eba63 Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Mon, 11 Mar 2024 11:31:42 -0700
Subject: [PATCH 1/9] PERF: Avoid np.divmod in RangeIndex._shallow_copy

---
 pandas/_libs/lib.pyx         |  6 +++---
 pandas/core/indexes/range.py | 13 +++++--------
 2 files changed, 8 insertions(+), 11 deletions(-)

diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index 00668576d5d53..4a692de802804 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -659,7 +659,7 @@ ctypedef fused int6432_t:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
+def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n, int diff=1) -> bool:
     """
     Perform an element by element comparison on 1-d integer arrays, meant for indexer
     comparisons
@@ -667,12 +667,12 @@ def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
     cdef:
         Py_ssize_t i
 
-    if left.size != n:
+    if left.size != n or diff == 0:
         return False
 
     for i in range(n):
 
-        if left[i] != i:
+        if left[i] != left[0] + i * diff:
             return False
 
     return True
diff --git a/pandas/core/indexes/range.py b/pandas/core/indexes/range.py
index 24f53f16e1985..13af2de28fe13 100644
--- a/pandas/core/indexes/range.py
+++ b/pandas/core/indexes/range.py
@@ -476,14 +476,11 @@ def _shallow_copy(self, values, name: Hashable = no_default):
             # GH 46675 & 43885: If values is equally spaced, return a
             # more memory-compact RangeIndex instead of Index with 64-bit dtype
             diff = values[1] - values[0]
-            if not missing.isna(diff) and diff != 0:
-                maybe_range_indexer, remainder = np.divmod(values - values[0], diff)
-                if (
-                    lib.is_range_indexer(maybe_range_indexer, len(maybe_range_indexer))
-                    and not remainder.any()
-                ):
-                    new_range = range(values[0], values[-1] + diff, diff)
-                    return type(self)._simple_new(new_range, name=name)
+            if not missing.isna(diff) and lib.is_range_indexer(
+                values, len(values), diff
+            ):
+                new_range = range(values[0], values[-1] + diff, diff)
+                return type(self)._simple_new(new_range, name=name)
         return self._constructor._simple_new(values, name=name)
 
     def _view(self) -> Self:

From 9bcc4eacb6ba7d1adeafbb331fab753a5eb0616f Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Mon, 11 Mar 2024 12:48:36 -0700
Subject: [PATCH 2/9] Make is_range

---
 pandas/_libs/lib.pyi         |  4 ++++
 pandas/_libs/lib.pyx         | 26 +++++++++++++++++++++++---
 pandas/core/indexes/range.py |  4 +---
 3 files changed, 28 insertions(+), 6 deletions(-)

diff --git a/pandas/_libs/lib.pyi b/pandas/_libs/lib.pyi
index 34193a9b1d231..d7ba16d5e9d69 100644
--- a/pandas/_libs/lib.pyi
+++ b/pandas/_libs/lib.pyi
@@ -231,3 +231,7 @@ def is_range_indexer(
     left: np.ndarray,
     n: int,  # np.ndarray[np.int64, ndim=1]
 ) -> bool: ...
+def is_range(
+    left: np.ndarray,
+    diff: int,  # np.ndarray[np.int64, ndim=1]
+) -> bool: ...
diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index 4a692de802804..8412d3b4fd0e7 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -659,7 +659,7 @@ ctypedef fused int6432_t:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n, int diff=1) -> bool:
+def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
     """
     Perform an element by element comparison on 1-d integer arrays, meant for indexer
     comparisons
@@ -667,12 +667,32 @@ def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n, int diff=1)
     cdef:
         Py_ssize_t i
 
-    if left.size != n or diff == 0:
+    if left.size != n:
         return False
 
     for i in range(n):
 
-        if left[i] != left[0] + i * diff:
+        if left[i] != i:
+            return False
+
+    return True
+
+
+@cython.wraparound(False)
+@cython.boundscheck(False)
+def is_range(ndarray[int6432_t, ndim=1] sequence, int64_t diff) -> bool:
+    """
+    Check if sequence is equivalent to a range with the specified diff.
+    """
+    cdef:
+        Py_ssize_t i, n = len(sequence)
+
+    if diff == 0:
+        return False
+
+    for i in range(n):
+
+        if sequence[i] != sequence[0] + i * diff:
             return False
 
     return True
diff --git a/pandas/core/indexes/range.py b/pandas/core/indexes/range.py
index 13af2de28fe13..a66cfabaee171 100644
--- a/pandas/core/indexes/range.py
+++ b/pandas/core/indexes/range.py
@@ -476,9 +476,7 @@ def _shallow_copy(self, values, name: Hashable = no_default):
             # GH 46675 & 43885: If values is equally spaced, return a
             # more memory-compact RangeIndex instead of Index with 64-bit dtype
             diff = values[1] - values[0]
-            if not missing.isna(diff) and lib.is_range_indexer(
-                values, len(values), diff
-            ):
+            if not missing.isna(diff) and lib.is_range(values, diff):
                 new_range = range(values[0], values[-1] + diff, diff)
                 return type(self)._simple_new(new_range, name=name)
         return self._constructor._simple_new(values, name=name)

From c9339d56c662f384e6be5fae43dec198ad8f263f Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Mon, 11 Mar 2024 12:50:17 -0700
Subject: [PATCH 3/9] pyi error

---
 pandas/_libs/lib.pyi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/_libs/lib.pyi b/pandas/_libs/lib.pyi
index d7ba16d5e9d69..3e14bf12c43f8 100644
--- a/pandas/_libs/lib.pyi
+++ b/pandas/_libs/lib.pyi
@@ -232,6 +232,6 @@ def is_range_indexer(
     n: int,  # np.ndarray[np.int64, ndim=1]
 ) -> bool: ...
 def is_range(
-    left: np.ndarray,
+    sequence: np.ndarray,
     diff: int,  # np.ndarray[np.int64, ndim=1]
 ) -> bool: ...

From 0b484bdb270168422dd678ad08171257ac10d186 Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Wed, 13 Mar 2024 10:18:07 -0700
Subject: [PATCH 4/9] Use step

---
 pandas/_libs/lib.pyi         | 4 ++--
 pandas/_libs/lib.pyx         | 8 ++++----
 pandas/core/indexes/range.py | 2 +-
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/pandas/_libs/lib.pyi b/pandas/_libs/lib.pyi
index 3e14bf12c43f8..b39d32d069619 100644
--- a/pandas/_libs/lib.pyi
+++ b/pandas/_libs/lib.pyi
@@ -231,7 +231,7 @@ def is_range_indexer(
     left: np.ndarray,
     n: int,  # np.ndarray[np.int64, ndim=1]
 ) -> bool: ...
-def is_range(
+def is_sequence_range(
     sequence: np.ndarray,
-    diff: int,  # np.ndarray[np.int64, ndim=1]
+    step: int,  # np.ndarray[np.int64, ndim=1]
 ) -> bool: ...
diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index 8412d3b4fd0e7..0780c4f0c454c 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -680,19 +680,19 @@ def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def is_range(ndarray[int6432_t, ndim=1] sequence, int64_t diff) -> bool:
+def is_sequence_range(ndarray[intp_t, ndim=1] sequence, int64_t step) -> bool:
     """
-    Check if sequence is equivalent to a range with the specified diff.
+    Check if sequence is equivalent to a range with the specified step.
     """
     cdef:
         Py_ssize_t i, n = len(sequence)
 
-    if diff == 0:
+    if step == 0:
         return False
 
     for i in range(n):
 
-        if sequence[i] != sequence[0] + i * diff:
+        if sequence[i] != sequence[0] + i * step:
             return False
 
     return True
diff --git a/pandas/core/indexes/range.py b/pandas/core/indexes/range.py
index 64c855347dc7a..2e7c3a1ad6e01 100644
--- a/pandas/core/indexes/range.py
+++ b/pandas/core/indexes/range.py
@@ -482,7 +482,7 @@ def _shallow_copy(self, values, name: Hashable = no_default):
                 new_range = range(start, start + self.step, self.step)
                 return type(self)._simple_new(new_range, name=name)
             diff = values[1] - values[0]
-            if not missing.isna(diff) and lib.is_range(values, diff):
+            if not missing.isna(diff) and lib.is_sequence_range(values, diff):
                 new_range = range(values[0], values[-1] + diff, diff)
                 return type(self)._simple_new(new_range, name=name)
         return self._constructor._simple_new(values, name=name)

From 997416f848e13318e63cdfc787d209982be40746 Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Wed, 13 Mar 2024 12:38:11 -0700
Subject: [PATCH 5/9] Switch back to int6432

---
 pandas/_libs/lib.pyx | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index 0780c4f0c454c..5196148cfd468 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -680,7 +680,7 @@ def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def is_sequence_range(ndarray[intp_t, ndim=1] sequence, int64_t step) -> bool:
+def is_sequence_range(ndarray[int6432_t, ndim=1] sequence, int64_t step) -> bool:
     """
     Check if sequence is equivalent to a range with the specified step.
     """

From b8ea98ca75b06fb072d55b4a25d619f9c03a837e Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Thu, 14 Mar 2024 09:26:04 -0700
Subject: [PATCH 6/9] try int64_t

---
 pandas/_libs/lib.pyx | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index 5196148cfd468..bc2a2e5f99e53 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -680,7 +680,7 @@ def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def is_sequence_range(ndarray[int6432_t, ndim=1] sequence, int64_t step) -> bool:
+def is_sequence_range(ndarray[int64_t, ndim=1] sequence, int64_t step) -> bool:
     """
     Check if sequence is equivalent to a range with the specified step.
     """

From 21f13d9d9a80e7b63489b061ee7880783b0f465b Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Thu, 14 Mar 2024 11:34:12 -0700
Subject: [PATCH 7/9] Revert "try int64_t"

This reverts commit b8ea98ca75b06fb072d55b4a25d619f9c03a837e.
---
 pandas/_libs/lib.pyx | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index bc2a2e5f99e53..5196148cfd468 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -680,7 +680,7 @@ def is_range_indexer(ndarray[int6432_t, ndim=1] left, Py_ssize_t n) -> bool:
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def is_sequence_range(ndarray[int64_t, ndim=1] sequence, int64_t step) -> bool:
+def is_sequence_range(ndarray[int6432_t, ndim=1] sequence, int64_t step) -> bool:
     """
     Check if sequence is equivalent to a range with the specified step.
     """

From 503d96f061dd44cf54f029048ea96b81cbda88cb Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Tue, 19 Mar 2024 15:58:26 -0700
Subject: [PATCH 8/9] Adjust maybe_sequence_to_range

---
 pandas/core/indexes/base.py | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/pandas/core/indexes/base.py b/pandas/core/indexes/base.py
index 111263f588439..8185e5abde321 100644
--- a/pandas/core/indexes/base.py
+++ b/pandas/core/indexes/base.py
@@ -7179,13 +7179,7 @@ def maybe_sequence_to_range(sequence) -> Any | range:
     diff = np_sequence[1] - np_sequence[0]
     if diff == 0:
         return sequence
-    elif len(np_sequence) == 2:
-        return range(np_sequence[0], np_sequence[1] + diff, diff)
-    maybe_range_indexer, remainder = np.divmod(np_sequence - np_sequence[0], diff)
-    if (
-        lib.is_range_indexer(maybe_range_indexer, len(maybe_range_indexer))
-        and not remainder.any()
-    ):
+    elif len(np_sequence) == 2 or lib.is_sequence_range(np_sequence, diff):
         return range(np_sequence[0], np_sequence[-1] + diff, diff)
     else:
         return sequence

From 5f31dd85273f1a8329d00746b4332a5c61a984b4 Mon Sep 17 00:00:00 2001
From: Matthew Roeschke <10647082+mroeschke@users.noreply.github.com>
Date: Thu, 21 Mar 2024 09:14:34 -0700
Subject: [PATCH 9/9] Access first element once

---
 pandas/_libs/lib.pyx | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/pandas/_libs/lib.pyx b/pandas/_libs/lib.pyx
index 5196148cfd468..a2205454a5a46 100644
--- a/pandas/_libs/lib.pyx
+++ b/pandas/_libs/lib.pyx
@@ -686,15 +686,17 @@ def is_sequence_range(ndarray[int6432_t, ndim=1] sequence, int64_t step) -> bool
     """
     cdef:
         Py_ssize_t i, n = len(sequence)
+        int6432_t first_element
 
     if step == 0:
         return False
+    if n == 0:
+        return True
 
-    for i in range(n):
-
-        if sequence[i] != sequence[0] + i * step:
+    first_element = sequence[0]
+    for i in range(1, n):
+        if sequence[i] != first_element + i * step:
             return False
-
     return True