Issue python#23573: Increased performance of string search operations (str.find,

str.index, str.count, the in operator, str.split, str.partition) with
arguments of different kinds (UCS1, UCS2, UCS4).

Author: serhiy-storchaka

Misc/NEWS

@@ -10,6 +10,10 @@ Release date: 2015-03-28
 Core and Builtins
 -----------------
 
+- Issue #23573: Increased performance of string search operations (str.find,
+  str.index, str.count, the in operator, str.split, str.partition) with
+  arguments of different kinds (UCS1, UCS2, UCS4).
+
 - Issue #23753: Python doesn't support anymore platforms without stat() or
   fstat(), these functions are always required.
 

Objects/bytearrayobject.c

@@ -1142,7 +1142,7 @@ bytearray_find_internal(PyByteArrayObject *self, PyObject *args, int dir)
     char byte;
     Py_buffer subbuf;
     const char *sub;
-    Py_ssize_t sub_len;
+    Py_ssize_t len, sub_len;
     Py_ssize_t start=0, end=PY_SSIZE_T_MAX;
     Py_ssize_t res;
 
@@ -1161,15 +1161,30 @@ bytearray_find_internal(PyByteArrayObject *self, PyObject *args, int dir)
         sub = &byte;
         sub_len = 1;
     }
+    len = PyByteArray_GET_SIZE(self);
 
-    if (dir > 0)
-        res = stringlib_find_slice(
-            PyByteArray_AS_STRING(self), PyByteArray_GET_SIZE(self),
-            sub, sub_len, start, end);
-    else
-        res = stringlib_rfind_slice(
-            PyByteArray_AS_STRING(self), PyByteArray_GET_SIZE(self),
-            sub, sub_len, start, end);
+    ADJUST_INDICES(start, end, len);
+    if (end - start < sub_len)
+        res = -1;
+    else if (sub_len == 1) {
+        unsigned char needle = *sub;
+        int mode = (dir > 0) ? FAST_SEARCH : FAST_RSEARCH;
+        res = stringlib_fastsearch_memchr_1char(
+            PyByteArray_AS_STRING(self) + start, end - start,
+            needle, needle, mode);
+        if (res >= 0)
+            res += start;
+    }
+    else {
+        if (dir > 0)
+            res = stringlib_find_slice(
+                PyByteArray_AS_STRING(self), len,
+                sub, sub_len, start, end);
+        else
+            res = stringlib_rfind_slice(
+                PyByteArray_AS_STRING(self), len,
+                sub, sub_len, start, end);
+    }
 
     if (subobj)
         PyBuffer_Release(&subbuf);

Objects/bytesobject.c

@@ -1914,7 +1914,7 @@ bytes_find_internal(PyBytesObject *self, PyObject *args, int dir)
     char byte;
     Py_buffer subbuf;
     const char *sub;
-    Py_ssize_t sub_len;
+    Py_ssize_t len, sub_len;
     Py_ssize_t start=0, end=PY_SSIZE_T_MAX;
     Py_ssize_t res;
 
@@ -1933,15 +1933,30 @@ bytes_find_internal(PyBytesObject *self, PyObject *args, int dir)
         sub = &byte;
         sub_len = 1;
     }
+    len = PyBytes_GET_SIZE(self);
 
-    if (dir > 0)
-        res = stringlib_find_slice(
-            PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self),
-            sub, sub_len, start, end);
-    else
-        res = stringlib_rfind_slice(
-            PyBytes_AS_STRING(self), PyBytes_GET_SIZE(self),
-            sub, sub_len, start, end);
+    ADJUST_INDICES(start, end, len);
+    if (end - start < sub_len)
+        res = -1;
+    else if (sub_len == 1) {
+        unsigned char needle = *sub;
+        int mode = (dir > 0) ? FAST_SEARCH : FAST_RSEARCH;
+        res = stringlib_fastsearch_memchr_1char(
+            PyBytes_AS_STRING(self) + start, end - start,
+            needle, needle, mode);
+        if (res >= 0)
+            res += start;
+    }
+    else {
+        if (dir > 0)
+            res = stringlib_find_slice(
+                PyBytes_AS_STRING(self), len,
+                sub, sub_len, start, end);
+        else
+            res = stringlib_rfind_slice(
+                PyBytes_AS_STRING(self), len,
+                sub, sub_len, start, end);
+    }
 
     if (subobj)
         PyBuffer_Release(&subbuf);

Objects/stringlib/fastsearch.h

@@ -36,7 +36,7 @@
 Py_LOCAL_INLINE(Py_ssize_t)
 STRINGLIB(fastsearch_memchr_1char)(const STRINGLIB_CHAR* s, Py_ssize_t n,
                                    STRINGLIB_CHAR ch, unsigned char needle,
-                                   Py_ssize_t maxcount, int mode)
+                                   int mode)
 {
     if (mode == FAST_SEARCH) {
         const STRINGLIB_CHAR *ptr = s;
@@ -115,7 +115,7 @@ FASTSEARCH(const STRINGLIB_CHAR* s, Py_ssize_t n,
             if (needle != 0)
 #endif
                 return STRINGLIB(fastsearch_memchr_1char)
-                       (s, n, p[0], needle, maxcount, mode);
+                       (s, n, p[0], needle, mode);
         }
         if (mode == FAST_COUNT) {
             for (i = 0; i < n; i++)

Objects/stringlib/find.h

@@ -11,8 +11,7 @@ STRINGLIB(find)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
 {
     Py_ssize_t pos;
 
-    if (str_len < 0)
-        return -1;
+    assert(str_len >= 0);
     if (sub_len == 0)
         return offset;
 
@@ -31,8 +30,7 @@ STRINGLIB(rfind)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
 {
     Py_ssize_t pos;
 
-    if (str_len < 0)
-        return -1;
+    assert(str_len >= 0);
     if (sub_len == 0)
         return str_len + offset;
 
@@ -44,27 +42,11 @@ STRINGLIB(rfind)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
     return pos;
 }
 
-/* helper macro to fixup start/end slice values */
-#define ADJUST_INDICES(start, end, len)         \
-    if (end > len)                              \
-        end = len;                              \
-    else if (end < 0) {                         \
-        end += len;                             \
-        if (end < 0)                            \
-            end = 0;                            \
-    }                                           \
-    if (start < 0) {                            \
-        start += len;                           \
-        if (start < 0)                          \
-            start = 0;                          \
-    }
-
 Py_LOCAL_INLINE(Py_ssize_t)
 STRINGLIB(find_slice)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
                      const STRINGLIB_CHAR* sub, Py_ssize_t sub_len,
                      Py_ssize_t start, Py_ssize_t end)
 {
-    ADJUST_INDICES(start, end, str_len);
     return STRINGLIB(find)(str + start, end - start, sub, sub_len, start);
 }
 
@@ -73,7 +55,6 @@ STRINGLIB(rfind_slice)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
                       const STRINGLIB_CHAR* sub, Py_ssize_t sub_len,
                       Py_ssize_t start, Py_ssize_t end)
 {
-    ADJUST_INDICES(start, end, str_len);
     return STRINGLIB(rfind)(str + start, end - start, sub, sub_len, start);
 }