Rework get_string_data to cleaner get_c_string_buf_and_size

Author: vnlitvinov

pandas/_libs/hashtable_class_helper.pxi.in

@@ -9,7 +9,7 @@ WARNING: DO NOT edit .pxi FILE directly, .pxi is generated from .pxi.in
 # VectorData
 # ----------------------------------------------------------------------
 
-from pandas._libs.tslibs.util cimport get_string_data, get_string_data_checked
+from pandas._libs.tslibs.util cimport get_c_string
 
 {{py:
 
@@ -597,7 +597,7 @@ cdef class StringHashTable(HashTable):
         cdef:
             khiter_t k
             const char *v
-        v = get_string_data_checked(val, NULL)
+        v = get_c_string(val)
 
         k = kh_get_str(self.table, v)
         if k != self.table.n_buckets:
@@ -611,7 +611,7 @@ cdef class StringHashTable(HashTable):
             int ret = 0
             const char *v
 
-        v = get_string_data_checked(val, NULL)
+        v = get_c_string(val)
 
         k = kh_put_str(self.table, v, &ret)
         self.table.keys[k] = key
@@ -634,7 +634,7 @@ cdef class StringHashTable(HashTable):
         vecs = <const char **>malloc(n * sizeof(char *))
         for i in range(n):
             val = values[i]
-            v = get_string_data_checked(val, NULL)
+            v = get_c_string(val)
             vecs[i] = v
 
         with nogil:
@@ -664,9 +664,9 @@ cdef class StringHashTable(HashTable):
             val = values[i]
 
             if isinstance(val, (str, unicode)):
-                v = get_string_data(val, NULL)
+                v = get_c_string(val)
             else:
-                v = get_string_data(self.na_string_sentinel, NULL)
+                v = get_c_string(self.na_string_sentinel)
             vecs[i] = v
 
         with nogil:
@@ -697,9 +697,9 @@ cdef class StringHashTable(HashTable):
             val = values[i]
 
             if isinstance(val, (str, unicode)):
-                v = get_string_data(val, NULL)
+                v = get_c_string(val)
             else:
-                v = get_string_data(self.na_string_sentinel, NULL)
+                v = get_c_string(self.na_string_sentinel)
             vecs[i] = v
 
         with nogil:
@@ -778,7 +778,7 @@ cdef class StringHashTable(HashTable):
                 labels[i] = na_sentinel
             else:
                 # if ignore_na is False, we also stringify NaN/None/etc.
-                v = get_string_data_checked(val, NULL)
+                v = get_c_string(val)
                 vecs[i] = v
 
         # compute

pandas/_libs/tslibs/np_datetime.pyx

@@ -12,7 +12,7 @@ from cpython.datetime cimport (datetime, date,
 PyDateTime_IMPORT
 
 from numpy cimport int64_t
-from pandas._libs.tslibs.util cimport get_string_data_checked
+from pandas._libs.tslibs.util cimport get_c_string_buf_and_size
 
 cdef extern from "src/datetime/np_datetime.h":
     int cmp_npy_datetimestruct(npy_datetimestruct *a,
@@ -175,8 +175,8 @@ cdef inline int _string_to_dts(object val, npy_datetimestruct* dts,
                                int* out_local, int* out_tzoffset) except? -1:
     cdef:
         Py_ssize_t length
-        const char* tmp
+        const char* buf
 
-    tmp = get_string_data_checked(val, &length)
-    return parse_iso_8601_datetime(tmp, length,
+    buf = get_c_string_buf_and_size(val, &length)
+    return parse_iso_8601_datetime(buf, length,
                                    dts, out_local, out_tzoffset)

pandas/_libs/tslibs/util.pxd

@@ -18,7 +18,6 @@ cdef extern from "Python.h":
     # Note: importing extern-style allows us to declare these as nogil
     # functions, whereas `from cpython cimport` does not.
     bint PyUnicode_Check(object obj) nogil
-    bint PyBytes_Check(object obj) nogil
     bint PyString_Check(object obj) nogil
     bint PyBool_Check(object obj) nogil
     bint PyFloat_Check(object obj) nogil
@@ -30,8 +29,8 @@ cdef extern from "Python.h":
     # potentially allocate memory inside in unlikely case of when underlying
     # unicode object was stored as non-utf8 and utf8 wasn't requested before.
     bint PyBytes_AsStringAndSize(object obj, char** buf,
-                                 Py_ssize_t* length)
-    char* PyUnicode_AsUTF8AndSize(object obj, Py_ssize_t* length)
+                                 Py_ssize_t* length) except -1
+    char* PyUnicode_AsUTF8AndSize(object obj, Py_ssize_t* length) except NULL
 
 from numpy cimport int64_t
 
@@ -238,18 +237,16 @@ cdef inline bint is_nan(object val):
     return (is_float_object(val) or is_complex_object(val)) and val != val
 
 
-cdef inline const char* get_string_data(object s, Py_ssize_t *length):
+cdef inline const char* get_c_string_buf_and_size(object s,
+                                                  Py_ssize_t *length):
     """
-    Extract internal char * buffer of unicode or bytes object `s` to `buf` with
+    Extract internal char * buffer of unicode or bytes object `s` with
     getting length of this internal buffer saved in `length`.
-    Returns `False` if it failed to extract such buffer for whatever reason,
-    otherwise returns `True`.
 
     Notes
     -----
-    Python object owns memory, `buf` should not be freed.
+    Python object owns memory, thus returned char* must not be freed.
     `length` can be NULL if getting buffer length is not needed.
-    This function should only raise exceptions in out-of-memory cases.
 
     Parameters
     ----------
@@ -265,20 +262,10 @@ cdef inline const char* get_string_data(object s, Py_ssize_t *length):
 
     if PyUnicode_Check(s):
         buf = PyUnicode_AsUTF8AndSize(s, length)
-    if PyBytes_Check(s):
-        if PyBytes_AsStringAndSize(s, <char**>&buf, length) != 0:
-            return NULL
+    else:
+        PyBytes_AsStringAndSize(s, <char**>&buf, length)
     return buf
 
 
-cdef inline const char* get_string_data_checked(object s, Py_ssize_t *length):
-    """
-    This is a wrapper for get_string_data() that raises TypeError
-    when supplied with neither unicode nor bytes object
-    """
-    cdef:
-        const char *buf = get_string_data(s, length)
-
-    if not buf:
-        PyErr_BadArgument()
-    return buf
+cdef inline const char* get_c_string(object s):
+    return get_c_string_buf_and_size(s, NULL)