PERF: added no exception versions of '_string_to_dts' and 'parse_iso_8601_datetime' functions (pandas-dev#26220)

Author: anmyachev

pandas/_libs/tslib.pyx

@@ -204,7 +204,7 @@ def _test_parse_iso8601(object ts):
     elif ts == 'today':
         return Timestamp.now().normalize()
 
-    _string_to_dts(ts, &obj.dts, &out_local, &out_tzoffset)
+    _string_to_dts(ts, &obj.dts, &out_local, &out_tzoffset, True)
     obj.value = dtstruct_to_dt64(&obj.dts)
     check_dts_bounds(&obj.dts)
     if out_local == 1:
@@ -511,6 +511,7 @@ cpdef array_to_datetime(ndarray[object] values, str errors='raise',
         int out_local=0, out_tzoffset=0
         float offset_seconds, tz_offset
         set out_tzoffset_vals = set()
+        bint string_to_dts_failed
 
     # specify error conditions
     assert is_raise or is_ignore or is_coerce
@@ -578,10 +579,12 @@ cpdef array_to_datetime(ndarray[object] values, str errors='raise',
                         iresult[i] = NPY_NAT
                         continue
 
-                    try:
-                        _string_to_dts(val, &dts, &out_local, &out_tzoffset)
-                    except ValueError:
-                        # A ValueError at this point is a _parsing_ error
+                    string_to_dts_failed = _string_to_dts(
+                        val, &dts, &out_local,
+                        &out_tzoffset, False
+                    )
+                    if string_to_dts_failed:
+                        # An error at this point is a _parsing_ error
                         # specifically _not_ OutOfBoundsDatetime
                         if _parse_today_now(val, &iresult[i]):
                             continue
@@ -623,14 +626,8 @@ cpdef array_to_datetime(ndarray[object] values, str errors='raise',
 
                         _ts = convert_datetime_to_tsobject(py_dt, None)
                         iresult[i] = _ts.value
-                    except:
-                        # TODO: What exception are we concerned with here?
-                        if is_coerce:
-                            iresult[i] = NPY_NAT
-                            continue
-                        raise
-                    else:
-                        # No error raised by string_to_dts, pick back up
+                    if not string_to_dts_failed:
+                        # No error reported by string_to_dts, pick back up
                         # where we left off
                         value = dtstruct_to_dt64(&dts)
                         if out_local == 1:

pandas/_libs/tslibs/conversion.pyx

@@ -392,6 +392,44 @@ cdef _TSObject convert_datetime_to_tsobject(datetime ts, object tz,
     return obj
 
 
+cdef _TSObject create_tsobject_tz_using_offset(int64_t value,
+                                               int tzoffset, object tz=None):
+    """
+    Convert a numpy datetime64 `value`, along with initial timezone offset
+    `tzoffset` to a _TSObject (with timezone object `tz` - optional).
+
+    Parameters
+    ----------
+    value: int64_t
+        numpy dt64
+    tzoffset: int
+    tz : tzinfo or None
+        timezone for the timezone-aware output.
+
+    Returns
+    -------
+    obj : _TSObject
+    """
+    cdef:
+        _TSObject obj
+        datetime dt
+
+    tzinfo = pytz.FixedOffset(tzoffset)
+    value = tz_convert_single(value, tzinfo, UTC)
+    obj = convert_to_tsobject(value, tzinfo, None, 0, 0)
+    if tz is None:
+        check_overflows(obj)
+        return obj
+
+    # Keep the converter same as PyDateTime's
+    dt = datetime(obj.dts.year, obj.dts.month, obj.dts.day,
+                  obj.dts.hour, obj.dts.min, obj.dts.sec,
+                  obj.dts.us, obj.tzinfo)
+    obj = convert_datetime_to_tsobject(
+        dt, tz, nanos=obj.dts.ps // 1000)
+    return obj
+
+
 cdef _TSObject convert_str_to_tsobject(object ts, object tz, object unit,
                                        bint dayfirst=False,
                                        bint yearfirst=False):
@@ -420,15 +458,14 @@ cdef _TSObject convert_str_to_tsobject(object ts, object tz, object unit,
     obj : _TSObject
     """
     cdef:
-        _TSObject obj
+        npy_datetimestruct dts
+        int64_t value  # numpy dt64
         int out_local = 0, out_tzoffset = 0
-        datetime dt
+        bint do_parse_datetime_string = False
 
     if tz is not None:
         tz = maybe_get_tz(tz)
 
-    obj = _TSObject()
-
     assert isinstance(ts, str)
 
     if len(ts) == 0 or ts in nat_strings:
@@ -443,34 +480,23 @@ cdef _TSObject convert_str_to_tsobject(object ts, object tz, object unit,
         ts = datetime.now(tz)
         # equiv: datetime.today().replace(tzinfo=tz)
     else:
+        string_to_dts_failed = _string_to_dts(
+            ts, &dts, &out_local,
+            &out_tzoffset, False
+        )
         try:
-            _string_to_dts(ts, &obj.dts, &out_local, &out_tzoffset)
-            obj.value = dtstruct_to_dt64(&obj.dts)
-            check_dts_bounds(&obj.dts)
-            if out_local == 1:
-                obj.tzinfo = pytz.FixedOffset(out_tzoffset)
-                obj.value = tz_convert_single(obj.value, obj.tzinfo, UTC)
-                if tz is None:
-                    check_dts_bounds(&obj.dts)
-                    check_overflows(obj)
-                    return obj
+            if not string_to_dts_failed:
+                check_dts_bounds(&dts)
+                value = dtstruct_to_dt64(&dts)
+                if out_local == 1:
+                    return create_tsobject_tz_using_offset(value,
+                                                           out_tzoffset, tz)
                 else:
-                    # Keep the converter same as PyDateTime's
-                    obj = convert_to_tsobject(obj.value, obj.tzinfo,
-                                              None, 0, 0)
-                    dt = datetime(obj.dts.year, obj.dts.month, obj.dts.day,
-                                  obj.dts.hour, obj.dts.min, obj.dts.sec,
-                                  obj.dts.us, obj.tzinfo)
-                    obj = convert_datetime_to_tsobject(
-                        dt, tz, nanos=obj.dts.ps // 1000)
-                    return obj
-
-            else:
-                ts = obj.value
-                if tz is not None:
-                    # shift for localize_tso
-                    ts = tz_localize_to_utc(np.array([ts], dtype='i8'), tz,
-                                            ambiguous='raise')[0]
+                    ts = value
+                    if tz is not None:
+                        # shift for localize_tso
+                        ts = tz_localize_to_utc(np.array([ts], dtype='i8'), tz,
+                                                ambiguous='raise')[0]
 
         except OutOfBoundsDatetime:
             # GH#19382 for just-barely-OutOfBounds falling back to dateutil
@@ -479,6 +505,9 @@ cdef _TSObject convert_str_to_tsobject(object ts, object tz, object unit,
             raise
 
         except ValueError:
+            do_parse_datetime_string = True
+
+        if string_to_dts_failed or do_parse_datetime_string:
             try:
                 ts = parse_datetime_string(ts, dayfirst=dayfirst,
                                            yearfirst=yearfirst)

pandas/_libs/tslibs/np_datetime.pxd

@@ -73,4 +73,5 @@ cdef npy_timedelta get_timedelta64_value(object obj) nogil
 cdef NPY_DATETIMEUNIT get_datetime64_unit(object obj) nogil
 
 cdef int _string_to_dts(object val, npy_datetimestruct* dts,
-                        int* out_local, int* out_tzoffset) except? -1
+                        int* out_local, int* out_tzoffset,
+                        bint want_exc) except? -1

pandas/_libs/tslibs/np_datetime.pyx

@@ -31,7 +31,7 @@ cdef extern from "src/datetime/np_datetime.h":
     npy_datetimestruct _NS_MIN_DTS, _NS_MAX_DTS
 
 cdef extern from "src/datetime/np_datetime_strings.h":
-    int parse_iso_8601_datetime(const char *str, int len,
+    int parse_iso_8601_datetime(const char *str, int len, int want_exc,
                                 npy_datetimestruct *out,
                                 int *out_local, int *out_tzoffset)
 
@@ -170,11 +170,12 @@ cdef inline int64_t pydate_to_dt64(date val, npy_datetimestruct *dts):
 
 
 cdef inline int _string_to_dts(object val, npy_datetimestruct* dts,
-                               int* out_local, int* out_tzoffset) except? -1:
+                               int* out_local, int* out_tzoffset,
+                               bint want_exc) except? -1:
     cdef:
         Py_ssize_t length
         const char* buf
 
     buf = get_c_string_buf_and_size(val, &length)
-    return parse_iso_8601_datetime(buf, length,
+    return parse_iso_8601_datetime(buf, length, want_exc,
                                    dts, out_local, out_tzoffset)

pandas/_libs/tslibs/src/datetime/np_datetime_strings.c

@@ -66,7 +66,7 @@ This file implements string parsing and creation for NumPy datetime.
  *
  * Returns 0 on success, -1 on failure.
  */
-int parse_iso_8601_datetime(const char *str, int len,
+int parse_iso_8601_datetime(const char *str, int len, int want_exc,
                             npy_datetimestruct *out,
                             int *out_local, int *out_tzoffset) {
     int year_leap = 0;
@@ -173,8 +173,10 @@ int parse_iso_8601_datetime(const char *str, int len,
         goto parse_error;
     }
     if (out->month < 1 || out->month > 12) {
-        PyErr_Format(PyExc_ValueError,
-                     "Month out of range in datetime string \"%s\"", str);
+        if (want_exc) {
+            PyErr_Format(PyExc_ValueError,
+                        "Month out of range in datetime string \"%s\"", str);
+        }
         goto error;
     }
 
@@ -217,8 +219,10 @@ int parse_iso_8601_datetime(const char *str, int len,
     }
     if (out->day < 1 ||
         out->day > days_per_month_table[year_leap][out->month - 1]) {
-        PyErr_Format(PyExc_ValueError,
-                     "Day out of range in datetime string \"%s\"", str);
+        if (want_exc) {
+            PyErr_Format(PyExc_ValueError,
+                        "Day out of range in datetime string \"%s\"", str);
+        }
         goto error;
     }
 
@@ -251,8 +255,11 @@ int parse_iso_8601_datetime(const char *str, int len,
         ++substr;
         --sublen;
         if (out->hour >= 24) {
-            PyErr_Format(PyExc_ValueError,
-                         "Hours out of range in datetime string \"%s\"", str);
+            if (want_exc) {
+                PyErr_Format(PyExc_ValueError,
+                             "Hours out of range in datetime string \"%s\"",
+                             str);
+            }
             goto error;
         }
     }
@@ -291,8 +298,11 @@ int parse_iso_8601_datetime(const char *str, int len,
         ++substr;
         --sublen;
         if (out->min >= 60) {
-            PyErr_Format(PyExc_ValueError,
-                         "Minutes out of range in datetime string \"%s\"", str);
+            if (want_exc) {
+                PyErr_Format(PyExc_ValueError,
+                             "Minutes out of range in datetime string \"%s\"",
+                             str);
+            }
             goto error;
         }
     } else if (!has_hms_sep) {
@@ -328,8 +338,11 @@ int parse_iso_8601_datetime(const char *str, int len,
         ++substr;
         --sublen;
         if (out->sec >= 60) {
-            PyErr_Format(PyExc_ValueError,
-                         "Seconds out of range in datetime string \"%s\"", str);
+            if (want_exc) {
+                PyErr_Format(PyExc_ValueError,
+                             "Seconds out of range in datetime string \"%s\"",
+                             str);
+            }
             goto error;
         }
     } else if (!has_hms_sep) {
@@ -438,10 +451,12 @@ int parse_iso_8601_datetime(const char *str, int len,
             substr += 2;
             sublen -= 2;
             if (offset_hour >= 24) {
-                PyErr_Format(PyExc_ValueError,
-                             "Timezone hours offset out of range "
-                             "in datetime string \"%s\"",
-                             str);
+                if (want_exc) {
+                    PyErr_Format(PyExc_ValueError,
+                                "Timezone hours offset out of range "
+                                "in datetime string \"%s\"",
+                                str);
+                }
                 goto error;
             }
         } else if (sublen >= 1 && isdigit(substr[0])) {
@@ -466,10 +481,12 @@ int parse_iso_8601_datetime(const char *str, int len,
                 substr += 2;
                 sublen -= 2;
                 if (offset_minute >= 60) {
-                    PyErr_Format(PyExc_ValueError,
-                                 "Timezone minutes offset out of range "
-                                 "in datetime string \"%s\"",
-                                 str);
+                    if (want_exc) {
+                        PyErr_Format(PyExc_ValueError,
+                                    "Timezone minutes offset out of range "
+                                    "in datetime string \"%s\"",
+                                    str);
+                    }
                     goto error;
                 }
             } else if (sublen >= 1 && isdigit(substr[0])) {
@@ -507,9 +524,11 @@ int parse_iso_8601_datetime(const char *str, int len,
     return 0;
 
 parse_error:
-    PyErr_Format(PyExc_ValueError,
-                 "Error parsing datetime string \"%s\" at position %d", str,
-                 (int)(substr - str));
+    if (want_exc) {
+        PyErr_Format(PyExc_ValueError,
+                    "Error parsing datetime string \"%s\" at position %d", str,
+                    (int)(substr - str));
+    }
     return -1;
 
 error:

pandas/_libs/tslibs/src/datetime/np_datetime_strings.h

@@ -54,7 +54,7 @@ This file implements string parsing and creation for NumPy datetime.
  * Returns 0 on success, -1 on failure.
  */
 int
-parse_iso_8601_datetime(const char *str, int len,
+parse_iso_8601_datetime(const char *str, int len, int want_exc,
                         npy_datetimestruct *out,
                         int *out_local,
                         int *out_tzoffset);
@@ -79,5 +79,4 @@ get_datetime_iso_8601_strlen(int local, NPY_DATETIMEUNIT base);
 int
 make_iso_8601_datetime(npy_datetimestruct *dts, char *outstr, int outlen,
                        NPY_DATETIMEUNIT base);
-
 #endif  // PANDAS__LIBS_TSLIBS_SRC_DATETIME_NP_DATETIME_STRINGS_H_