Add except declaration in Cython interface for regex_program::create (#…

…13054)

Add the `except +` declaration to the `cudf::strings::regex_program::create()` function in the Cython `regex_program.pxd` interface since invalid regex patterns are thrown by this call. This allows the normal Cython exception handling to pass the exception to the Python logic without aborting the process.

Closes #13052

Authors:
  - David Wendt (https://github.com/davidwendt)

Approvers:
  - Bradley Dice (https://github.com/bdice)
  - Ashwin Srinath (https://github.com/shwina)

URL: #13054

python/cudf/cudf/_lib/cpp/strings/regex_program.pxd

@@ -1,4 +1,4 @@
-# Copyright (c) 2022, NVIDIA CORPORATION.
+# Copyright (c) 2022-2023, NVIDIA CORPORATION.
 
 from libcpp.memory cimport unique_ptr
 from libcpp.string cimport string
@@ -12,4 +12,7 @@ cdef extern from "cudf/strings/regex/regex_program.hpp" \
     cdef cppclass regex_program:
 
         @staticmethod
-        unique_ptr[regex_program] create(string pattern, regex_flags flags)
+        unique_ptr[regex_program] create(
+            string pattern,
+            regex_flags flags
+        ) except +

python/cudf/cudf/_lib/strings/contains.pyx

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2023, NVIDIA CORPORATION.
 
 from cython.operator cimport dereference
 from libc.stdint cimport uint32_t
@@ -35,10 +35,10 @@ def contains_re(Column source_strings, object reg_ex, uint32_t flags):
 
     cdef string reg_ex_string = <string>str(reg_ex).encode()
     cdef regex_flags c_flags = <regex_flags>flags
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(reg_ex_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(reg_ex_string, c_flags))
         c_result = move(cpp_contains_re(
             source_view,
             dereference(c_prog)
@@ -58,10 +58,10 @@ def count_re(Column source_strings, object reg_ex, uint32_t flags):
 
     cdef string reg_ex_string = <string>str(reg_ex).encode()
     cdef regex_flags c_flags = <regex_flags>flags
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(reg_ex_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(reg_ex_string, c_flags))
         c_result = move(cpp_count_re(
             source_view,
             dereference(c_prog)
@@ -81,10 +81,10 @@ def match_re(Column source_strings, object reg_ex, uint32_t flags):
 
     cdef string reg_ex_string = <string>str(reg_ex).encode()
     cdef regex_flags c_flags = <regex_flags>flags
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(reg_ex_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(reg_ex_string, c_flags))
         c_result = move(cpp_matches_re(
             source_view,
             dereference(c_prog)

python/cudf/cudf/_lib/strings/extract.pyx

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2023, NVIDIA CORPORATION.
 
 from cython.operator cimport dereference
 from libc.stdint cimport uint32_t
@@ -30,10 +30,10 @@ def extract(Column source_strings, object pattern, uint32_t flags):
 
     cdef string pattern_string = <string>str(pattern).encode()
     cdef regex_flags c_flags = <regex_flags>flags
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_extract(
             source_view,
             dereference(c_prog)

python/cudf/cudf/_lib/strings/findall.pyx

@@ -1,4 +1,4 @@
-# Copyright (c) 2019-2022, NVIDIA CORPORATION.
+# Copyright (c) 2019-2023, NVIDIA CORPORATION.
 
 from cython.operator cimport dereference
 from libc.stdint cimport uint32_t
@@ -27,10 +27,10 @@ def findall(Column source_strings, object pattern, uint32_t flags):
 
     cdef string pattern_string = <string>str(pattern).encode()
     cdef regex_flags c_flags = <regex_flags>flags
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_findall(
             source_view,
             dereference(c_prog)

python/cudf/cudf/_lib/strings/replace_re.pyx

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2023, NVIDIA CORPORATION.
 
 from cython.operator cimport dereference
 from libcpp.memory cimport unique_ptr
@@ -43,10 +43,10 @@ def replace_re(Column source_strings,
     cdef const string_scalar* scalar_repl = \
         <const string_scalar*>(repl.get_raw_ptr())
     cdef regex_flags c_flags = regex_flags.DEFAULT
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_replace_re(
             source_view,
             dereference(c_prog),
@@ -73,10 +73,10 @@ def replace_with_backrefs(
     cdef string pattern_string = <string>str(pattern).encode()
     cdef string repl_string = <string>str(repl).encode()
     cdef regex_flags c_flags = regex_flags.DEFAULT
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_replace_with_backrefs(
             source_view,
             dereference(c_prog),

python/cudf/cudf/_lib/strings/split/split.pyx

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2023, NVIDIA CORPORATION.
 
 from cython.operator cimport dereference
 from libcpp.memory cimport unique_ptr
@@ -163,10 +163,10 @@ def split_re(Column source_strings,
     cdef column_view source_view = source_strings.view()
     cdef string pattern_string = <string>str(pattern).encode()
     cdef regex_flags c_flags = regex_flags.DEFAULT
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_split_re(
             source_view,
             dereference(c_prog),
@@ -192,10 +192,10 @@ def rsplit_re(Column source_strings,
     cdef column_view source_view = source_strings.view()
     cdef string pattern_string = <string>str(pattern).encode()
     cdef regex_flags c_flags = regex_flags.DEFAULT
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_rsplit_re(
             source_view,
             dereference(c_prog),
@@ -220,10 +220,10 @@ def split_record_re(Column source_strings,
     cdef column_view source_view = source_strings.view()
     cdef string pattern_string = <string>str(pattern).encode()
     cdef regex_flags c_flags = regex_flags.DEFAULT
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_split_record_re(
             source_view,
             dereference(c_prog),
@@ -248,10 +248,10 @@ def rsplit_record_re(Column source_strings,
     cdef column_view source_view = source_strings.view()
     cdef string pattern_string = <string>str(pattern).encode()
     cdef regex_flags c_flags = regex_flags.DEFAULT
-    cdef unique_ptr[regex_program] c_prog = \
-        regex_program.create(pattern_string, c_flags)
+    cdef unique_ptr[regex_program] c_prog
 
     with nogil:
+        c_prog = move(regex_program.create(pattern_string, c_flags))
         c_result = move(cpp_rsplit_record_re(
             source_view,
             dereference(c_prog),

python/cudf/cudf/tests/test_string.py

@@ -784,6 +784,12 @@ def test_string_extract(ps_gs, pat, expand, flags, flags_raise):
         assert_eq(expect, got)
 
 
+def test_string_invalid_regex():
+    gs = cudf.Series(["a"])
+    with pytest.raises(RuntimeError):
+        gs.str.extract(r"{\}")
+
+
 @pytest.mark.parametrize(
     "pat,regex",
     [