Remove cudf._lib.string.convert/split in favor of inlining pylibcudf

python/cudf/cudf/_lib/CMakeLists.txt

@@ -50,4 +50,3 @@ target_link_libraries(interop PUBLIC nanoarrow)
 
 add_subdirectory(io)
 add_subdirectory(nvtext)
-add_subdirectory(strings)

python/cudf/cudf/_lib/__init__.py

@@ -14,7 +14,6 @@
     sort,
     stream_compaction,
     string_casting,
-    strings,
     strings_udf,
     text,
 )

python/cudf/cudf/_lib/strings/__init__.py

@@ -32,18 +32,3 @@
     detokenize,
     tokenize_with_vocabulary,
 )
-from cudf._lib.strings.convert.convert_fixed_point import to_decimal
-from cudf._lib.strings.convert.convert_floats import is_float
-from cudf._lib.strings.convert.convert_integers import is_integer
-from cudf._lib.strings.convert.convert_urls import url_decode, url_encode
-from cudf._lib.strings.split.partition import partition, rpartition
-from cudf._lib.strings.split.split import (
-    rsplit,
-    rsplit_re,
-    rsplit_record,
-    rsplit_record_re,
-    split,
-    split_re,
-    split_record,
-    split_record_re,
-)