Change pattern parameter for regex APIs from std::string to std::string_view

cpp/include/cudf/strings/contains.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2022, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -51,7 +51,7 @@ namespace strings {
  */
 std::unique_ptr<column> contains_re(
   strings_column_view const& strings,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -78,7 +78,7 @@ std::unique_ptr<column> contains_re(
  */
 std::unique_ptr<column> matches_re(
   strings_column_view const& strings,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -105,7 +105,7 @@ std::unique_ptr<column> matches_re(
  */
 std::unique_ptr<column> count_re(
   strings_column_view const& strings,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 

cpp/include/cudf/strings/extract.hpp

@@ -55,7 +55,7 @@ namespace strings {
  */
 std::unique_ptr<table> extract(
   strings_column_view const& strings,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -90,7 +90,7 @@ std::unique_ptr<table> extract(
  */
 std::unique_ptr<column> extract_all_record(
   strings_column_view const& strings,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 

cpp/include/cudf/strings/findall.hpp

@@ -56,7 +56,7 @@ namespace strings {
  */
 std::unique_ptr<table> findall(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -90,7 +90,7 @@ std::unique_ptr<table> findall(
  */
 std::unique_ptr<column> findall_record(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 

cpp/include/cudf/strings/replace_re.hpp

@@ -50,7 +50,7 @@ namespace strings {
  */
 std::unique_ptr<column> replace_re(
   strings_column_view const& strings,
-  std::string const& pattern,
+  std::string_view pattern,
   string_scalar const& replacement           = string_scalar(""),
   std::optional<size_type> max_replace_count = std::nullopt,
   regex_flags const flags                    = regex_flags::DEFAULT,
@@ -98,8 +98,8 @@ std::unique_ptr<column> replace_re(
  */
 std::unique_ptr<column> replace_with_backrefs(
   strings_column_view const& strings,
-  std::string const& pattern,
-  std::string const& replacement,
+  std::string_view pattern,
+  std::string_view replacement,
   regex_flags const flags             = regex_flags::DEFAULT,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 

cpp/include/cudf/strings/split/split_re.hpp

@@ -71,7 +71,7 @@ namespace strings {
  */
 std::unique_ptr<table> split_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   size_type maxsplit                  = -1,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -121,7 +121,7 @@ std::unique_ptr<table> split_re(
  */
 std::unique_ptr<table> rsplit_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   size_type maxsplit                  = -1,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -173,7 +173,7 @@ std::unique_ptr<table> rsplit_re(
  */
 std::unique_ptr<column> split_record_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   size_type maxsplit                  = -1,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
@@ -227,7 +227,7 @@ std::unique_ptr<column> split_record_re(
  */
 std::unique_ptr<column> rsplit_record_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   size_type maxsplit                  = -1,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 

cpp/src/strings/contains.cu

@@ -56,7 +56,7 @@ struct contains_fn {
 };
 
 std::unique_ptr<column> contains_impl(strings_column_view const& input,
-                                      std::string const& pattern,
+                                      std::string_view pattern,
                                       regex_flags const flags,
                                       bool const beginning_only,
                                       rmm::cuda_stream_view stream,
@@ -85,7 +85,7 @@ std::unique_ptr<column> contains_impl(strings_column_view const& input,
 
 std::unique_ptr<column> contains_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags,
   rmm::cuda_stream_view stream,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource())
@@ -95,7 +95,7 @@ std::unique_ptr<column> contains_re(
 
 std::unique_ptr<column> matches_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags,
   rmm::cuda_stream_view stream,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource())
@@ -105,7 +105,7 @@ std::unique_ptr<column> matches_re(
 
 std::unique_ptr<column> count_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags,
   rmm::cuda_stream_view stream,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource())
@@ -128,7 +128,7 @@ std::unique_ptr<column> count_re(
 // external APIs
 
 std::unique_ptr<column> contains_re(strings_column_view const& strings,
-                                    std::string const& pattern,
+                                    std::string_view pattern,
                                     regex_flags const flags,
                                     rmm::mr::device_memory_resource* mr)
 {
@@ -137,7 +137,7 @@ std::unique_ptr<column> contains_re(strings_column_view const& strings,
 }
 
 std::unique_ptr<column> matches_re(strings_column_view const& strings,
-                                   std::string const& pattern,
+                                   std::string_view pattern,
                                    regex_flags const flags,
                                    rmm::mr::device_memory_resource* mr)
 {
@@ -146,7 +146,7 @@ std::unique_ptr<column> matches_re(strings_column_view const& strings,
 }
 
 std::unique_ptr<column> count_re(strings_column_view const& strings,
-                                 std::string const& pattern,
+                                 std::string_view pattern,
                                  regex_flags const flags,
                                  rmm::mr::device_memory_resource* mr)
 {

cpp/src/strings/extract/extract.cu

@@ -85,7 +85,7 @@ struct extract_fn {
 
 //
 std::unique_ptr<table> extract(strings_column_view const& input,
-                               std::string const& pattern,
+                               std::string_view pattern,
                                regex_flags const flags,
                                rmm::cuda_stream_view stream,
                                rmm::mr::device_memory_resource* mr)
@@ -130,7 +130,7 @@ std::unique_ptr<table> extract(strings_column_view const& input,
 // external API
 
 std::unique_ptr<table> extract(strings_column_view const& strings,
-                               std::string const& pattern,
+                               std::string_view pattern,
                                regex_flags const flags,
                                rmm::mr::device_memory_resource* mr)
 {

cpp/src/strings/extract/extract_all.cu

@@ -96,7 +96,7 @@ struct extract_fn {
  */
 std::unique_ptr<column> extract_all_record(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags,
   rmm::cuda_stream_view stream,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource())
@@ -165,7 +165,7 @@ std::unique_ptr<column> extract_all_record(
 // external API
 
 std::unique_ptr<column> extract_all_record(strings_column_view const& strings,
-                                           std::string const& pattern,
+                                           std::string_view pattern,
                                            regex_flags const flags,
                                            rmm::mr::device_memory_resource* mr)
 {

cpp/src/strings/regex/regex.cuh

@@ -88,7 +88,7 @@ class reprog_device {
    * @return The program device object.
    */
   static std::unique_ptr<reprog_device, std::function<void(reprog_device*)>> create(
-    std::string const& pattern, rmm::cuda_stream_view stream);
+    std::string_view pattern, rmm::cuda_stream_view stream);
 
   /**
    * @brief Create the device program instance from a regex pattern.
@@ -99,7 +99,7 @@ class reprog_device {
    * @return The program device object.
    */
   static std::unique_ptr<reprog_device, std::function<void(reprog_device*)>> create(
-    std::string const& pattern, regex_flags const re_flags, rmm::cuda_stream_view stream);
+    std::string_view pattern, regex_flags const re_flags, rmm::cuda_stream_view stream);
 
   /**
    * @brief Called automatically by the unique_ptr returned from create().

cpp/src/strings/regex/regexec.cu

@@ -43,14 +43,14 @@ reprog_device::reprog_device(reprog& prog)
 }
 
 std::unique_ptr<reprog_device, std::function<void(reprog_device*)>> reprog_device::create(
-  std::string const& pattern, rmm::cuda_stream_view stream)
+  std::string_view pattern, rmm::cuda_stream_view stream)
 {
   return reprog_device::create(pattern, regex_flags::MULTILINE, stream);
 }
 
 // Create instance of the reprog that can be passed into a device kernel
 std::unique_ptr<reprog_device, std::function<void(reprog_device*)>> reprog_device::create(
-  std::string const& pattern, regex_flags const flags, rmm::cuda_stream_view stream)
+  std::string_view pattern, regex_flags const flags, rmm::cuda_stream_view stream)
 {
   // compile pattern into host object
   reprog h_prog = reprog::create_from(pattern, flags);

cpp/src/strings/replace/backref_re.cu

@@ -46,13 +46,14 @@ namespace {
  *
  * Reference: https://www.regular-expressions.info/refreplacebackref.html
  */
-std::string get_backref_pattern(std::string const& repl)
+std::string get_backref_pattern(std::string_view repl)
 {
   std::string const backslash_pattern = "\\\\(\\d+)";
   std::string const bracket_pattern   = "\\$\\{(\\d+)\\}";
   std::smatch m;
-  return std::regex_search(repl, m, std::regex(backslash_pattern)) ? backslash_pattern
-                                                                   : bracket_pattern;
+  std::string r{repl};
+  return std::regex_search(r, m, std::regex(backslash_pattern)) ? backslash_pattern
+                                                                : bracket_pattern;
 }
 /**
  * @brief Parse the back-ref index and position values from a given replace format.
@@ -66,11 +67,11 @@ std::string get_backref_pattern(std::string const& repl)
  * For example, for input string 'hello \2 and \1' the returned `backref_type` vector
  * contains `[(2,6),(1,11)]` and the returned string is 'hello  and '.
  */
-std::pair<std::string, std::vector<backref_type>> parse_backrefs(std::string const& repl,
+std::pair<std::string, std::vector<backref_type>> parse_backrefs(std::string_view repl,
                                                                  int const group_count)
 {
   std::vector<backref_type> backrefs;
-  std::string str = repl;  // make a modifiable copy
+  std::string str{repl};  // make a modifiable copy
   std::smatch m;
   std::regex ex(get_backref_pattern(repl));
   std::string rtn;
@@ -100,8 +101,8 @@ std::pair<std::string, std::vector<backref_type>> parse_backrefs(std::string con
 
 //
 std::unique_ptr<column> replace_with_backrefs(strings_column_view const& input,
-                                              std::string const& pattern,
-                                              std::string const& replacement,
+                                              std::string_view pattern,
+                                              std::string_view replacement,
                                               regex_flags const flags,
                                               rmm::cuda_stream_view stream,
                                               rmm::mr::device_memory_resource* mr)
@@ -144,8 +145,8 @@ std::unique_ptr<column> replace_with_backrefs(strings_column_view const& input,
 // external API
 
 std::unique_ptr<column> replace_with_backrefs(strings_column_view const& strings,
-                                              std::string const& pattern,
-                                              std::string const& replacement,
+                                              std::string_view pattern,
+                                              std::string_view replacement,
                                               regex_flags const flags,
                                               rmm::mr::device_memory_resource* mr)
 {

cpp/src/strings/replace/replace_re.cu

@@ -101,7 +101,7 @@ struct replace_regex_fn {
 //
 std::unique_ptr<column> replace_re(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   string_scalar const& replacement,
   std::optional<size_type> max_replace_count,
   regex_flags const flags,
@@ -135,7 +135,7 @@ std::unique_ptr<column> replace_re(
 // external API
 
 std::unique_ptr<column> replace_re(strings_column_view const& strings,
-                                   std::string const& pattern,
+                                   std::string_view pattern,
                                    string_scalar const& replacement,
                                    std::optional<size_type> max_replace_count,
                                    regex_flags const flags,

cpp/src/strings/search/findall.cu

@@ -86,7 +86,7 @@ struct findall_fn {
 }  // namespace
 
 std::unique_ptr<table> findall(strings_column_view const& input,
-                               std::string const& pattern,
+                               std::string_view pattern,
                                regex_flags const flags,
                                rmm::cuda_stream_view stream,
                                rmm::mr::device_memory_resource* mr)
@@ -147,7 +147,7 @@ std::unique_ptr<table> findall(strings_column_view const& input,
 // external API
 
 std::unique_ptr<table> findall(strings_column_view const& input,
-                               std::string const& pattern,
+                               std::string_view pattern,
                                regex_flags const flags,
                                rmm::mr::device_memory_resource* mr)
 {

cpp/src/strings/search/findall_record.cu

@@ -93,7 +93,7 @@ std::unique_ptr<column> findall_util(column_device_view const& d_strings,
 //
 std::unique_ptr<column> findall_record(
   strings_column_view const& input,
-  std::string const& pattern,
+  std::string_view pattern,
   regex_flags const flags,
   rmm::cuda_stream_view stream,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource())
@@ -133,7 +133,7 @@ std::unique_ptr<column> findall_record(
 // external API
 
 std::unique_ptr<column> findall_record(strings_column_view const& input,
-                                       std::string const& pattern,
+                                       std::string_view pattern,
                                        regex_flags const flags,
                                        rmm::mr::device_memory_resource* mr)
 {