refactor(clp-s): Replace clp_s::StringUtils function calls with clp::string_utils equivalents and remove redundant implementations. (#1103)

Author: gibber9809

components/core/cmake/Options/options.cmake

@@ -198,6 +198,7 @@ endfunction()
 
 function(validate_clp_s_archivereader_dependencies)
     validate_clp_dependencies_for_target(CLP_BUILD_CLP_S_ARCHIVEREADER
+        CLP_BUILD_CLP_STRING_UTILS
         CLP_BUILD_CLP_S_CLP_DEPENDENCIES
         CLP_BUILD_CLP_S_IO
         CLP_BUILD_CLP_S_TIMESTAMPPATTERN
@@ -300,6 +301,7 @@ endfunction()
 
 function(validate_clp_s_search_dependencies)
     validate_clp_dependencies_for_target(CLP_BUILD_CLP_S_SEARCH
+        CLP_BUILD_CLP_STRING_UTILS
         CLP_BUILD_CLP_S_ARCHIVEREADER
         CLP_BUILD_CLP_S_CLP_DEPENDENCIES
         CLP_BUILD_CLP_S_SEARCH_AST

components/core/src/clp_s/CMakeLists.txt

@@ -302,6 +302,7 @@ if(CLP_BUILD_CLP_S_ARCHIVEREADER)
                 clp_s_archive_reader
                 PUBLIC
                 absl::flat_hash_map
+                clp::string_utils
                 clp_s::io
                 msgpack-cxx
                 nlohmann_json::nlohmann_json

components/core/src/clp_s/DictionaryReader.hpp

@@ -6,6 +6,7 @@
 #include <unordered_set>
 
 #include <boost/algorithm/string/case_conv.hpp>
+#include <string_utils/string_utils.hpp>
 
 #include "ArchiveReaderAdaptor.hpp"
 #include "DictionaryEntry.hpp"
@@ -191,7 +192,12 @@ void DictionaryReader<DictionaryIdType, EntryType>::get_entries_matching_wildcar
         std::unordered_set<EntryType const*>& entries
 ) const {
     for (auto const& entry : m_entries) {
-        if (StringUtils::wildcard_match_unsafe(entry.get_value(), wildcard_string, !ignore_case)) {
+        if (clp::string_utils::wildcard_match_unsafe(
+                    entry.get_value(),
+                    wildcard_string,
+                    !ignore_case
+            ))
+        {
             entries.insert(&entry);
         }
     }

components/core/src/clp_s/Utils.cpp

@@ -9,6 +9,7 @@
 #include <boost/url.hpp>
 #include <fmt/core.h>
 #include <spdlog/spdlog.h>
+#include <string_utils/string_utils.hpp>
 
 #include "archive_constants.hpp"
 
@@ -188,9 +189,9 @@ bool StringUtils::get_bounds_of_next_var(string const& msg, size_t& begin_pos, s
         end_pos = begin_pos;
         for (; end_pos < msg_length; ++end_pos) {
             char c = msg[end_pos];
-            if (is_decimal_digit(c)) {
+            if (clp::string_utils::is_decimal_digit(c)) {
                 contains_decimal_digit = true;
-            } else if (is_alphabet(c)) {
+            } else if (clp::string_utils::is_alphabet(c)) {
                 contains_alphabet = true;
             } else if (is_delim(c)) {
                 break;
@@ -212,308 +213,6 @@ bool StringUtils::get_bounds_of_next_var(string const& msg, size_t& begin_pos, s
     return (msg_length != begin_pos);
 }
 
-size_t StringUtils::find_first_of(
-        string const& haystack,
-        char const* needles,
-        size_t search_start_pos,
-        size_t& needle_ix
-) {
-    size_t haystack_length = haystack.length();
-    size_t needles_length = strlen(needles);
-    for (size_t i = search_start_pos; i < haystack_length; ++i) {
-        for (needle_ix = 0; needle_ix < needles_length; ++needle_ix) {
-            if (haystack[i] == needles[needle_ix]) {
-                return i;
-            }
-        }
-    }
-
-    return string::npos;
-}
-
-string StringUtils::replace_characters(
-        char const* characters_to_escape,
-        char const* replacement_characters,
-        string const& value,
-        bool escape
-) {
-    string new_value;
-    size_t search_start_pos = 0;
-    while (true) {
-        size_t replace_char_ix;
-        size_t char_to_replace_pos
-                = find_first_of(value, characters_to_escape, search_start_pos, replace_char_ix);
-        if (string::npos == char_to_replace_pos) {
-            new_value.append(value, search_start_pos, string::npos);
-            break;
-        } else {
-            new_value.append(value, search_start_pos, char_to_replace_pos - search_start_pos);
-            if (escape) {
-                new_value += "\\";
-            }
-            new_value += replacement_characters[replace_char_ix];
-            search_start_pos = char_to_replace_pos + 1;
-        }
-    }
-    return new_value;
-}
-
-void StringUtils::to_lower(string& str) {
-    std::transform(str.cbegin(), str.cend(), str.begin(), [](unsigned char c) {
-        return std::tolower(c);
-    });
-}
-
-bool StringUtils::is_wildcard(char c) {
-    static constexpr char cWildcards[] = "?*";
-    for (size_t i = 0; i < strlen(cWildcards); ++i) {
-        if (cWildcards[i] == c) {
-            return true;
-        }
-    }
-    return false;
-}
-
-string StringUtils::clean_up_wildcard_search_string(string_view str) {
-    string cleaned_str;
-
-    bool is_escaped = false;
-    auto str_end = str.cend();
-    for (auto current = str.cbegin(); current != str_end;) {
-        auto c = *current;
-        if (is_escaped) {
-            is_escaped = false;
-
-            if (is_wildcard(c) || '\\' == c) {
-                // Keep escaping if c is a wildcard character or an escape character
-                cleaned_str += '\\';
-            }
-            cleaned_str += c;
-            ++current;
-        } else if ('*' == c) {
-            cleaned_str += c;
-
-            // Skip over all '*' to find the next non-'*'
-            do {
-                ++current;
-            } while (current != str_end && '*' == *current);
-        } else {
-            if ('\\' == c) {
-                is_escaped = true;
-            } else {
-                cleaned_str += c;
-            }
-            ++current;
-        }
-    }
-
-    return cleaned_str;
-}
-
-bool StringUtils::advance_tame_to_next_match(
-        char const*& tame_current,
-        char const*& tame_bookmark,
-        char const* tame_end,
-        char const*& wild_current,
-        char const*& wild_bookmark
-) {
-    auto w = *wild_current;
-    if ('?' != w) {
-        // No need to check for '*' since the caller ensures wild doesn't
-        // contain consecutive '*'
-
-        // Handle escaped characters
-        if ('\\' == w) {
-            ++wild_current;
-            // This is safe without a bounds check since this the caller
-            // ensures there are no dangling escape characters
-            w = *wild_current;
-        }
-
-        // Advance tame_current until it matches wild_current
-        while (true) {
-            if (tame_end == tame_current) {
-                // Wild group is longer than last group in tame, so
-                // can't match
-                // e.g. "*abc" doesn't match "zab"
-                return false;
-            }
-            auto t = *tame_current;
-            if (t == w) {
-                break;
-            }
-            ++tame_current;
-        }
-    }
-
-    tame_bookmark = tame_current;
-
-    return true;
-}
-
-bool
-StringUtils::wildcard_match_unsafe(string_view tame, string_view wild, bool case_sensitive_match) {
-    if (case_sensitive_match) {
-        return wildcard_match_unsafe_case_sensitive(tame, wild);
-    } else {
-        // We convert to lowercase (rather than uppercase) anticipating that
-        // callers use lowercase more frequently, so little will need to change.
-        string lowercase_tame(tame);
-        to_lower(lowercase_tame);
-        string lowercase_wild(wild);
-        to_lower(lowercase_wild);
-        return wildcard_match_unsafe_case_sensitive(lowercase_tame, lowercase_wild);
-    }
-}
-
-/**
- * The algorithm basically works as follows:
- * Given a wild string "*abc*def*ghi*", it can be broken into groups of
- * characters delimited by one or more '*' characters. The goal of the
- * algorithm is then to determine whether the tame string contains each of
- * those groups in the same order.
- *
- * Thus, the algorithm:
- * 1. searches for the start of one of these groups in wild,
- * 2. searches for a group in tame starting with the same character, and then
- * 3. checks if the two match. If not, the search repeats with the next group in
- *    tame.
- */
-bool StringUtils::wildcard_match_unsafe_case_sensitive(string_view tame, string_view wild) {
-    auto const tame_length = tame.length();
-    auto const wild_length = wild.length();
-    char const* tame_current = tame.data();
-    char const* wild_current = wild.data();
-    char const* tame_bookmark = nullptr;
-    char const* wild_bookmark = nullptr;
-    char const* tame_end = tame_current + tame_length;
-    char const* wild_end = wild_current + wild_length;
-
-    // Handle wild or tame being empty
-    if (0 == wild_length) {
-        return 0 == tame_length;
-    } else {
-        if (0 == tame_length) {
-            return "*" == wild;
-        }
-    }
-
-    char w;
-    char t;
-    bool is_escaped = false;
-    while (true) {
-        w = *wild_current;
-        if ('*' == w) {
-            ++wild_current;
-            if (wild_end == wild_current) {
-                // Trailing '*' means everything remaining in tame will match
-                return true;
-            }
-
-            // Set wild and tame bookmarks
-            wild_bookmark = wild_current;
-            if (!advance_tame_to_next_match(
-                        tame_current,
-                        tame_bookmark,
-                        tame_end,
-                        wild_current,
-                        wild_bookmark
-                ))
-            {
-                return false;
-            }
-        } else {
-            // Handle escaped characters
-            if ('\\' == w) {
-                is_escaped = true;
-                ++wild_current;
-                // This is safe without a bounds check since this the caller
-                // ensures there are no dangling escape characters
-                w = *wild_current;
-            }
-
-            // Handle a mismatch
-            t = *tame_current;
-            if (false == ((false == is_escaped && '?' == w) || t == w)) {
-                if (nullptr == wild_bookmark) {
-                    // No bookmark to return to
-                    return false;
-                }
-
-                wild_current = wild_bookmark;
-                tame_current = tame_bookmark + 1;
-                if (!advance_tame_to_next_match(
-                            tame_current,
-                            tame_bookmark,
-                            tame_end,
-                            wild_current,
-                            wild_bookmark
-                    ))
-                {
-                    return false;
-                }
-            }
-        }
-
-        ++tame_current;
-        ++wild_current;
-
-        // Handle reaching the end of tame or wild
-        if (tame_end == tame_current) {
-            return (wild_end == wild_current
-                    || ('*' == *wild_current && (wild_current + 1) == wild_end));
-        } else {
-            if (wild_end == wild_current) {
-                if (nullptr == wild_bookmark) {
-                    // No bookmark to return to
-                    return false;
-                } else {
-                    wild_current = wild_bookmark;
-                    tame_current = tame_bookmark + 1;
-                    if (!advance_tame_to_next_match(
-                                tame_current,
-                                tame_bookmark,
-                                tame_end,
-                                wild_current,
-                                wild_bookmark
-                        ))
-                    {
-                        return false;
-                    }
-                }
-            }
-        }
-    }
-}
-
-bool StringUtils::convert_string_to_int64(std::string_view raw, int64_t& converted) {
-    auto raw_end = raw.cend();
-    auto result = std::from_chars(raw.cbegin(), raw_end, converted);
-    if (raw_end != result.ptr) {
-        return false;
-    } else {
-        return result.ec == std::errc();
-    }
-}
-
-bool StringUtils::convert_string_to_double(std::string const& raw, double& converted) {
-    if (raw.empty()) {
-        // Can't convert an empty string
-        return false;
-    }
-
-    char const* c_str = raw.c_str();
-    char* end_ptr;
-    // Reset errno so we can detect a new error
-    errno = 0;
-    double raw_as_double = strtod(c_str, &end_ptr);
-    if (ERANGE == errno || (end_ptr - c_str) < raw.length()) {
-        return false;
-    }
-    converted = raw_as_double;
-    return true;
-}
-
 void StringUtils::escape_json_string(std::string& destination, std::string_view const source) {
     // Escaping is implemented using this `append_unescaped_slice` approach to offer a fast path
     // when strings are mostly or entirely valid escaped JSON. Benchmarking shows that this offers