diff options
Diffstat (limited to 'source/blender/blenlib/intern/string_search.cc')
-rw-r--r-- | source/blender/blenlib/intern/string_search.cc | 23 |
1 files changed, 11 insertions, 12 deletions
diff --git a/source/blender/blenlib/intern/string_search.cc b/source/blender/blenlib/intern/string_search.cc index 14d85b99739..f87304863c9 100644 --- a/source/blender/blenlib/intern/string_search.cc +++ b/source/blender/blenlib/intern/string_search.cc @@ -11,14 +11,14 @@ #include "BLI_timeit.hh" /* Right arrow, keep in sync with #UI_MENU_ARROW_SEP in `UI_interface.h`. */ -#define UI_MENU_ARROW_SEP "\xe2\x96\xb6" -#define UI_MENU_ARROW_SEP_UNICODE 0x25b6 +#define UI_MENU_ARROW_SEP "\xe2\x96\xb8" +#define UI_MENU_ARROW_SEP_UNICODE 0x25b8 namespace blender::string_search { static int64_t count_utf8_code_points(StringRef str) { - return static_cast<int64_t>(BLI_strnlen_utf8(str.data(), static_cast<size_t>(str.size()))); + return int64_t(BLI_strnlen_utf8(str.data(), size_t(str.size()))); } int damerau_levenshtein_distance(StringRef a, StringRef b) @@ -205,7 +205,7 @@ static bool match_word_initials(StringRef query, StringRef word = words[word_index]; /* Try to match the current character with the current word. */ - if (static_cast<int>(char_index) < word.size()) { + if (int(char_index) < word.size()) { const uint32_t char_unicode = BLI_str_utf8_as_unicode_step( word.data(), word.size(), &char_index); if (query_unicode == char_unicode) { @@ -345,7 +345,7 @@ void extract_normalized_words(StringRef str, LinearAllocator<> &allocator, Vector<StringRef, 64> &r_words) { - const uint32_t unicode_space = (uint32_t)' '; + const uint32_t unicode_space = uint32_t(' '); const uint32_t unicode_right_triangle = UI_MENU_ARROW_SEP_UNICODE; BLI_assert(unicode_space == BLI_str_utf8_as_unicode(" ")); @@ -358,7 +358,7 @@ void extract_normalized_words(StringRef str, /* Make a copy of the string so that we can edit it. */ StringRef str_copy = allocator.copy_string(str); char *mutable_copy = const_cast<char *>(str_copy.data()); - const size_t str_size_in_bytes = static_cast<size_t>(str.size()); + const size_t str_size_in_bytes = size_t(str.size()); BLI_str_tolower_ascii(mutable_copy, str_size_in_bytes); /* Iterate over all unicode code points to split individual words. */ @@ -371,8 +371,7 @@ void extract_normalized_words(StringRef str, size -= offset; if (is_separator(unicode)) { if (is_in_word) { - r_words.append( - str_copy.substr(static_cast<int>(word_start), static_cast<int>(offset - word_start))); + r_words.append(str_copy.substr(int(word_start), int(offset - word_start))); is_in_word = false; } } @@ -386,7 +385,7 @@ void extract_normalized_words(StringRef str, } /* If the last word is not followed by a separator, it has to be handled separately. */ if (is_in_word) { - r_words.append(str_copy.drop_prefix(static_cast<int>(word_start))); + r_words.append(str_copy.drop_prefix(int(word_start))); } } @@ -419,7 +418,7 @@ void BLI_string_search_add(StringSearch *search, StringRef str_ref{str}; string_search::extract_normalized_words(str_ref, search->allocator, words); search->items.append({search->allocator.construct_array_copy(words.as_span()), - (int)str_ref.size(), + int(str_ref.size()), user_data, weight}); } @@ -458,7 +457,7 @@ int BLI_string_search_query(StringSearch *search, const char *query, void ***r_d if (score == found_scores[0] && !query_str.is_empty()) { /* Sort items with best score by length. Shorter items are more likely the ones you are * looking for. This also ensures that exact matches will be at the top, even if the query is - * a substring of another item. */ + * a sub-string of another item. */ std::sort(indices.begin(), indices.end(), [&](int a, int b) { return search->items[a].length < search->items[b].length; }); @@ -472,7 +471,7 @@ int BLI_string_search_query(StringSearch *search, const char *query, void ***r_d } void **sorted_data = static_cast<void **>( - MEM_malloc_arrayN(static_cast<size_t>(sorted_result_indices.size()), sizeof(void *), AT)); + MEM_malloc_arrayN(size_t(sorted_result_indices.size()), sizeof(void *), AT)); for (const int i : sorted_result_indices.index_range()) { const int result_index = sorted_result_indices[i]; SearchItem &item = search->items[result_index]; |