Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Use an unordered_map with std::string keys

Browse Source 
This is needed because we use references to values in the map and
std::map invalidates references on insertion while std::unordered_map
does not
This commit is contained in:
Kovid Goyal 2021-06-23 10:37:06 +05:30
parent 78dea8e439
commit 53d0d1c249
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 8 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/calibre/db/sqlite_extension.cpp
View File

@ -11,7 +11,7 @@
#include <string> #include <string>
#include <locale> #include <locale>
#include <vector> #include <vector>
#include <map> #include <unordered_map>
#include <mutex> #include <mutex>
#include <cstring> #include <cstring>
#include <sqlite3ext.h> #include <sqlite3ext.h>
@ -98,13 +98,6 @@ class IteratorDescription {
UScriptCode script; UScriptCode script;
}; };
struct char_cmp {
bool operator () (const char *a, const char *b) const
{
return strcmp(a,b)<0;
}
};
class Stemmer { class Stemmer {
private: private:
struct sb_stemmer *handle; struct sb_stemmer *handle;
@ -142,6 +135,7 @@ public:
typedef std::unique_ptr<icu::BreakIterator> BreakIterator; typedef std::unique_ptr<icu::BreakIterator> BreakIterator;
typedef std::unique_ptr<Stemmer> StemmerPtr; typedef std::unique_ptr<Stemmer> StemmerPtr;
static const std::string empty_string("");
class Tokenizer { class Tokenizer {
private: private:
@ -151,8 +145,8 @@ private:
std::string token_buf, current_ui_language; std::string token_buf, current_ui_language;
token_callback_func current_callback; token_callback_func current_callback;
void *current_callback_ctx; void *current_callback_ctx;
std::map<const char*, BreakIterator, char_cmp> iterators; std::unordered_map<std::string, BreakIterator> iterators;
std::map<const char*, StemmerPtr, char_cmp> stemmers; std::unordered_map<std::string, StemmerPtr> stemmers;
bool is_token_char(UChar32 ch) const { bool is_token_char(UChar32 ch) const {
switch(u_charType(ch)) { switch(u_charType(ch)) {
@ -225,11 +219,11 @@ private:
void ensure_basic_iterator(void) { void ensure_basic_iterator(void) {
std::lock_guard<std::mutex> lock(global_mutex); std::lock_guard<std::mutex> lock(global_mutex);
if (current_ui_language != ui_language || iterators.find("") == iterators.end()) { if (current_ui_language != ui_language || iterators.find(empty_string) == iterators.end()) {
current_ui_language.clear(); current_ui_language = ui_language; current_ui_language.clear(); current_ui_language = ui_language;
icu::ErrorCode status; icu::ErrorCode status;
if (current_ui_language.empty()) { if (current_ui_language.empty()) {
iterators[""] = BreakIterator(icu::BreakIterator::createWordInstance(icu::Locale::getDefault(), status)); iterators[empty_string] = BreakIterator(icu::BreakIterator::createWordInstance(icu::Locale::getDefault(), status));
} else { } else {
ensure_lang_iterator(ui_language); ensure_lang_iterator(ui_language);
} }
@ -297,7 +291,8 @@ public:
Tokenizer(const char **args, int nargs, bool stem_words = false) : Tokenizer(const char **args, int nargs, bool stem_words = false) :
remove_diacritics(true), stem_words(stem_words), diacritics_remover(), remove_diacritics(true), stem_words(stem_words), diacritics_remover(),
byte_offsets(), token_buf(), current_ui_language(""), byte_offsets(), token_buf(), current_ui_language(""),
current_callback(NULL), current_callback_ctx(NULL), iterators(), current_callback(NULL), current_callback_ctx(NULL),
iterators(), stemmers(),
constructor_error(SQLITE_OK) constructor_error(SQLITE_OK)
{ {