1
0
Fork 0
libreoffice/lingucomponent/source/languageguessing/guess.cxx
Daniel Baumann 8e63e14cf6
Adding upstream version 4:25.2.3.
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
2025-06-22 16:20:04 +02:00

100 lines
3.5 KiB
C++

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/*
* This file is part of the LibreOffice project.
*
* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/.
*
* This file incorporates work covered by the following license notice:
*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed
* with this work for additional information regarding copyright
* ownership. The ASF licenses this file to you under the Apache
* License, Version 2.0 (the "License"); you may not use this file
* except in compliance with the License. You may obtain a copy of
* the License at http://www.apache.org/licenses/LICENSE-2.0 .
*/
#include <sal/config.h>
#include <cassert>
#include <string.h>
#ifdef SYSTEM_LIBEXTTEXTCAT
#include <libexttextcat/textcat.h>
#else
#include <textcat.h>
#endif
#include "guess.hxx"
/* Old textcat.h versions defined bad spelled constants. */
#ifndef TEXTCAT_RESULT_UNKNOWN_STR
#define TEXTCAT_RESULT_UNKNOWN_STR _TEXTCAT_RESULT_UNKOWN
#endif
#ifndef TEXTCAT_RESULT_SHORT_STR
#define TEXTCAT_RESULT_SHORT_STR _TEXTCAT_RESULT_SHORT
#endif
Guess::Guess()
: language_str(DEFAULT_LANGUAGE)
, country_str(DEFAULT_COUNTRY)
{
}
/*
* this use a char * string to build the guess object
* a string like those is made as : [language-country-encoding]...
*
*/
Guess::Guess(const char * guess_str)
: language_str(DEFAULT_LANGUAGE)
, country_str(DEFAULT_COUNTRY)
{
//if the guess is not like "UNKNOWN" or "SHORT", go into the brackets
if(strcmp(guess_str + 1, TEXTCAT_RESULT_UNKNOWN_STR) == 0
|| strcmp(guess_str + 1, TEXTCAT_RESULT_SHORT_STR) == 0)
return;
// From how this ctor is called from SimpleGuesser::GuessLanguage and
// SimpleGuesser::GetManagedLanguages in
// lingucomponent/source/languageguessing/simpleguesser.cxx, guess_str must start with "[":
assert(guess_str[0] == GUESS_SEPARATOR_OPEN);
auto const start = guess_str + 1;
// Only look at the prefix of guess_str, delimited by the next "]" or "[" or end-of-string;
// split it into at most three segments separated by "-" (where excess occurrences of "-"
// would become part of the third segment), like "en-US-utf8"; the first segment denotes the
// language; if there are three segments, the second denotes the country and the third the
// encoding; otherwise, the second segment, if any (e.g., in "haw-utf8"), denotes the
// encoding:
char const * dash1 = nullptr;
char const * dash2 = nullptr;
auto p = start;
for (;; ++p) {
auto const c = *p;
if (c == '\0' || c == GUESS_SEPARATOR_OPEN || c == GUESS_SEPARATOR_CLOSE) {
break;
}
if (c == GUESS_SEPARATOR_SEP) {
if (dash1 == nullptr) {
dash1 = p;
} else {
dash2 = p;
// The encoding is ignored, so we can stop as soon as we found the second "-":
break;
}
}
}
auto const langLen = (dash1 == nullptr ? p : dash1) - start;
if (langLen != 0) { // if not we use the default value
language_str.assign(start, langLen);
}
if (dash2 != nullptr) {
country_str.assign(dash1 + 1, dash2 - (dash1 + 1));
}
}
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */