Thu, 22 Jan 2015 13:21:57 +0100
Incorporate requested changes from Mozilla in review:
https://bugzilla.mozilla.org/show_bug.cgi?id=1123480#c6
1 /*
2 **********************************************************************
3 * Copyright (C) 2005-2012, International Business Machines
4 * Corporation and others. All Rights Reserved.
5 **********************************************************************
6 */
8 #ifndef __CSRECOG_H
9 #define __CSRECOG_H
11 #include "unicode/uobject.h"
13 #if !UCONFIG_NO_CONVERSION
15 #include "inputext.h"
17 U_NAMESPACE_BEGIN
19 class CharsetMatch;
21 class CharsetRecognizer : public UMemory
22 {
23 public:
24 /**
25 * Get the IANA name of this charset.
26 * Note that some recognizers can recognize more than one charset, but that this API
27 * assumes just one name per recognizer.
28 * TODO: need to account for multiple names in public API that enumerates over the
29 * known detectable charsets.
30 * @return the charset name.
31 */
32 virtual const char *getName() const = 0;
34 /**
35 * Get the ISO language code for this charset.
36 * @return the language code, or <code>null</code> if the language cannot be determined.
37 */
38 virtual const char *getLanguage() const;
40 /*
41 * Try the given input text against this Charset, and fill in the results object
42 * with the quality of the match plus other information related to the match.
43 *
44 * Return TRUE if the the input bytes are a potential match, and
45 * FALSE if the input data is not compatible with, or illegal in this charset.
46 */
47 virtual UBool match(InputText *textIn, CharsetMatch *results) const = 0;
49 virtual ~CharsetRecognizer();
50 };
52 U_NAMESPACE_END
54 #endif
55 #endif /* __CSRECOG_H */