intl/icu/source/i18n/inputext.h@b8a032363ba2 (annotated)
intl/icu/source/i18n/inputext.h
Thu, 22 Jan 2015 13:21:57 +0100
- author
- Michael Schloh von Bennewitz <michael@schloh.com>
- date
- Thu, 22 Jan 2015 13:21:57 +0100
- branch
- TOR_BUG_9701
- changeset 15
- b8a032363ba2
- permissions
- -rw-r--r--
Incorporate requested changes from Mozilla in review:
https://bugzilla.mozilla.org/show_bug.cgi?id=1123480#c6
| michael@0 | 1 | /* |
| michael@0 | 2 | ********************************************************************** |
| michael@0 | 3 | * Copyright (C) 2005-2008, International Business Machines |
| michael@0 | 4 | * Corporation and others. All Rights Reserved. |
| michael@0 | 5 | ********************************************************************** |
| michael@0 | 6 | */ |
| michael@0 | 7 | |
| michael@0 | 8 | #ifndef __INPUTEXT_H |
| michael@0 | 9 | #define __INPUTEXT_H |
| michael@0 | 10 | |
| michael@0 | 11 | /** |
| michael@0 | 12 | * \file |
| michael@0 | 13 | * \internal |
| michael@0 | 14 | * |
| michael@0 | 15 | * This is an internal header for the Character Set Detection code. The |
| michael@0 | 16 | * name is probably too generic... |
| michael@0 | 17 | */ |
| michael@0 | 18 | |
| michael@0 | 19 | |
| michael@0 | 20 | #include "unicode/uobject.h" |
| michael@0 | 21 | |
| michael@0 | 22 | #if !UCONFIG_NO_CONVERSION |
| michael@0 | 23 | |
| michael@0 | 24 | U_NAMESPACE_BEGIN |
| michael@0 | 25 | |
| michael@0 | 26 | class InputText : public UMemory |
| michael@0 | 27 | { |
| michael@0 | 28 | // Prevent copying |
| michael@0 | 29 | InputText(const InputText &); |
| michael@0 | 30 | public: |
| michael@0 | 31 | InputText(UErrorCode &status); |
| michael@0 | 32 | ~InputText(); |
| michael@0 | 33 | |
| michael@0 | 34 | void setText(const char *in, int32_t len); |
| michael@0 | 35 | void setDeclaredEncoding(const char *encoding, int32_t len); |
| michael@0 | 36 | UBool isSet() const; |
| michael@0 | 37 | void MungeInput(UBool fStripTags); |
| michael@0 | 38 | |
| michael@0 | 39 | // The text to be checked. Markup will have been |
| michael@0 | 40 | // removed if appropriate. |
| michael@0 | 41 | uint8_t *fInputBytes; |
| michael@0 | 42 | int32_t fInputLen; // Length of the byte data in fInputBytes. |
| michael@0 | 43 | // byte frequency statistics for the input text. |
| michael@0 | 44 | // Value is percent, not absolute. |
| michael@0 | 45 | // Value is rounded up, so zero really means zero occurences. |
| michael@0 | 46 | int16_t *fByteStats; |
| michael@0 | 47 | UBool fC1Bytes; // True if any bytes in the range 0x80 - 0x9F are in the input;false by default |
| michael@0 | 48 | char *fDeclaredEncoding; |
| michael@0 | 49 | |
| michael@0 | 50 | const uint8_t *fRawInput; // Original, untouched input bytes. |
| michael@0 | 51 | // If user gave us a byte array, this is it. |
| michael@0 | 52 | // If user gave us a stream, it's read to a |
| michael@0 | 53 | // buffer here. |
| michael@0 | 54 | int32_t fRawLength; // Length of data in fRawInput array. |
| michael@0 | 55 | |
| michael@0 | 56 | }; |
| michael@0 | 57 | |
| michael@0 | 58 | U_NAMESPACE_END |
| michael@0 | 59 | |
| michael@0 | 60 | #endif |
| michael@0 | 61 | #endif /* __INPUTEXT_H */ |
