The Tor Browser / file comparison
comparison: intl/lwbrk/src/nsSemanticUnitScanner.cpp
intl/lwbrk/src/nsSemanticUnitScanner.cpp
- changeset 0
- 6474c204b198
equal
deleted
inserted
replaced
| -1:000000000000 | 0:c08404ea3158 |
|---|---|
| 1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ | |
| 2 /* This Source Code Form is subject to the terms of the Mozilla Public | |
| 3 * License, v. 2.0. If a copy of the MPL was not distributed with this | |
| 4 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ | |
| 5 | |
| 6 #include "nsSemanticUnitScanner.h" | |
| 7 | |
| 8 NS_IMPL_ISUPPORTS(nsSemanticUnitScanner, nsISemanticUnitScanner) | |
| 9 | |
| 10 nsSemanticUnitScanner::nsSemanticUnitScanner() : nsSampleWordBreaker() | |
| 11 { | |
| 12 /* member initializers and constructor code */ | |
| 13 } | |
| 14 | |
| 15 nsSemanticUnitScanner::~nsSemanticUnitScanner() | |
| 16 { | |
| 17 /* destructor code */ | |
| 18 } | |
| 19 | |
| 20 | |
| 21 /* void start (in string characterSet); */ | |
| 22 NS_IMETHODIMP nsSemanticUnitScanner::Start(const char *characterSet) | |
| 23 { | |
| 24 // do nothing for now. | |
| 25 return NS_OK; | |
| 26 } | |
| 27 | |
| 28 /* void next (in wstring text, in long length, in long pos, out boolean hasMoreUnits, out long begin, out long end); */ | |
| 29 NS_IMETHODIMP nsSemanticUnitScanner::Next(const char16_t *text, int32_t length, int32_t pos, bool isLastBuffer, int32_t *begin, int32_t *end, bool *_retval) | |
| 30 { | |
| 31 // xxx need to bullet proff and check input pointer | |
| 32 // make sure begin, end and _retval is not nullptr here | |
| 33 | |
| 34 // if we reach the end, just return | |
| 35 if (pos >= length) { | |
| 36 *begin = pos; | |
| 37 *end = pos; | |
| 38 *_retval = false; | |
| 39 return NS_OK; | |
| 40 } | |
| 41 | |
| 42 uint8_t char_class = nsSampleWordBreaker::GetClass(text[pos]); | |
| 43 | |
| 44 // if we are in chinese mode, return one han letter at a time | |
| 45 // we should not do this if we are in Japanese or Korean mode | |
| 46 if (kWbClassHanLetter == char_class) { | |
| 47 *begin = pos; | |
| 48 *end = pos+1; | |
| 49 *_retval = true; | |
| 50 return NS_OK; | |
| 51 } | |
| 52 | |
| 53 int32_t next; | |
| 54 // find the next "word" | |
| 55 next = NextWord(text, (uint32_t) length, (uint32_t) pos); | |
| 56 | |
| 57 // if we don't have enough text to make decision, return | |
| 58 if (next == NS_WORDBREAKER_NEED_MORE_TEXT) { | |
| 59 *begin = pos; | |
| 60 *end = isLastBuffer ? length : pos; | |
| 61 *_retval = isLastBuffer; | |
| 62 return NS_OK; | |
| 63 } | |
| 64 | |
| 65 // if what we got is space or punct, look at the next break | |
| 66 if ((char_class == kWbClassSpace) || (char_class == kWbClassPunct)) { | |
| 67 // if the next "word" is not letters, | |
| 68 // call itself recursively with the new pos | |
| 69 return Next(text, length, next, isLastBuffer, begin, end, _retval); | |
| 70 } | |
| 71 | |
| 72 // for the rest, return | |
| 73 *begin = pos; | |
| 74 *end = next; | |
| 75 *_retval = true; | |
| 76 return NS_OK; | |
| 77 } | |
| 78 |
