Magic-Mirror/gecko-dev
1
0
Fork 0
mirror of https://github.com/mozilla/gecko-dev.git synced 2024-11-29 07:42:04 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Bug 755480 - Part 3: Make mozInlineSpellWordUtil::SplitDOMWord O(n) for strings with no special charts (as opposed to O(n2)); r=roc

Browse Source
This commit is contained in:
Ehsan Akhgari 2012-05-15 16:38:43 -04:00
parent 512ba908f5
commit 720ebd36cd
1 changed files with 23 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

52
extensions/spellcheck/src/mozInlineSpellWordUtil.cpp
View File

@ -761,7 +761,7 @@ enum CharClass {
CHAR_CLASS_END_OF_INPUT };
// Encapsulates DOM-word to real-word splitting
struct WordSplitState
struct NS_STACK_CLASS WordSplitState
{
mozInlineSpellWordUtil* mWordUtil;
const nsDependentSubstring mDOMWordText;
@ -782,9 +782,9 @@ struct WordSplitState
// current position, and returns their length, or 0 if not found. This allows
// arbitrary word breaking rules to be used for these special entities, as
// long as they can not contain whitespace.
PRInt32 FindSpecialWord();
bool IsSpecialWord();
// Similar to FindSpecialWord except that this takes a split word as
// Similar to IsSpecialWord except that this takes a split word as
// input. This checks for things that do not require special word-breaking
// rules.
bool ShouldSkipWord(PRInt32 aStart, PRInt32 aLength);
@ -912,18 +912,16 @@ WordSplitState::AdvanceThroughWord()
}
// WordSplitState::FindSpecialWord
// WordSplitState::IsSpecialWord
PRInt32
WordSplitState::FindSpecialWord()
bool
WordSplitState::IsSpecialWord()
{
PRInt32 i;
// Search for email addresses. We simply define these as any sequence of
// characters with an '@' character in the middle. The DOM word is already
// split on whitepace, so we know that everything to the end is the address
PRInt32 firstColon = -1;
for (i = mDOMWordOffset;
for (PRInt32 i = mDOMWordOffset;
i < PRInt32(mDOMWordText.Length()); i ++) {
if (mDOMWordText[i] == '@') {
// only accept this if there are unambiguous word characters (don't bother
@ -938,9 +936,9 @@ WordSplitState::FindSpecialWord()
// current position for potentially removing a spelling range.
if (i > 0 && ClassifyCharacter(i - 1, false) == CHAR_CLASS_WORD &&
i < (PRInt32)mDOMWordText.Length() - 1 &&
ClassifyCharacter(i + 1, false) == CHAR_CLASS_WORD)
return mDOMWordText.Length() - mDOMWordOffset;
ClassifyCharacter(i + 1, false) == CHAR_CLASS_WORD) {
return true;
}
} else if (mDOMWordText[i] == ':' && firstColon < 0) {
firstColon = i;
@ -948,7 +946,7 @@ WordSplitState::FindSpecialWord()
// This will catch things like asdf://foo.com
if (firstColon < (PRInt32)mDOMWordText.Length() - 1 &&
mDOMWordText[firstColon + 1] == '/') {
return mDOMWordText.Length() - mDOMWordOffset;
return true;
}
}
}
@ -967,12 +965,12 @@ WordSplitState::FindSpecialWord()
protocol.EqualsIgnoreCase("javascript") ||
protocol.EqualsIgnoreCase("data") ||
protocol.EqualsIgnoreCase("ftp")) {
return mDOMWordText.Length() - mDOMWordOffset;
return true;
}
}
// not anything special
return -1;
return false;
}
// WordSplitState::ShouldSkipWord
@ -1002,25 +1000,21 @@ mozInlineSpellWordUtil::SplitDOMWord(PRInt32 aStart, PRInt32 aEnd)
WordSplitState state(this, mSoftText, aStart, aEnd - aStart);
state.mCurCharClass = state.ClassifyCharacter(0, true);
state.AdvanceThroughSeparators();
if (state.mCurCharClass != CHAR_CLASS_END_OF_INPUT &&
state.IsSpecialWord()) {
PRInt32 specialWordLength = state.mDOMWordText.Length() - state.mDOMWordOffset;
mRealWords.AppendElement(
RealWord(aStart + state.mDOMWordOffset, specialWordLength, false));
return;
}
while (state.mCurCharClass != CHAR_CLASS_END_OF_INPUT) {
state.AdvanceThroughSeparators();
if (state.mCurCharClass == CHAR_CLASS_END_OF_INPUT)
break;
PRInt32 specialWordLength = state.FindSpecialWord();
if (specialWordLength > 0) {
mRealWords.AppendElement(
RealWord(aStart + state.mDOMWordOffset, specialWordLength, false));
// skip the special word
state.mDOMWordOffset += specialWordLength;
if (state.mDOMWordOffset + aStart >= aEnd)
state.mCurCharClass = CHAR_CLASS_END_OF_INPUT;
else
state.mCurCharClass = state.ClassifyCharacter(state.mDOMWordOffset, true);
continue;
}
// save the beginning of the word
PRInt32 wordOffset = state.mDOMWordOffset;