i2p/haiku-beta6
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Improve the word wrapping a bit. This should eventually use the LocalKit when

Browse Source 
it's ready. For now, I just tried to fix stuff like breaking before
punctuations. I had trouble wrapping my mind around the wrapping code, so I
didn't include as much cases as I wanted. Basically, the wrapping code assumes
that a spot where CanEndLine() returns true must be a white space character,
which wouldn't be the case.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30075 a95241bf-73f2-0310-859d-f6bbb57e9c96
This commit is contained in:
Stephan Aßmus
2009-04-09 17:54:01 +00:00
parent e7cdf7f3f5
commit bedfcd63af
1 changed files with 43 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/kits/interface/TextView.cpp
+43 -8
View File
@@ -100,7 +100,9 @@ static const uint32 kFlattenedTextRunArrayVersion = 0;
enum {
B_SEPARATOR_CHARACTER,
B_OTHER_CHARACTER
B_PUNCTUATION_CHARACTER,
B_OTHER_CHARACTER,
B_END_OF_TEXT
};
@@ -1953,8 +1955,35 @@ BTextView::FindWord(int32 inOffset, int32 *outFromOffset, int32 *outToOffset)
bool
BTextView::CanEndLine(int32 offset)
{
// TODO: Could be improved, the bebook says there are other checks to do
return (_CharClassification(offset) == B_SEPARATOR_CHARACTER);
// TODO: This has to be improved a lot, but also the wrapping code.
// It should use the forthcomming LocalKit.
uint32 classification = _CharClassification(offset);
if (classification == B_END_OF_TEXT
|| classification == B_SEPARATOR_CHARACTER) {
return true;
}
uint32 nextClassification = _CharClassification(offset + 1);
if (nextClassification == B_END_OF_TEXT)
return true;
if (classification == B_PUNCTUATION_CHARACTER
&& nextClassification == B_OTHER_CHARACTER) {
return true;
}
// TODO: This cannot be enabled, since the wrapping code things the char
// is a trailing space or something. Otherwise it would allow to treat
// something like "..." as a word.
// uint32 nextNextClassification = _CharClassification(offset + 2);
//
// if (classification == B_OTHER_CHARACTER
// && nextClassification == B_PUNCTUATION_CHARACTER
// && nextClassification == nextNextClassification) {
// return true;
// }
return false;
}
@@ -3679,9 +3708,9 @@ BTextView::_FindLineBreak(int32 fromOffset, float *outAscent, float *outDescent,
int32 offset = fromOffset;
// Text wrapping is turned off.
// Just find the offset of the first \n character
if (!fWrap) {
// Text wrapping is turned off.
// Just find the offset of the first \n character
offset = limit - fromOffset;
fText->FindChar(B_ENTER, fromOffset, &offset);
offset += fromOffset;
@@ -4823,16 +4852,17 @@ BTextView::_SetRunArray(int32 startOffset, int32 endOffset,
uint32
BTextView::_CharClassification(int32 offset) const
{
// TODO:Should check against a list of characters containing also
// TODO: Should check against a list of characters containing also
// japanese word breakers.
// And what about other languages ? Isn't there a better way to check
// for separator characters ?
// Andrew suggested to have a look at UnicodeBlockObject.h
switch (fText->RealCharAt(offset)) {
case '\0':
return B_END_OF_TEXT;
case B_SPACE:
case '_':
case '.':
case '\0':
case B_TAB:
case B_ENTER:
case '&':
@@ -4847,6 +4877,11 @@ BTextView::_CharClassification(int32 offset) const
case '^':
case '|':
return B_SEPARATOR_CHARACTER;
case '.':
case ',':
return B_PUNCTUATION_CHARACTER;
default:
return B_OTHER_CHARACTER;
}