Lars Gullik Bjønnes wrote:
Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| Abdel.
|
| Index: buffer_funcs.cpp
| ===================================================================
| --- buffer_funcs.cpp (revision 19485)
| +++ buffer_funcs.cpp (working copy)
| @@ -614,4 +616,25 @@
| }
| }
|
| +
| +void loadChildDocuments(Buffer const & buf)
| +{
| + bool parse_error = false;
| +
| + for (InsetIterator it = inset_iterator_begin(buf.inset()); it; ++it) {
Should the test be comparing against the end-iterator instead?
I just copied and pasted this part. (it) test is already used elsewere
in the code.
Would
be clearer if possible, nand fit better with other uses of iterators.
But I see that this is done in old code as well, so don't mind to
much. Should probably we changed later on though.
Yes.
| Index: BufferView.cpp
| ===================================================================
| --- BufferView.cpp (revision 19485)
| +++ BufferView.cpp (working copy)
| @@ -872,21 +752,21 @@
It seems that almost all changes to BufferView is buffer_-> to
buffer_. changes.
| Index: BufferView.h
| ===================================================================
| --- BufferView.h (revision 19485)
| +++ BufferView.h (working copy)
| @@ -269,7 +263,7 @@
| ///
| CoordCache coord_cache_;
| ///
| - Buffer * buffer_;
| + Buffer & buffer_;
To keep the size of the patch down I think you should leave this as
Buffer * buffer_;
As I said, I won't revert more of this patch. There is absolutely no
reason why buffer_ should be a pointer.
| Index: frontends/controllers/ControlTabular.cpp
| ===================================================================
| --- frontends/controllers/ControlTabular.cpp (revision 19485)
| +++ frontends/controllers/ControlTabular.cpp (working copy)
| @@ -39,7 +39,7 @@
| // assume that it is "ours"
| for (int i = cur.depth() - 1; i >= 0; --i)
| if (cur[i].inset().lyxCode() == Inset::TABULAR_CODE) {
| - current_inset = static_cast<InsetTabular const
*>(&cur[i].inset());
| + current_inset = static_cast<InsetTabular
*>(&cur[i].inset());
Why is const wrong?
I reverted that in my most current patch.
Your patch is very big (size wise) and I am afraid that at least to me
that muddles a bit what is happening in it, and it really is tiresome
to read such huge patches.
Which is exactly the reason why I suggested to read the svn diff instead.
Abdel.
