Lars Gullik Bjønnes wrote:
Abdelrazak Younes <[EMAIL PROTECTED]> writes:

| Abdelrazak Younes wrote:
| > Lars Gullik Bjønnes wrote:
| >> Abdelrazak Younes <[EMAIL PROTECTED]> writes:
| >>
| >> | OK, here is my last patch before holidays (hopefully).
| >> | | This one continue the drastic diet operated on BufferView. It removes
| >> | all boost signal from BufferView::pimpl and put them in LyXView.
| >> | LyXView already inherits from boost::signals::trackable so this is not
| >> | a big deal. If you look closely at the signal connections, it all
| >> | makes sense IMO.
| >> | | With this change, the only difference between BufferView.h and
| >> | BufferView_pimpl.h is "cursor.h" so I think it is time to merge at
| >> | last the two classes.
| >> | | Comments? Objection?
| >>
| >> Yes. I belive you have moved a bit too much. In particular the
| >> errorList itself. That is not a frontend thing really it is a
| >> bufferview thing.
| > OK. Perhaps even Buffer?
| | Actually, after reading the Buffer code, I think it makes a lot of
| sense to put the error_list_ in there. Right now there is only three
| type of errors in Buffer.C:

I do not agree.
errorList is primarly for recording latex errors, the buffer should
have as little as possible to do with that.
(Or other viewing/typesetting/external errors (typesetting engine
related))
)

I agree but so does the BufferView. I just send a patch that is a first step toward a proper cleanup.



| one in "bool Buffer::readDocument(LyXLex & lex)":
| | string const s = _("\\begin_document is missing");
|       error(ErrorItem(_("Document header error"), s, -1, 0, 0));
| | two in "int Buffer::readHeader(LyXLex & lex)": | | string const s = bformat(_("Unknown token: "...
|       error(ErrorItem(_("Document header error"), s, -1, 0, 0));
| | string const s = _("\\begin_header is missing");
|       error(ErrorItem(_("Document header error"), s, -1, 0, 0));

I'd rather get rid of this, or use a different mechanishm for these
errors. (After all these errors should never happen. It is an invalid
lyx document.)

I am in line with you. But let's proceed iteratively please. My last mail sets a plan for cleaning this up. Let's discuss the issue in this other thread.

Abdel.

Reply via email to