For print preview header/footer, I would like to be able to put several special codes at the same place (eg headerLeft, footerCenter, etc.), and have a special code to put total number of pages.

Comment on attachment 167416 [details] [diff] [review] &PT1 special code and no return after &PT treatment If you want to allow any arbitrary combination of codes (which makes more sense to me than allowing some combinations and not others), you need to remove the return at the end of each block. I would rather have a single letter code for "page total", to preserve the pattern that single letter codes insert one piece of data and the double letter code &PT inserts two. &T is already taken, of course, but you could use &O, &A, or &L. The change also needs to be reflected in the Help files, but that could be done in a separate bug.

> you need to remove the > return at the end of each block. The problem is that the SubstValueForCode function used for the next patterns do a Truncate() on the string passed if there is no title or url for the document.

So if I set one of the header fields to "Title: &T" and there is no <title>, no header at all will appear? That seems bogus to me.

Alexandre, if I were you I would remove SubstValueForCode completely and replace calls to SubstValueForCode(aNewStr, foo, bar); by aNewStr.ReplaceSubstring(foo, bar); As far as I can tell everything that SubstValueForCode is doing is either wrong or a workaround for some problem in string code which no longer exists.

(In reply to comment #5) > ... replace calls to > SubstValueForCode(aNewStr, foo, bar); > by > aNewStr.ReplaceSubstring(foo, bar); ... that might be going too far: better at least do a null-check on "bar" like the current code for &D

(In reply to comment #6) > (In reply to comment #5) > > ... replace calls to > > SubstValueForCode(aNewStr, foo, bar); > > by > > aNewStr.ReplaceSubstring(foo, bar); > > ... that might be going too far: better at least do a null-check on "bar" like > the current code for &D Thanks Simon, I think that's what I'm going to do. That SubstValueForCode function seemed strange to me because of that 'fixingSubstr' stuff. Makes sense that this was to avoid a bogus behavior of ReplaceSubstring.

Comment on attachment 167623 [details] [diff] [review] replaced call for SubstValueForCode by ReplaceSubstring, removed all "return;" statements Hi Simon, Can you review this new patch for me? I can't find any doc on setting header footer using nsIPrintSettings. I looked in embedding docs. Do you know where this might be? Alex

Comment on attachment 167623 [details] [diff] [review] replaced call for SubstValueForCode by ReplaceSubstring, removed all "return;" statements r=smontagu. Thank you, Alexandre, this is an excellent enhancement. The help files that mention custom headers and footers are extensions/help/resources/locale/en-US/nav_help.xhtml and browser/components/help/locale/en-US/using_firebird.xhtml

Does this still need to be checked in?

Let me know if you'd like me to check this in. It looks as if bug 232850 is about to be fixed, so you may not need me :)

Simon, Kenneth, It would be nice if this could be checked in, but I thought checkin was only for drivers until 1.8 is released. So even when I get my write access I'll have to wait, won't I?

Do I need to change the patch to update the path of the nsPageFrame file?

The freeze seems to have been delayed until after 1.8b2, so this can be checked in now. You don't need to attach a new patch against the changed location of the source file.

Alexandre, as Simon said, the tree is actually open right now. Even if it were closed, it would only mean having to get permission to check in the patch. And you don't have to check the patch in yourself; you can ask someone else to do it.

Is this still waiting to be checked in? It'd be nice to get this in before the tree branches...

Yes, Kenneth, Can you check this in for me? Or Simon maybe?

Comment on attachment 167623 [details] [diff] [review] replaced call for SubstValueForCode by ReplaceSubstring, removed all "return;" statements I will be happy to check this in, if drivers will approve it.

Comment on attachment 167623 [details] [diff] [review] replaced call for SubstValueForCode by ReplaceSubstring, removed all "return;" statements a=asa

Checked in.