On Tue, Jan 25, 2005 at 03:12:42PM +0100, Lars Gullik Bjønnes wrote:
> So this isn't the case of a missing initialization of rows_?
>
> Shouldn't a paragraph _always_ have at least one row if it exists?
Not anymore. It used to be like that but nowadays there might be (lots
of) paragraph in the do
Asger Ottar Alstrup <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>> Unless _really_ big, do you have to attach as octed-stream and bzip?
>
| Is 61k ok to send as raw?
for me: semi-ok
>
| I used Thunderbird as my mailer. Teach me how to change the MIME and I
| will. Do you prefer som
Asger Ottar Alstrup <[EMAIL PROTECTED]> writes:
| Hi,
>
| The attached patch to dociterator.C should be obviously correct. You
| can not access a textRow if there are none. If you apply this, you
| will also get an assertion failure in text.C breakParagraph as I do
| without even this patch, becau
Lars Gullik Bjønnes wrote:
Unless _really_ big, do you have to attach as octed-stream and bzip?
Is 61k ok to send as raw?
I used Thunderbird as my mailer. Teach me how to change the MIME and I
will. Do you prefer something else than bzip?
Regards,
Asger
Asger Ottar Alstrup <[EMAIL PROTECTED]> writes:
| Well, I meant to include a patch to text.C, but it has been trying to
| make a diff for half an hour without any success. I just attach the
| latest version instead. You can make your own diff.
Unless _really_ big, do you have to attach as octed-s
Hi,
The attached patch to dociterator.C should be obviously correct. You can
not access a textRow if there are none. If you apply this, you will also
get an assertion failure in text.C breakParagraph as I do without even
this patch, because MSVC is so kind to flag a memory access error due to
p
