On 27/05/2021 18:44, msk...@ansuz.sooke.bc.ca wrote:
On Thu, 27 May 2021, Philip Taylor wrote:
msk...@ansuz.sooke.bc.ca wrote:
Is this a function of the PDF writer "instructing" the reader to reload,
or is it something the PDF reader does independently?
Empirical observation suggests the former. If it were the latter, then
would it (Adobe Acrobat, that is) not do the same when XeTeX + (x)dvipdfm(x)
attempts to write to the file ? At the moment, the latter simply aborts and
the file remains unchanged.
Well, if the writer is aborting, then there must be some kind of
communication between it and the reader already.
My guess would be that Acrobat has the file open, and Windows doesn't
let the would-be writer open it for writing while someone else already
has it open for reading.
Under Linux it would
normally be impossible for the reader to prevent the writer from rewriting
the file; but that's a different issue from what I thought was the
question, about the reader knowing to reload the file *after* the writer
has written it. Seems like it will need some kind of Windows-specific
handling to resolve, then.
