Hello! Still able to reproduce the issue, and definitely willing to test it, given a binary. As of writing, compiling it myself isn't feasible though.
Also, is there a reasonable chance that the write succeeds after a > smaller delay than 1 second? For example, might we want to try again > after 0.25 seconds for the first time, 0.5 seconds after the second, 1 > second after the third, and then stay at 1 second for the fourth, fifth, > and sixth try? I can update the patch, if this change would be desired. > I have no intuition for the cost of trying more frequently, so I don't > have a strong opinion. It would probably suffice to remove the temporary file, once the "Retry" succeeds. Currently, even the "Retry" leaves the temporary file; Presumably the "Retry" creates a second temporary file and forgets about the first one. - Klaus On Wed, Aug 8, 2018 at 5:38 AM Richard Kimberly Heck <rikih...@lyx.org> wrote: > On 08/07/2018 07:13 PM, Scott Kostyshak wrote: > > The following issue seems to affect a lot of users: > > > > https://www.lyx.org/trac/ticket/10091 > > > > Riki posted a patch (see comment 42), but from what I can tell it has > > not had any testing on Windows since no one who can reproduce the issue > > can also compile on Windows. > > > > It seems that the risk to including the patch is that it does not fix > > the issue and actually makes the issue worse by introducing a 5 second > > delay. However, perhaps that risk is worth it. > > > > Riki, can you provide the commands that I need to make a test binary for > > Windows users? This way I can try to save you some time from creating it > > yourself since you have already done so much. Well, actually it might > > take you more time to explain to me how to do it, but I would volunteer > > to make test binaries for Windows users in the future. > > Yes, I am thinking it might take more time to explain it. The code is > currenty in a > branch in my developers' repo. I'll put it into 2.3.x at some point > soon, and then others > will be able to use it. > > Riki > >
