Hi, On Mon, Feb 26, 2024 at 5:46 PM Douglas Anderson <diand...@chromium.org> wrote: > > If we're using 'hole-detection=raw' then we're going to be manually > looking for sparseness in our files. We don't really care about > whether the OS thinks they're sparse, so don't do the check about the > number of blocks the OS thinks the file takes. > > Without this change if I use "tar --hole-detection=raw -cf" to tar up > a file that's got lots of zeros (but isn't marked by the OS as > sparse), when I untar the file then it's still not sparse. With this > change then the zeros are correctly transformed into sparseness. > > --- > This is the same patch that was submitted via Savannah [1], but I'm > not terribly familiar with the process for submitting tar patches so I > thought I'd maybe post to this list too. Sorry if I've got this wrong. > > [1] https://savannah.gnu.org/patch/index.php?10432 > --- > src/create.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-)
I'm hoping for some sort of response, either positive or negative, about this patch. I'm happy to make changes and submit a new version if there is something folks don't like about it. IMO this is fixing a real problem with the "--hole-detection=raw" mode and it would be nice to get it in. If people disagree I'm happy to discuss or alternatively go back to the drawing board and try to find a different way to solve my problem. If people simply need more time to evaluate this patch then I can also step back and wait patiently. I'm not trying to be a pest but am trying to figure out how to move forward. Please let me know! Thanks! -Doug
