On Sun, Oct 01, 2017 at 08:54:54PM +0200, Christoph Berg wrote: > Re: Osamu Aoki 2017-09-30 > <[email protected]> > > But I don't find strong reason nor have guts to take this chance without > > others encouraging me. Let me discuss each case here with you to sort > > this out. > > > > * --download-version set in the command line option > > --> maybe set $download=2 unconditionally > > * --download-debversion set in the command line option > > --> maybe set $download=2 unconditionally > > * --download-current-version set in the command line option > > --> maybe set $download=2 unconditionally > > Is there a use case for using one of the --download-* switches without > actually downloading something?
This is the question I am worried and I needed some time to think so I will not make wrong decision. uscan is used to scan upstream without downloading upstream. The case when the normal scan with --no-download fails is when there is no new version is available. So exit code itself is not useful for scanning but by combining it with the output of uscan, they can be used to determine availability current or new version and goodness of watch file. > If not, they should force --download So anytime users use --download-*, they wish to set --download too. > and hence override USCAN_DOWNLOAD=no. I guess I need to write manual accordingly. Osamu _______________________________________________ devscripts-devel mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/devscripts-devel
