Jens Lehmann <Jens.Lehmann@xxxxxx> writes: > Am 13.05.2011 22:03, schrieb Junio C Hamano: >> * jl/read-tree-m-dry-run (2011-05-11) 1 commit >> - Teach read-tree the -n|--dry-run option >> >> Looked good. We _might_ want to make -n simply ignore -u, though. > > Fine by me. Do you want me to send an updated version? If so, > should a warning be issued in that case or should that just > happen silently? Later we _may_ have a condition where "read-tree -m" with some options (except "-n") may succeed but "read-tree -m -u" with the same other options may fail, and the reason to have "-n" is so that the caller can tell if the operation may or may not fail in advance without changing the state. Singling out "do not write INDEX" was a mistake to begin with. How about "simply ignore" aka "silently"? We do not even have to make it sound like an user error, if we explain the option like this. I think the change to the implementation would be trivial, but you would probably want to update the tests in 1002 (remove the one at the end that checks -u/-n incompatibility, and perhaps make sure "-n -u" combination does not touch anything as advertised). Thanks. diff --git a/Documentation/git-read-tree.txt b/Documentation/git-read-tree.txt index a35849f..46a96f2 100644 --- a/Documentation/git-read-tree.txt +++ b/Documentation/git-read-tree.txt @@ -55,8 +55,8 @@ OPTIONS -n:: --dry-run:: - Don't write the index file. This option isn't allowed together - with -u. + Check if the command would error out, without updating the index + nor the files in the working tree for real. -v:: Show the progress of checking files out. -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html