| Message ID | cover-0.5-00000000000-20210505T122816Z-avarab@gmail.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | streaming.c: refactor for smaller + easier to understand code | expand |
On Wed, May 05, 2021 at 02:33:27PM +0200, Ævar Arnfjörð Bjarmason wrote: > This is a prep series for my yet-to-be-sent re-roll of [1], > aka. ab/fsck-unexpected-type. > > Jeff King had a comment in [2] saying it was odd to have an "oi2" > variable, that was because there was on an "oi" name already defined > via the new-gone open_method_decl(loose) macro. As we'll see we could > do without the initial "oi" and a few other types of indirection in > this interface. > > Junio: Could you eject ab/fsck-unexpected-type while this is being > considered/cooked? I didn't really see how to address Jeff's feedback > about that variable name in a way that wouldn't just make something > like this refactoring part of an even bigger series. I think ejecting > the functional changes & trying to get this in first is the least > worst approach at this point. Eventually I'll submit a re-roll of > ab/fsck-unexpected-type either based on this, or master (if it's > landed already). I think my feedback was really just: if that patch didn't refactor parse_loose_header() to get rid of the "simple" and extended versions, then you wouldn't have to touch streaming.c at all. It could just continue to use the simple version. That said, I don't mind the cleanups here, especially getting rid of the macros. -Peff
This is a prep series for my yet-to-be-sent re-roll of [1], aka. ab/fsck-unexpected-type. Jeff King had a comment in [2] saying it was odd to have an "oi2" variable, that was because there was on an "oi" name already defined via the new-gone open_method_decl(loose) macro. As we'll see we could do without the initial "oi" and a few other types of indirection in this interface. Junio: Could you eject ab/fsck-unexpected-type while this is being considered/cooked? I didn't really see how to address Jeff's feedback about that variable name in a way that wouldn't just make something like this refactoring part of an even bigger series. I think ejecting the functional changes & trying to get this in first is the least worst approach at this point. Eventually I'll submit a re-roll of ab/fsck-unexpected-type either based on this, or master (if it's landed already). 1. https://lore.kernel.org/git/cover-0.6-00000000000-20210413T093734Z-avarab@gmail.com/#t 2. https://lore.kernel.org/git/YILZHiuUyj0mt958@coredump.intra.peff.net/ Ævar Arnfjörð Bjarmason (5): streaming.c: avoid forward declarations streaming.c: remove enum/function/vtbl indirection streaming.c: remove {open,close,read}_method_decl() macros streaming.c: stop passing around "object_info *" to open() streaming.c: move {open,close,read} from vtable to "struct git_istream" streaming.c | 268 ++++++++++++++++++++++------------------------------ 1 file changed, 115 insertions(+), 153 deletions(-)