| 1 |
On Sun, Aug 16, 2015 at 12:33:05PM -0500, William Hubbs wrote: |
| 2 |
> On Sat, Aug 15, 2015 at 02:44:47PM +0200, Peter Stuge wrote: |
| 3 |
> > Hi and happy Git days! :) |
| 4 |
> > |
| 5 |
> > |
| 6 |
> > Robin H. Johnson wrote: |
| 7 |
> > > It expands to the hash of the blob of that file; and from that, you can |
| 8 |
> > > identify which commits the blob exists in. |
| 9 |
> > |
| 10 |
> > $ git ls-tree HEAD README |
| 11 |
> > 100644 blob 08ae16956b8944da2fef75fee892dcba457cf4f0 README |
| 12 |
> > $ |
| 13 |
> > |
| 14 |
> > $ (stat --printf='blob %s\0' README; cat README) | sha1sum |
| 15 |
> > 08ae16956b8944da2fef75fee892dcba457cf4f0 - |
| 16 |
> > $ |
| 17 |
> > |
| 18 |
> > This is so simple to generate that it doesn't really need a |
| 19 |
> > placeholder in every ebuild in the repository. |
| 20 |
It's ONLY applicable if the file has never been modified. If the file |
| 21 |
has been modified, then unless you know exactly what the modifications |
| 22 |
were (or have a backup before modifications), you cannot figure out the |
| 23 |
exact source. |
| 24 |
|
| 25 |
> If this is this simple to regenerate, consider my response here a very |
| 26 |
> strong +1; let's kill $id$ out of the git repository and generate this |
| 27 |
> some how at the rsync level if users need it or give them a tool that |
| 28 |
> will generate it. |
| 29 |
Ebuilds are NOT the only place where it's relevant, but they have a |
| 30 |
similar use case to the other places. |
| 31 |
|
| 32 |
Ebuild: |
| 33 |
- User copies .ebuild with expanded $Id$ to /usr/local/portage/... |
| 34 |
- User modifies local ebuild copy |
| 35 |
- User submits changed ebuild to bugzilla |
| 36 |
- Developer uses Header/Id to figure out what to base a diff on for |
| 37 |
merging. |
| 38 |
|
| 39 |
Non-ebuild |
| 40 |
- Package installs /etc/init.d/foobar, with some bug |
| 41 |
- User reports it is buggy |
| 42 |
- User modifies it to temporarily work around the bug |
| 43 |
- User submits modified version back |
| 44 |
- Developer asks what the original version was if it didn't have some |
| 45 |
useful marker in the file. |
| 46 |
|
| 47 |
Example of old marker: |
| 48 |
/etc/init.d/nfs:# $Header: /var/cvsroot/gentoo-x86/net-fs/nfs-utils/files/nfs.initd,v 1.26 2011/09/18 01:51:51 vapier Exp $ |
| 49 |
|
| 50 |
> I don't object to someone having a way to get this information if it |
| 51 |
> is useful to them, but I don't think we need a placeholder for it in the |
| 52 |
> git repository. |
| 53 |
If we don't have a placeholder for where to inject it, how do we know |
| 54 |
where it's safe to inject? |
| 55 |
|
| 56 |
-- |
| 57 |
Robin Hugh Johnson |
| 58 |
Gentoo Linux: Developer, Infrastructure Lead |
| 59 |
E-Mail : robbat2@g.o |
| 60 |
GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85 |
Archives