1 I'd strongly recommend this modification to ikiwiki. Any particular limitations that anyone can think of?
3 I might even have a try at this patch, though I'd have to hack the user preferences page to include author name...
5 As to the question of whether the committer was the 'script' or the wiki editor... I'm not sure. Marking it as the script somehow (`ikiwiki-cgi <ikiwiki@sitename>`)? seems to make sense and would make it easier to manage.
9 I've been thinking a bit about the GIT attribution in ikiwiki...
11 If no email set, I think "$USERNAME" is reasonable... no point in the
13 >> **adjustement wrt comments**: leave the '<>' in due to requirements in git
15 If no username set... then something like '@[IPADDR]' makes sense...
16 (not in email brackets).
18 > Why not put it in email brackets? --[[Joey]]
20 In the case of OpenID login.. I think that's a special case... I don't
21 think attempting to munge something meaningful out of the OpenID makes
22 sense... but I think some massaging might need to be done.
24 Ex: I've noticed in the current mode where logging in w/
25 harningt.eharning.us/ shows up in the logs w/o HTTP and if I login w/
26 http://harningt.eharning.us/ is shows up w/ the http... causing some
27 inconsistency. I think it oughtta make sure that it has the properly
28 discovered, canonicalized form (ex: if there's a redirect to another
29 site (harningt.eharning.us -> www.eharning.us) then technically the
30 target site is the 'real' openid (at least according to how most OpenID
35 For OpenID edits, I think there should be a way to tell it what
36 username to show in the preferences dialog (so you can have a 'normal'
37 $USER <$EMAIL> setup.) This could by default be filled in w/ sreg
38 nickname value (as well as email for that matter)...
40 To convey the openid used to make the edit, I think it would be
41 important that some sort of footer line along the lines of the
42 Signed-off: $USER <$EMAIL> conventions I've seen.
44 Perhaps an OpenID: $OPENID_URL would make sense. This could help w/
45 making sure that no one irrefutably spoofs a post by someone (since w/
46 the setup where email and effective username are configurable, there's
47 no determination of uniqueness)
48 >> **adj re git req**: "$OPENID_URL <>"
52 [[madduck]]: git requires `Name <Email@address>` format, as far as I know.
56 > joey@kodama:~/tmp/foo/bar>git commit --author "foo"
57 > fatal: malformed --author parameter
59 > It seems to be happy with anything of the form "foo <foo>" -- doesn't seem to
60 > do any kind of strict checking. Even "http://joey.kitenet.net <>" will be
61 > accepted. --[[Joey]]
67 > I think the thing to do is, as Josh suggested originally, use
68 > GIT_AUTHOR_NAME and GIT_AUTHOR_EMAIL. Note that setting these
69 > individually is best, so git can independently validate/sanitize both
70 > (which it does do somewhat). Always put the username/openid/IP in
71 > GIT_AUTHOR_NAME; if the user has configured an email address,
72 > GIT_AUTHOR_EMAIL can also be set.
74 > There is one thing yet to be solved, and that is how to tell the
75 > difference between a web commit by 'Joey Hess <joey@kitenet.net>',
76 > and a git commit by the same. I think we do want to differentiate these,
77 > and the best way to do it seems to be to add a line to the end of the
78 > commit message. Something like: "\n\nWeb-commit: true"
80 > For backwards compatability, the code that parses the current stuff needs
81 > to be left in. But it will need to take care to only parse that if the
82 > commit isn't flagged as a web commit! Else web committers could forge
83 > commits from others. --[[Joey]]
85 > BTW, I decided not to use the user's email address in the commit, because
86 > then the email becomes part of project history, and you don't really
87 > expect that to happen when you give your email address on signup to a web
90 > The problem with leaving the email empty is that it confuses some things
91 > that try to parse it, including:
92 > * cia (wants a username in there):
93 > * git pull --rebase (?)
94 > * github pushes to twitter ;-)
96 > So while I tried that way at first, I'm now leaning toward encoding the
97 > username in the email address. Like "user <user@web>", or
98 > "joey <http://joey.kitenet.net/@web>".