You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Neels Janosch Hofmeyr <ne...@hofmeyr.de> on 2009/06/03 20:03:50 UTC

naming in repos_diff.c

Hi Arfrever, Stsp,

would you guys be fine with the attached patch? Calling that thing
"diff_editor" got me confused there for a while.

Furthermore, could you explain why you chuck the local svnpatch_* functions
into an (also local) svn_delta_editor_t in the first place? It makes much
more sense to me to just call them directly without the editor struct.

Thanks,
~Neels

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359211

Re: [OT] channel moderation -- was: Re: naming in repos_diff.c

Posted by "C. Michael Pilato" <cm...@collab.net>.
Branko Cibej wrote:
> Stefan Sperling wrote:
>> On Thu, Jun 04, 2009 at 03:25:30PM +0200, Neels Janosch Hofmeyr wrote:
>>   
>>> Yeah, I was wondering myself just now. Accidentally sent it from my
>>> "private" mail address, while I'm subscribed with my "work" address.
>>>
>>> And this leads me to another question: How can I see whether a mail went
>>> through the moderating channel anyway?
>>>     
>> No idea. But to prevent this from happening in the future, you can
>> log into tigris, go to http://www.tigris.org/servlets/UserEdit
>> and add your private address as a "secondary" email address on that page.
>>   
> 
> Why? I thought that once a mail from an address was moderated through,
> all others from that address would skip the moderation queue. At least
> that's how it used to work before the famous mailing system rewrite.

The current moderation methods match the previous ones.  There are
individual moderator responses for "reject", "accept", and "allow future
posts from this address".  So it kinda depends on the moderators doing the
right thing here.

-- 
C. Michael Pilato <cm...@collab.net>
CollabNet   <>   www.collab.net   <>   Distributed Development On Demand

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359462

Re: [OT] channel moderation -- was: Re: naming in repos_diff.c

Posted by Mark Phippard <ma...@gmail.com>.
On Thu, Jun 4, 2009 at 10:39 AM, Branko Cibej<br...@xbc.nu> wrote:
> Stefan Sperling wrote:
>> On Thu, Jun 04, 2009 at 03:25:30PM +0200, Neels Janosch Hofmeyr wrote:
>>
>>> Yeah, I was wondering myself just now. Accidentally sent it from my
>>> "private" mail address, while I'm subscribed with my "work" address.
>>>
>>> And this leads me to another question: How can I see whether a mail went
>>> through the moderating channel anyway?
>>>
>>
>> No idea. But to prevent this from happening in the future, you can
>> log into tigris, go to http://www.tigris.org/servlets/UserEdit
>> and add your private address as a "secondary" email address on that page.
>>
>
> Why? I thought that once a mail from an address was moderated through,
> all others from that address would skip the moderation queue. At least
> that's how it used to work before the famous mailing system rewrite.

It is still the case.  Just as with the old system, Reply to a
moderation email allows that one message, and Reply to All, allows
that message and all future messages.

-- 
Thanks

Mark Phippard
http://markphip.blogspot.com/

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359463

Re: [OT] channel moderation -- was: Re: naming in repos_diff.c

Posted by Branko Cibej <br...@xbc.nu>.
Stefan Sperling wrote:
> On Thu, Jun 04, 2009 at 03:25:30PM +0200, Neels Janosch Hofmeyr wrote:
>   
>> Yeah, I was wondering myself just now. Accidentally sent it from my
>> "private" mail address, while I'm subscribed with my "work" address.
>>
>> And this leads me to another question: How can I see whether a mail went
>> through the moderating channel anyway?
>>     
>
> No idea. But to prevent this from happening in the future, you can
> log into tigris, go to http://www.tigris.org/servlets/UserEdit
> and add your private address as a "secondary" email address on that page.
>   

Why? I thought that once a mail from an address was moderated through,
all others from that address would skip the moderation queue. At least
that's how it used to work before the famous mailing system rewrite.

-- Brane

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359456

Re: [OT] channel moderation -- was: Re: naming in repos_diff.c

Posted by Stefan Sperling <st...@elego.de>.
On Thu, Jun 04, 2009 at 03:25:30PM +0200, Neels Janosch Hofmeyr wrote:
> Yeah, I was wondering myself just now. Accidentally sent it from my
> "private" mail address, while I'm subscribed with my "work" address.
> 
> And this leads me to another question: How can I see whether a mail went
> through the moderating channel anyway?

No idea. But to prevent this from happening in the future, you can
log into tigris, go to http://www.tigris.org/servlets/UserEdit
and add your private address as a "secondary" email address on that page.

Stefan

[OT] channel moderation -- was: Re: naming in repos_diff.c

Posted by Neels Janosch Hofmeyr <ne...@elego.de>.
Branko Cibej wrote:
> Edmund Wong wrote:
>> Neels Janosch Hofmeyr wrote:
>>   
>>> Hi Arfrever, Stsp,
>>>
>>> would you guys be fine with the attached patch? Calling that thing
>>> "diff_editor" got me confused there for a while.
>>>
>>> Furthermore, could you explain why you chuck the local svnpatch_* functions
>>> into an (also local) svn_delta_editor_t in the first place? It makes much
>>> more sense to me to just call them directly without the editor struct.
>>>
>>> Thanks,
>>> ~Neels
>>>
>>> ------------------------------------------------------
>>> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359211
>>>
>>>     
>> Pardon my offtopicness, but I don't get why this message went
>> through the moderator's channel.  Normally Neels' messages
>> go directly to the list without the need for moderation.  This
>> went through the mods channel.   Weird.
>>
>> Clarifications appreciated.
>>   
> 
> Could be because he didn't send it from his subscribed address (@elego.de)?

Yeah, I was wondering myself just now. Accidentally sent it from my
"private" mail address, while I'm subscribed with my "work" address.

And this leads me to another question: How can I see whether a mail went
through the moderating channel anyway?

;)
Thanks
~Neels

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359429

Re: naming in repos_diff.c

Posted by Branko Cibej <br...@xbc.nu>.
Edmund Wong wrote:
> Neels Janosch Hofmeyr wrote:
>   
>> Hi Arfrever, Stsp,
>>
>> would you guys be fine with the attached patch? Calling that thing
>> "diff_editor" got me confused there for a while.
>>
>> Furthermore, could you explain why you chuck the local svnpatch_* functions
>> into an (also local) svn_delta_editor_t in the first place? It makes much
>> more sense to me to just call them directly without the editor struct.
>>
>> Thanks,
>> ~Neels
>>
>> ------------------------------------------------------
>> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359211
>>
>>     
>
> Pardon my offtopicness, but I don't get why this message went
> through the moderator's channel.  Normally Neels' messages
> go directly to the list without the need for moderation.  This
> went through the mods channel.   Weird.
>
> Clarifications appreciated.
>   

Could be because he didn't send it from his subscribed address (@elego.de)?


-- Brane

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359279

Re: naming in repos_diff.c

Posted by Edmund Wong <ed...@kdtc.net>.
Neels Janosch Hofmeyr wrote:
> Hi Arfrever, Stsp,
> 
> would you guys be fine with the attached patch? Calling that thing
> "diff_editor" got me confused there for a while.
> 
> Furthermore, could you explain why you chuck the local svnpatch_* functions
> into an (also local) svn_delta_editor_t in the first place? It makes much
> more sense to me to just call them directly without the editor struct.
> 
> Thanks,
> ~Neels
> 
> ------------------------------------------------------
> http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359211
> 

Pardon my offtopicness, but I don't get why this message went
through the moderator's channel.  Normally Neels' messages
go directly to the list without the need for moderation.  This
went through the mods channel.   Weird.

Clarifications appreciated.

Edmund

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359264

Re: naming in repos_diff.c

Posted by Stefan Sperling <st...@elego.de>.
On Thu, Jun 04, 2009 at 05:16:11PM +0200, Neels Janosch Hofmeyr wrote:
> Stefan Sperling wrote:
> > But I'd also like to point out (again) that I don't like the fact that
> > "patch" implements its own editor. Because that duplicates a lot of code,
> > and it's hard to maintain in the long term.
> 
> Which editor in particular are you referring to?

I was referring to the one in libsvn_client/patch.c actually.
You probably haven't seen that one yet? :)
It might not be directly related to your patch.

> Clarifying: `svn patch' applies a patch file to a working copy, right?

svn patch does more than just that.

It also interprets special "SVNPATCH" blocks inside the unidiff file,
which contain encoded svn protocol commands to describe file additions
and deletions. It uses the editor interface to apply svn protocol
commands encoded in those SVNPATCH blocks. See notes/svnpatch for details.

> So why is there "svnpatch" code in diff?

That I don't really know, because I've not yet looked into how
exactly svnpatch is implemented. But it must have got something
to do with applying SVNPATCH blocks.

> Because `merge' also uses the diff
> framework, or because `svn patch' also needs to do stuff during `svn diff'?

I'd say because merge uses the diff callbacks, yes.
The editor in lisvn_client/patch.c also implements the diff callbacks.

> And, how does all this relate to your comments above? I'm not really
> grokking the interactions yet.
> 
> I'm on the ambitious path of trying to disentangle diff from merge in the
> process of enabling arbitrary diffs and introducing editor v2... I'm still
> at the drawing board, so I'm grateful for any details you've got :)

I've already told you all that I know :)

Stefan

Re: naming in repos_diff.c

Posted by Neels Janosch Hofmeyr <ne...@elego.de>.
Stefan Sperling wrote:
> On Wed, Jun 03, 2009 at 10:03:50PM +0200, Neels Janosch Hofmeyr wrote:
>> Hi Arfrever, Stsp,
>>
>> would you guys be fine with the attached patch? Calling that thing
>> "diff_editor" got me confused there for a while.
>>
>> Furthermore, could you explain why you chuck the local svnpatch_* functions
>> into an (also local) svn_delta_editor_t in the first place? It makes much
>> more sense to me to just call them directly without the editor struct.
> 
> I'm fine with this patch.
> 
> But I'd also like to point out (again) that I don't like the fact that
> "patch" implements its own editor. Because that duplicates a lot of code,
> and it's hard to maintain in the long term.

Which editor in particular are you referring to?

> 
> E.g. it's not clear to me yet whether the patch editor should be caring
> about tree conflicts, for example, which the update and merge editors do.
> 
> Maybe patch should somehow re-use the merge editor.
> I've been told this is impossible with the current API.
> 
> Anyway I think we should solve this problem before the patch feature
> gets released. So I hope that the code you're patching will be obsoleted
> eventually. But maybe that's just me :)
> 
> Stefan
> 

Clarifying: `svn patch' applies a patch file to a working copy, right?
So why is there "svnpatch" code in diff? Because `merge' also uses the diff
framework, or because `svn patch' also needs to do stuff during `svn diff'?

And, how does all this relate to your comments above? I'm not really
grokking the interactions yet.

I'm on the ambitious path of trying to disentangle diff from merge in the
process of enabling arbitrary diffs and introducing editor v2... I'm still
at the drawing board, so I'm grateful for any details you've got :)

~Neels

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=2359469

Re: naming in repos_diff.c

Posted by Stefan Sperling <st...@elego.de>.
On Wed, Jun 03, 2009 at 10:03:50PM +0200, Neels Janosch Hofmeyr wrote:
> Hi Arfrever, Stsp,
> 
> would you guys be fine with the attached patch? Calling that thing
> "diff_editor" got me confused there for a while.
> 
> Furthermore, could you explain why you chuck the local svnpatch_* functions
> into an (also local) svn_delta_editor_t in the first place? It makes much
> more sense to me to just call them directly without the editor struct.

I'm fine with this patch.

But I'd also like to point out (again) that I don't like the fact that
"patch" implements its own editor. Because that duplicates a lot of code,
and it's hard to maintain in the long term.

E.g. it's not clear to me yet whether the patch editor should be caring
about tree conflicts, for example, which the update and merge editors do.

Maybe patch should somehow re-use the merge editor.
I've been told this is impossible with the current API.

Anyway I think we should solve this problem before the patch feature
gets released. So I hope that the code you're patching will be obsoleted
eventually. But maybe that's just me :)

Stefan