[p4] Is there a way to have individual comments in a changeli st? - Gordon Broom - com.perforce.perforce-user

12 messages in com.perforce.perforce-user[p4] Is there a way to have individua...

From	Sent On	Attachments
Tyler, Tom	08 Aug 2001 14:09
Tom Kludy	08 Aug 2001 14:29
Jeff A. Bowles	08 Aug 2001 15:03
Gordon Broom	08 Aug 2001 15:29
Robert Lin	08 Aug 2001 15:37
wiv...@us.itmasters.com	08 Aug 2001 15:42
Mike Castle	08 Aug 2001 16:07
Stephen Vance	08 Aug 2001 16:58
Stephen Vance	08 Aug 2001 17:06
Arnt Gulbrandsen	09 Aug 2001 02:42
Tom Kludy	09 Aug 2001 07:42
Robert O'Brien	10 Aug 2001 03:11

Subject:	[p4] Is there a way to have individual comments in a changeli st?
From:	Gordon Broom (Gord...@pml.com)
Date:	08/08/2001 03:29:38 PM
List:	com.perforce.perforce-user

I tend to agree with Jeff here. If the comment is "THAT important" to someone working on the file, it had better be in the file rather than in the changelist description.

The main reason I lean this way is that a changelist description is metadata, and can be changed at will *at a later date* by a Perforce superuser, and you will have (almost) no record of it being changed. Whereas if the comment is part of the file itself, it cannot change without leaving an auditable trail in the metadata.

I would also recommend that you look at the p4 job mechanism, which gives you a way of grouping related changelists together.

Gord Broom Configuration Management Engineer Power Measurement Ltd. http://www.pml.com/ +1 250 652 7100 x7608

-----Original Message----- From: Jeff A. Bowles [mailto:jab at piccoloeng.com] Sent: August 8, 2001 3:04 PM To: Tom Kludy; perforce-user at perforce.com Subject: RE: [p4] Is there a way to have individual comments in a changeli st?

At 05:29 PM 8/8/2001 -0400, Tom Kludy wrote:

A simple way to get around this is to use good commenting; perhaps a comment at the top of the file that says "revision 3: changed blah", or even better, a comment next to each individual change in the file itself. A simple diff between changes will show the comments.

In all seriousness, I ask the following:

Are the people who use development branches snickering at this thread?

One workaround, I suppose, might be to check in each file separately (with its own change comments) into a branch, and then merge up into the real codeline you want to modify.

I'm afraid that I see this thread as two sides of a potentially heated discussion. I fall onto the "if it's a comment that is THAT important to someone working on the file, make it a real comment in the file, but don't complain about not being able to make file-specific parts of a change description" camp.

Really, are per-file change descriptions that helpful or necessary? If the modification is so important that a later developer will need the per-file change info, shouldn't that just be in a comment of the form: /* Knuth Vol III Algorithm D, Section 6.4. ** This is a double-hash, in which ** we use h1(x) to get the start point and h2(x) ** to set the step size. See the ref for more details. ** Changed 8/2001 (jab) to use this instead of the ** previous more simple-minded hash code that ** never really worked. */ Perhaps we're simply talking about personal styles, but I don't see why asking a developer to add a REAL comment (like above) for a significant file-specific note is a problem.

-Jeff Bowles Perforce Consulting Partner