14 messages in com.perforce.perforce-user[p4] VSS-like "Links" in Perfoce?| From | Sent On | Attachments |
|---|---|---|
| Tim Henrion | 18 Sep 2000 05:40 | |
| Steve Bennett | 18 Sep 2000 07:29 | |
| Jeff A. Bowles | 18 Sep 2000 10:13 | |
| Jeff A. Bowles | 18 Sep 2000 10:44 | |
| Russell Jackson | 18 Sep 2000 10:50 | |
| Jeff A. Bowles | 18 Sep 2000 10:58 | |
| Tim Henrion | 18 Sep 2000 13:19 | |
| Steve Bennett | 18 Sep 2000 13:59 | |
| Russell Jackson | 18 Sep 2000 14:08 | |
| Richard Brooksby | 18 Sep 2000 14:09 | |
| Steve Bennett | 18 Sep 2000 14:09 | |
| George Van Treeck | 18 Sep 2000 21:45 | |
| Berend de Boer | 19 Sep 2000 13:46 | |
| Richard Brooksby | 19 Sep 2000 14:26 |
| Subject: | [p4] VSS-like "Links" in Perfoce?![]() |
|---|---|
| From: | Jeff A. Bowles (ja...@pobox.com) |
| Date: | 09/18/2000 10:13:24 AM |
| List: | com.perforce.perforce-user |
At 08:41 AM 9/18/2000 -0400, Tim Henrion wrote:
Does anyone have a good method for duplicating Visual SourceSafe "links" in Perforce? I'd like to have links to global include files exist in multiple directories in my depot. All of the links essentially point to the same file. Whenever someone checks out/changes/checks in the file via any of the links, all links to the file automatically get updated to the new contents.
Look at the following scripts in the public.perforce.com:1666 depot: //guest/jeff_bowles/scripts/p4push.dat //guest/jeff_bowles/scripts/p4push.pl These implement the model that you can have a ``master copy'' of a file that gets updated, and propogates the changes to another set of files. The model is that one file changes, the rest are downstream copies that get updated as the result of an integrate.
You would just run this from crontab(1) or "at" or perhaps a "p4 review" script. And you might use "p4 protect" to forbid subordinate copies from being directly updated/edited.
Don't infer that I like this model: a couple of users asked me for this. I think that updating a file in location A and having three files in locations (B,C,D) get updated is poor form: the Perforce model is far more of the "you only update that which you mean to, and you do it explicitly."
So, use this as a stop-gap measure and then change your procedures.
-jab




