[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
cvs add & join inconsistency
From: |
Paul Edwards |
Subject: |
cvs add & join inconsistency |
Date: |
Sun, 25 May 2003 02:47:53 GMT |
With reference to:
http://www.google.com.au/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=0UB5a.8169%24VH5.7121%40news-server.bigpond.net.au
is this (patch against cvs1-11-5) what is required to get a conflict?
Note that existing comments say:
/* The file currently exists in the working directory, so we
have a conflict which we can not resolve. Note that this
is true even if the file is marked for addition or removal. */
ie they consider it to be a conflict. But did they expect it to be
marked as a conflict somewhere else? I didn't see too many
direct calls to this Register function to set a conflict. It looks
like conflicts are signified by updating the Entries file. I wonder
how CVS knows whether conflicts have been resolved. Is it
via updated timestamp or the absence of "<<<<<<" markup?
BFN. Paul.
Index: update.c
===================================================================
RCS file: /cvs/ccvs/src/update.c,v
retrieving revision 1.202
diff -c -r1.202 update.c
*** update.c 28 Dec 2002 17:56:59 -0000 1.202
--- update.c 25 May 2003 02:03:37 -0000
***************
*** 2425,2430 ****
--- 2425,2433 ----
"file %s exists, but has been added in revision %s",
finfo->fullname, jrev2);
+ Register (finfo->entries, finfo->file, mrev, vers->ts_rcs,
+ vers->options, vers->tag, vers->date, vers->ts_conflict);
+
return;
}
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- cvs add & join inconsistency,
Paul Edwards <=