Re: Successful branch-5-0-stable transplant

[Daniel J S. <dan...@ie...>]

On 10/24/2017 10:10 PM, Daniel J Sebald wrote:
> I'm going to look at that oldest branch and try to make sense of it.  As 
> I said before, I think that all of those checkins, which few if any seem 
> to have a counter-part in the root branch (as typically Ethan would do 
> with bug fixes across root and release branches).

I looked at this unresolved branch earlier this evening.  Within the 
branch itself there are no glaring "jumps", i.e., very large numbers of 
files that change.  I still can't resolve that root for "Windows 
linestyle fix."  From the cvs repository, here are the files that should 
have changed, the number of lines that changed, and the date it was changed:

=============================================================================

RCS file: /cvsroot/gnuplot/gnuplot/win/Attic/wgnuplib.h,v
Working file: win/wgnuplib.h
head: 1.6
branch:
locks: strict
access list:
symbolic names:
	GNUPLOT_RELEASE_3_7_3: 1.4.2.2
	GNUPLOT_RELEASE_3_7_2: 1.4.2.2
	pre-macport: 1.4.2.2
	GNUPLOT_37_20000507: 1.4.2.2
	GNUPLOT_RELEASE_3_7_1: 1.4.2.2
	GNUPLOT_RELEASE_19991103: 1.4.2.2
	branch-pre-3-7-1: 1.4.0.2
[snip, none of these matches 1.4.2.1
----------------------------
revision 1.4.2.1
date: 1999/08/19 14:13:09;  author: lhecking;  state: Exp;  lines: +4 -1
Windows linestyle fix.
=============================================================================

=============================================================================

RCS file: /cvsroot/gnuplot/gnuplot/win/Attic/wgraph.c,v
Working file: win/wgraph.c
head: 1.7
branch:
locks: strict
access list:
symbolic names:
	GNUPLOT_RELEASE_3_7_3: 1.5.2.2
	GNUPLOT_RELEASE_3_7_2: 1.5.2.2
	pre-macport: 1.5.2.2
	GNUPLOT_37_20000507: 1.5.2.2
	GNUPLOT_RELEASE_3_7_1: 1.5.2.1
	GNUPLOT_RELEASE_19991103: 1.5.2.1
	branch-pre-3-7-1: 1.5.0.2
[snip, there's a 1.5.2.1 for GNUPLOT_RELEASE_3_7_1 and 
GNUPLOT_RELEASE_19991103]
----------------------------
revision 1.5.2.1
date: 1999/08/19 14:13:09;  author: lhecking;  state: Exp;  lines: +12 -1
Windows linestyle fix.
=============================================================================

I've done comparisons for checkout's around 1999/08/19 and I don't see 
anything near only a dozen lines changed.  They are all above 1000 lines 
changed relative to the "Windows linestyle fix." version, and the diff 
shows no change for the two files listed above.

I could write a script that will do a comparison against every version. 
The git checkout commands are actually quite fast (fraction of a second) 
so it wouldn't take too long to compare against all versions, but I have 
a feeling there is an issue with the way the branches are constructed 
near the start of the repository.

The one odd thing that stands out is the following kind of pattern at 
the start of CVS files:

----------------------------
revision 1.1
date: 1998/04/15 19:16:26;  author: lhecking;  state: Exp;
branches:  1.1.1;
Initial revision
----------------------------
revision 1.1.1.2
date: 1998/04/15 19:21:47;  author: lhecking;  state: Exp;  lines: +107 -14
Import of beta 343.
----------------------------
revision 1.1.1.1
date: 1998/04/15 19:16:26;  author: lhecking;  state: Exp;  lines: +0 -0
Initial import of beta340.
----------------------------

In the git repository I do not see any changesets listed "Initial revision".

There is this entry that starts the master/trunk:

*** empty log message ***
Lars Hecking<lhe...@nm...>
4/15/98 2:21 PM
Initial import of beta340.
Include stdfn.h.
master (Remove all RCS/CVS cookies.)
origin/master (Remove all RCS/CVS cookies.)
origin/branch-5-2-stable (Release 5.2.1)
3.5 (Content from gnuplot-3.5.tar.gz)
4.0.2 (Massive change to all C sources:)

That trunk is branching from "3.5 (Content from gnuplot-3.5.tar.gz)" 
while all the others branch from "Import of beta 347.".  I wonder if the 
"Initial revision" files should have come ahead of all the "Import ..." 
changes.  For example, above both

revision 1.1
revision 1.1.1.1

have the same date/time stamp, which could lead to an arbitrary choice. 
Not sure what the issue is, but I'm confused about "Initial revision" 
not showing up as a changeset anywhere.

Dan