don't know about XP, but for 2000, my vote is for a TZSLIP2GO
I tried to unravel the HFSLIP source code once back when my brain was missing only a cylinder or two
HFSLIP had to be written before the days of rollups by tomasz86, wildbills, bristols, etc
We all benefitted!!
But, it had to be written for the general case of all possibilities, for all persons, for all situations, etc.
it seemed an impossible task, so it ended up in x-levels deep of FORs, SORTs, IFs, pipes, and ...
But that predicate, of scripting for unknown numbers of future unknowns is no longer the case
instead of 400 files with unknown names, we now have the case of updateRollup v1 or v2
all with known names and switches
I'm not a programmer so I should probably just keep my stupid mouth shut, but
Let a future TZSLIP2GO be divided into two parts
a core part -- TZSLIP2GO_partA_corefiles_v1.exe
and an options part TZSLIP2GO_partB_optionfiles_v1.exe
TZSLIP2GO_partA_corefiles_v1.exe
asks user to point to a place to setup TZSLIP root dir
use new name for root dir to prevent accidental use by HF
... I suggest TZSLIP, no disrespect to others
Then build the dir-file structure using current HFSLIP rules and habits,
then, load TZSLIP root dir with core files all of known fixitude,
basically tomasz86's current root dir for HF
but with daily rollups, patches, etc moved to partB
examples, w2ksp51, directX, media9, mdac, ...
Henry Ford used to say ...You can have any color car you want, as long as it's black
TZSLIP2GO_partB_optionfiles_v1.zip
asks user to point to the TZSLIP root dir
then builds dir-file structure for a new batch file
adding rollups, options, .net frameworks, etc.
TZSLIP2GG.cmd is a new batch written from scratch
no disrespect to the HF gods, but the future is not so unknown anymore
part A core files probably has 80 - 100 files
all are known by now
HF has to assume for every case, with unknowns adinfinitum
part B probably has 80-100 of rollups and option files
each of them is also known, with known switches
Abandon HFSLIP which has evolved into a hairball, now at version J
the new batch doesn't need not be written for every possible circumstance
rollups are actually written to solve the HFSLIP hariball problem
a new simple script would be maybe 300 lines
they could be linear and stupid with no sequencing problems
as new rollups are added, maybe one or two lines in the script need to change
each script included in a partB would already know exactly what's in itself, and in partA
and it would know the right sequences, and integrate switches
A TZSIP2GO would take the frustration out of 2k testing for noobs like me
and that would allow broad testing of rollups working towards SP52
as Henry Ford used to say ...You can have any color car you want, as long as it's black
This post has been edited by Molecule: 07 September 2012 - 11:52 AM