Has Media Center support been dropped now?

@Tomcat76,

70208a works without incident. Thanks!

Bug Report: I just wanted to make the IE7 executable with all updates, and for some reason modifyPE is required. I'll put modifyPE in HFTOOLS to bypass the error.

the_guy

EDIT: Maybe I should've read the information better. Why is modifyPE needed anyway?

If IE7 hotfixes are found in the HF folder, a new IE7 installer is created in which those hotfixes are slipstreamed. This is done with iexpress.exe, but Windows setup won't allow the output file to be copied over from CD to the hard disk if it isn't modifyPE'd first.

ModifyPE is only really needed if doing a non-SVCPACK install (not specifying IE7SVCPACK=1) of the newly created IE7 installer but I decided to keep things more simple by making it a prerequisite either way.

Thanks Tomcat.

Also, for some reason HFSLIP wanted to extract and recompress driver.cab. Is there any way that driver compression can be skipped over if no files need to go into the driver file and selected compression method is A?

the_guy

the_guy - if you have a slipstreamed source, there is typically a SP#.CAB, where the # is the service pack that is slipsteamed. The SP#.cab files have some binaries that are updates of the driver.cab binaries. With 2k, some of the dx9 binaries have to get packed in the driver.cab. Sometimes, there are hotfix binaries that have to get packed into the driver.cab too. Perhaps there is a workaround to create an optional hfslip pre-processing script to merge the sp#.cabs and driver.cab. If a hotfix and/or hfexpert drivers needs to get added during the normal hfslip process, an spx.cab will be created. Whattaya think? TC?

@tommyp: I understood that. What I meant is to not have driver processing if you just have IE7 updates in HF, or no files in HF, and option A was chosen for driver compression. What could be done is that there could be a variable in HFANSWER.INI such as RECOMPDRIVER=0 to cancel driver recompression, with the value ignored if DRIVERCOMP wasn't A.

the_guy

@the_guy

Check out the HFSLIP Preparation Tool

Interesting tool!

the_guy

hi TC! my apologies if what i'm sharing here were based on several results from testing 70203b ...
(1) posted some dotnet issues in Unattended forum: Silent .NET Maker (latest update: 20070201), Create silent switchless .NET installers for Win2K/XP/2K3 x86 ... could it be posted here in Test Release or in the new topic you started: How to slipstreasm .Netframework 1.1 and 2.0
(2) Windows PowerShell 1.0 English-Language Installation Package for Windows XP (KB926139) didn't install ... tried it in HF & HFGUIRUNONCE ... no go from both locations ... i recall trying to slipstream this before but you informed me back then it wasn't supported ... until i read in
(3) are there specific instructions in using HFGUIRUNONCE that i might have missed? do i really have to execute these files manually while the cmd window is active?
(4) are there specific HFSVCPACK_SW to do's prior to putting .msi files here? TimeZoneSetup.msi & UFDSetupWizard.msi didn't install (even if placed in HFGUIRUNONCE ... this was despite having dotnet1 & dotnet2 in HFSVCPACK (btw, aren't DNF11.exe & DNF20.exe, being switchless installers supposed to be placed in HFSVCPACK_SW?)
(5) is WindowsDesktopSearch-KB926356-V2-x86-ENU-6.0.5414.0 supported? i recall doing this the unorthodox way via HFSVCPACK before & if i recall it correctly, it worked ... officially though, is this download unsupported?
(6) i'm not sure if this has something to do with HFSLIP but installing drivers upon hitting GUI was problematic for me, specifically for older Install Shield Wizard ... they get stuck in the askbar without terminating automatically ... i always ended up rebooting the pc ...

please advise ...
thanks!

There are several problems with HFGUIRUNONCE in the previous test releases that have been fixed. Please use the latest version.

Any issues with HFGUIRUNONCE should be posted in this thread until this feature is available in a "final" version of HFSLIP. After that, start a separate thread. The main post in the .NET 1.1/2.0/3.0 thread is intended as an informative post.

Any hotfix goes in HF except otherwise noted. This is mentioned on the Basic how-to page and in several other locations. So the Power Shell installer goes in HF, and .NET 2.0 must be placed in HFSVCPACK. HFSLIP will instruct Windows setup to install Power Shell at first GUI logon.

HFSVCPACK_SW is for installers that need switches.

TimeZoneSetup.msi and UFDSetupWizard.msi require .NET 1.1. A limitation in HFSLIP requires you to place .NET 1.1 in HFSVCPACK and those two MSIs in HFGUIRUNONCE. I'll fix this in the next version so that you can place those MSIs in HFSVCPACK_SW.

Windows Desktop Search is unsupported as a hotfix. All binaries inside this executable need to go into Program Files and a lot of them exceed the 8.3 file naming standard, so support would have to be hard-coded into HFSLIP. You should try to find an addon. I'll make one if I have the time.

Don't know about 6). Try the latest test version with every file in the correct place.

MSIFIRSTLOGON won't return because it's less powerful than the HFGUIRUNONCE folder.

thanks TC! i'll try my luck again ... ... i've posted some concerns about dotnet installers too
HFGUIRUNONCE seems to be quite difficult to tame for me ... any additional info about HFGUIRUNONCE will definitely be helpful ...
would any of the HFs have anything to do with this?
Click to view attachment
this one comes up before showing the desktop
oops, i suddenly remembered, cmd window still comes up of UPHClean-Setup.msi ...
are we waiting for the new MS HFs to be released before our new final? just asking ...

Answered in the other thread.

Yes; WindowsXP-KB922120-v5-x86-ENU.exe.

Yes. And also for feedback on the usability of the "alternate input folders" feature.

hi again TC! my apologies if i've brought over some Qs i posted in another thread ...
Click to view attachment

Click to view attachment
Click to view attachment

Your comments are confusing!

Have you tried 70208a yet? Please use that version with all the files in the right places, like Tomcat76 said. Also, make sure your .NET files for the Silent .NET maker are in the correct location, the correct entries in the answer file are set, and ALL YOUR FILES are in the right place!

perhaps because of the quotes but my questions were really simple ... perhaps even too elementaryi certainly will ... when using HFGUIRUNONCE, does it really show a cmd window while still in the installation process ... you see that's what happened to me ... i'm not certain though if it was before T-13 or rightafter this part when cmd window came up ... i had to manually type names of .msis & .exe at the prompts (for those placed in HFGUIRUNONCE directory) for it to be executed ... i think there was 1 exe: DNF30 & 2 msis: timezone & ufd ... when everything else finished installing, i simply typed EXIT after this prompt: C:\WINDOWS\HFSLPGUI> to exit from this window to proceed with the normal installation of Windows ... there were 2 distinct results when putting (1) DNF11_30.exe or (2) DNF30.exe in HFGUIRUNONCE ... with (1) cmd window comes up with this prompt D:\$OEM$> while with (2)C:\WINDOWS\HFSLPGUI> ... you see i was under the impression that since this was for RunOnce, it should be prior to showing the desktop ... that's why i asked if it was how apps installation behaved for using HFGUIRUNONCE ... thanks!

I've never gotten command prompt windows during HFGUIRUNONCE. Try including cmdow in HFTOOLS if you haven't already.

The latest test release changes the order in which EXEs and MSIs in HFSVCPACK_SW are installed. Maybe that will help.

thanks S-M! never have removed cmdow tool in HFTOOLS ... i think dealing with .exes & .msis as shown in changelog is great ... i'll try the latest test release ... i'm readying myself too for the new HFs that MS would be releasing soon

Yep, MS is readying about a dozen of them patches.

Hi all,

any chance in having Branding.cab supported again for IE7 as IE7 is "recompiled" for updates ?

I'd also ask for an update in the ALTDIR folders, can you inclue a function to have different files/folder in HF :

I'm building CDs for XP Pro and XP Home.
XP Home has issues with WMP11, witch make me have to remove \WMP11 and WindowsXP-KB898543-v3-x86-ENU.exe from the HF folder and add mp10setup.exe, HMTCDWizard_FRA.exe, windowsmedia10-kb902344-x86-intl.exe, WindowsMedia10-KB917734-x86-FRA.exe, WindowsMedia-KB891122-x86-FRA.exe, WindowsXP-KB923689-x86-FRA.exe Into HF.
Can ou include a exclusion/remplace list/function for the HF folder . (Currently i've build an HF Common folder, along with HFWMP11 an HFWMP10 and a batch file that copies/replace file in the HF folder)

Thanks in advance.

Hi Tomcat, I have tested a bit this release, I have an issue (I think). I have used the tool for compact DRIVER.CAB before run hfslip script. When hfslip script goes to DRIVER.CAB repacking it shows this.

Click to view attachment

There is an issue or not? The script recompacts driver.cab even if I haven't put any new driver (I think)

hi S3pHiroTh! i'm not certain about this but wouldn't this:have anything to say about 0 files copied? which merging option did you use?

Before, when I had tested the previous version of hfslip, instead of "0 files copied" there was some files... but today, I have used HFSLIP_prep.cmd before of hfslip-70211b.cmd. So I think is a new issue related to DRIVER.CAB that is been compressed before hfslip script.
I have set this in HFANSWER.INI: DRIVERCOMP=A

S3pHiroTh--

I'm almost positive those errors don't come from the sections that handle DRIVER.CAB. Can you please try HFSLIP 70212a? It shows a bit more output text for the sections where those errors could be coming from. Maybe that'll help...

EmRoD--

You got BRANDING.CAB. I'll think about support for different HF folders but it won't be for the next final.

Slight bug with the NOLOGCOPY feature. I do not believe this was intended.

The log is no longer copied to SOURCESS until after the ISO is created. That essentially removes the usefulness of the variable and makes that feature the "default." This occurs regardless of whether the variable is actually set as there is no longer a copy command for HFSLIP.LOG before :MAKEISO is called.

Not really a bug with that feature specifically -- more a problem with the order of actions. But you're right: it's wrong

Fixed in 70212b... Hopefully...

@Tomcat76 thanks

Ok, I have tested once again with the latest build release, but the results are the same. I'll post the output messages that hfslip write, so, maybe, you can understand why this happen. (Can this happen if I slipstream, for example, KB888111 hotfix?)

been slipstreaming 888111 since it has been supported and i never had your results

This results are after I have compressed DRIVER.CAB with the tool that Tomcat has released some days ago. HFSLIP prep.cmd
Have you read my previous replies? The issue is ONLY when SP2.CAB doesn't exist because is already slipstreamed into DRIVER.CAB

new file still goes into HFCLEANUP, right?this is better! wouldn't just simply show in HFSLIP.log "HFCLEANUP detected" ... thanks!

It wasn't supposed to be fixed. I just added more text output (titles).

Can you give me the DRVINDEX.INF file that's in your SOURCE\I386 folder?

Added as you wish
I have uploaded the old one (before hfslip prep script) too.

Well, I'm stumped. Everything is looking fine...

Try this special version which echoes even more text.

Hi, tested the special version

I have see that some files are copied in DRIVER.CAB sections, but it displays "0 files copied". Maybe you can understand better

Weird. The only logical explanation I can give is that the files in your DRIVER.CAB package are fully up-to-date. Did you use a clean source with the prep tool?

Yes, it is clean. One thing I don't understand, why hfslip script recompress driver.cab if I don't add drivers? I have used hfslip prep tool because I test every version of the test build (and for corrections of my personalized windows cd), but if driver.cab will recompress is only a waste of time to use first the prep tool and after hfslip script.

The only thing that I have added to source directory is $OEM$ directory, that's all.

You have a full HF folder. Some hotfixes update driver binaries.

Ah ok

Nice to meet you.
Because a certain phenomenon was noticed, it reports.

The end code "0x1a" has not adhered at the end of file TXTSETUP.SIF generated with
HFSLIP.
It is in the one of the SOURCE folder.

The influence by this is uncertain now. However, it seems that you should not miss because it is confirmed not to be able to continue setting up normally with the file before it applies by cutting down this end code.

Are you talking about that little square at the bottom of the original TXTSETUP.SIF?

Well... With hundreds of runs with Windows 2000 SP4, XP SP1, XP SP2, 2003 Gold and 2003 SP1, I can positively say that it isn't required. Also, HFSLIP has been around for a very long time; people would've reported such problem...

Something weird happening here, I'm using 70212b with a hfslip-preped source; I'm was testing the iso creation with bbie.exe, mkisofs.exe and cygwin1.dll; after finishing creating the SOURCESS folder (which had driver.cab) the iso creation process started and then finished normally, but when I checked the SOURCESS folder driver.cab was nowhere to be found, like it was deleted by the iso creation process. (Just after the normal process said it copied driver.cab, I checked SOURCESS and it was there).

Alright. That's a very specific case. Should be fixed in 70214a.

Keep these bug reports coming...

Problem solved, thank you.

I have tested the last build release with February updates and I have had a strange issue. When I run Windows Update it shows 2 updates for KB890830.
Click to view attachment

Anyone has this little issue?

Tested hfslip-70212b with all the Feb hotfixes on XP pro, corp, and Home without issues

no issues with KB890830

After installing XP, with the 1.3 rc1, I get notified by Windows Updates that I don't have KB920872 and KB900485 even tough I have them both on my HF folder. Here is my HFSLIP.log. I have used a hfsliped-preped source BTW.

EDIT: After checking the log file I created I see this:

"
.....
Expandiendo source\i386\driver.cab a sourcess\i386\driver\xxwh1hlp.hlp.
Expandiendo source\i386\driver.cab a sourcess\i386\driver\zlogic.cyz.

4778 total de archivos.
Adding updated driver files
0 archivos copiados // copied files
0 archivos copiados
0 archivos copiados
0 archivos copiados
0 archivos copiados
0 archivos copiados
0 archivos copiados
0 archivos copiados
0 archivos copiados
0 archivos copiados

Basic cleanup"

So I guess the new .sys files aren't being copied to driver.cab.

Neo2x--

I've done extensive testing with those hotfixes, with merging options A and F, with and without the prep tool, and I can't duplicate this problem.

Does the problem still occur if you recopy the SOURCE from CD and just use the latest test release?

No it doesn't... In that moment the script copy some files into DRIVER.CAB working directory then recompress...