193 replies to this topic

[pscEx]

Do you mean to install a firadisk driver in the PE3?

 

Peter

[TrungNT_HUST]

Yes ! But the problem is I want it to load at startup WinPE !

[pscEx]

A driver can be "already installed" in a booting PE, when some registry enries are added to the PE boot ISU / disk.

 

When you mean that, I can try to realise.

 

But take care, that there is "only the working driver", nothing it mounts etc. at boot time.

 

Peter

[TrungNT_HUST]

Drivers can MiniXP installed with setup file txtsetup.sif

 

It works very well and efficiently. With Win PE 3.0 You have a way to load a driver as MiniXP

[steve6375]

If you use Easy2Boot with a removable USB drive, you just name the PE .iso file as .isope01 and when PE boots it will automatically load the PE ISO as drive Y:

Is that what you want?

[niche99]

Hi Peter,

 

Trying to build win7pe. I get this error:-

 

You need a version 2013.11.01 of WinBuilder or higher, in order to use all features of the installed win7pe project!

 

Where can i get the new verion? Can WinBuilder update itself? If so how?

 

Regards,

niche99

[pscEx]

As you can see, that is not an error. It is a warning, in this case a hint.

 

The new version is not yet published, but do not worry: Win7PE has workarounds, to produce the wanted PE with the current WinBuilder version. (look for hasOldWB).

 

"update itself":

 

The new (unpublished) WinBuilder will give a hint about a newer available WinBuilder version. You can download and auto install with the WinBuilder command UPDATE.

Additionally it will (config dependent) auto update Win7PE and plugins.

 

Peter

[niche99]

Hi Peter,

 

Thanks for the prompt reply.

You imply that I should be able to build with the old version of WinBuilder. However, the build will not complete. After the message about a newer version being required, I get a build ABORTED message.

Your previous information about hasOldWB means nothing to me, I can see it is present in 4 .bsh files in the win7pe project folder, but what I am supposed to do with it I do not know. I need more detailed nformation.

 

As I said the build aborts and goes no further. How to get a sucessful build?

 

Regards,

niche99

[Brito]

Peter, would you please disable the warning hint? This scares and confuses users.

 

Also, the download package uses "winbuilder_lastest" and this could be simplified. Could either be simply called "winbuilder" or add the version number like "winbuilder-2014.01.01" to ease referencing but "latest" is not so useful as reference after being downloaded to disk by end-users and causes confusion about what is being used since it is called as the "latest".

 

As I said the build aborts and goes no further. How to get a sucessful build?

 

Hi Niche,

 

For the moment don't look inside any .bsh file. First let's diagnose what is going on. For example, what is you log saying exactly?

 

I've tested the winbuilder project from a brand new folder and it worked without trouble. I see the messages you refer, they really are a bit misleading but don't prevent the ISO from being generated. Here is an example of my log:

21:38:01 [COMPLETED] The WIM file is available at D:\WinBuilder\2014-01-01\wb\output\bootdisk\sources\boot.wim and is sized in 274 Mb.
21:38:02 [RUNNING] Creating the ISO image at D:\WinBuilder\2014-01-01\wb\output\bootdisk.iso
21:38:09 [COMPLETED] The ISO file is available at D:\WinBuilder\2014-01-01\wb\output\bootdisk.iso and is sized in 279 Mb.
21:38:09 [INFO] *** You need a version 2013.11.01 of WinBuilder or higher, in order to use all features of the installed Win7PE project!
21:38:09 [INFO] *** A newer version of Win7PE is available for download!
21:38:09 [INFO] *** A newer version of Win7PE is available for download
21:38:09 [INFO] *** A newer version of Win7PE is available for download
21:38:09 [COMPLETED] Project build was concluded in 2 minutes and 53 seconds.

[pscEx]

ABORTED is usually only be done by the user.

 

Unfortunatelly we do not have a log file yet.

 

Can you post the (or some) screenshoot of the console window, with removed ART-ASCII lines?

 

The "hasOldWB" was only a hint for developers to look at the code what happens with old / new WinBuilder. Nothing to do.

 

Peter

[pscEx]

 

Peter, would you please disable the warning hint? This scares and confuses users.

 

Also, the download package uses "winbuilder_lastest" and this could be simplified. Could either be simply called "winbuilder" or add the version number like "winbuilder-2014.01.01" to ease referencing but "latest" is not so useful as reference after being downloaded to disk by end-users and causes confusion about what is being used since it is called as the "latest".

 

Part 1: I disagree. Maybe Win7PE does not work completelly like intended and developed for the 2013.11.xx WinBuilder when it uses the current WinBuilder version.

I tried to catch all possible issues in the workarounds, but I'm not sure whether everything is targeted.

IMO the warning is necessary.

 

Part2: done

 

Peter

[niche99]

Hi Peter,

 

I activated all components.

I see the following error:

[ERROR] Error processing MMC

Then message about newer version.

[ABORTED] Due to fatal errors project build was aborted after 53 seconds.

 

Seems to be a problem with the MMC component.

 

Edit: Confirmed. Deactivating the MMC component allows the build to complete sucessfully.

 

Regards,

niche99

[pscEx]

Thanks!

 

Can you do "CONFIG win7pe" and unselect MMC?

 

Maybe that is my fault. In factory settings MMC should be unselected.

 

Peter

[Brito]

Part 1: I disagree. Maybe Win7PE does not work completelly like intended and developed for the 2013.11.xx WinBuilder when it uses the current WinBuilder version.

 

What is the practical advantage of having them?

 

On the current state, I just see users being alarmed about something that is "not right" and pointing them to something not released. From my perspective, this is not a good thing and should simply be removed.

 

We need stability, we should never have these kind of warning signs scaring users and chasing them to get ghost versions. We had these same problems in the past with different beta and RC editions that caused quite a mess. Let's please avoid this situation, we can do better.

[pscEx]

What is the practical advantage of having them?

 

e.g modifyMulti could be executed inside WinBuilder rather than by a win7pe function

Spoiler

// for compatibility with old WinBuilder
/**
* workaround for registry multi_string bug
*/
void processMulti(RegHeader currentHive, String key, String value, String data) {
if (currentHive.existVal(key, value)) {
String current = currentHive.getValueString(key, value);
String[] actList = current.split(", ");
String[] newList = data.split(", ");
current = current.replace(", ", ",");
for (int n = 0; n < newList.length; n++) {
boolean found = false;
String cand = newList[n].trim().toLowerCase();
for (int a = 0; a < actList.length; a++) {
String act = actList[a].trim().toLowerCase();
if (cand.equals(act)) {
found = true;
break;
}
}
if (!found) {
current = current + "," + newList[n].trim();
}
}
currentHive.modifyValue(key, value, current);
} else {
currentHive.createValue(key, value, 0x7, data.replace(", ", ","));
}
};

e.g the WinBuilder registry function regimport could be used rather than a Win7PE series of single registry commands.

 

e.g. the UPGRADE command would work.

 

e.g. Win7 SP0 could be used.

 

e.g. Win8 WIMs could be expqanded.

 

e.g. the HOOK mechanism could be used to add plugins to Win7PE.

 

e.g. local projects and plugins could be developed rather than only download them from reboot.pro.

 

e.g. ...

 

Peter

[Brito]

Hi Peter,

Where are the advantages of pointing users to a ghost version? There are none, other than causing confusion. I would gladly trade new features for something that we can call stable.

I understand that you want a new version published. However, there is critical work missing to be done, the new version is not ready nor should a new version turn the older version released a few months ago completely obsolete.

The current win7PE needs the conversion to java-like scripts. I'm unable to do this on my own, would you be willing to work on this? Otherwise investing more effort onto this project on the current state is simply not good. What matters if you write something you're proud about when it just makes the next development step significantly harder?

My friend, when people write AUTO they expect something done without scary messages.

[pscEx]

Seems to be a problem with the MMC component.

There has been a nice problem:

 

The MMC component was simply lost in the download.

I uploaded a corrected version.

 

Peter

[Brito]

Peter, these scare messages are not good.

 

Messages removed on the most recent version of win7PE.

Edited by Nuno Brito, 02 January 2014 - 04:12 PM.
updated topic

[Baylink]

No one's interested in what I have to say around here, but I'm going to say it (again) anyway. :-)

Both of you gentlemen are *creating tools for building other tools*; Nuno two layers deep, and Peter one layer. This imposes upon you some constraints that are not common to those who build apps for users to use directly; there are a different set of expectations upon you.

Since, at the moment, Win7pe2013 seems to be the only mere-mortal usable toolset on top of WinBuilder2013, Nuno, Peter is your customer. Your job is -- without foreclosing on other useful things which might be desirable to other people building tools like his for people like me to use -- to give him what he needs to make his tool usable by people like me.

You have a certain interest in whether his tools make yours look bad, certainly, but there's a fair amount of play in that machinery.

Peter, for your part: your "does the underlying tool provide the facilities I need to give my end-users what they want" concern is valid -- and will become more valid as both tools iterate -- but perhaps there's a better way to phrase your tool's unhappiness about what it finds beneath it?

And both of you need to be *really clear* about which version of your toolset a person is running, both at runtime, and at download time; I cannot stress enough that *the harder you make it for your users to help themselves*, the more they're going to either a) dump that load on you or give up and move on.

Naming both of your current tools the same as the previous tools for which there is scads of documentation of various types on Google -- as much as you continue to want to not hear this and lock me out of threads -- is having a *major* negative impact on you on this point. The tools' interface *to the people who use them* is different -- and for you, Nuno, *hopefully* those people will not always be only Peter -- and that's what needs to drive the naming, as it controls the keywords which people will use to search for self-help.

Edited by Baylink, 02 January 2014 - 04:29 PM.

[Brito]

Naming both of your current tools the same as the previous tools for which there is scads of documentation of various types on Google -- as much as you continue to want to not hear this and lock me out of threads -- is having a *major* negative impact on you on this point. The tools' interface *to the people who use them* is different -- and for you, Nuno, *hopefully* those people will not always be only Peter -- and that's what needs to drive the naming, as it controls the keywords which people will use to search for self-help.

 

We need help in getting things improved, particularly by a native English speaker. You're welcome to join and help improve the user experience (guides, error messages) to make them more meaningful to other users.

[Baylink]

Well, I did try, before Christmas... but the both of you spent hours and hours telling me I was full of it, and I wrote too long, and locking me out of threads.

Peter was probably justified in blocking my PMs; I sort of went off on him, for which I apologize. But the thread-locks seemed entirely like "I don't like these opinions, so I'm not going to listen to them". Since you're trying to be at the root of a tool-ecosystem, if that's really the view you hold, I -- as a potential user, even two layers up -- would be very wary of piling onto that community.

The first batch of help I'd provide is to say "go back and re-read the posting you thought was too long, filtering out any of my unwarranted irritation; there's useful stuff in there". :-)

[pscEx]

Maybe it is caused by my bad English, but I felt in an unfair way agressed by what you wrote, and therefore I closed the topic.

 

Let's give us a new chance!

 

Peter

[Baylink]

It wasn't entirely 'unfairly', Peter; I was pretty cranky that day. :-}

[Baylink]

[ moved to the top of this reply: I do tend to write rather lyrical English; purposefully, though it's not necessarily helpful here. If you think you're hanging on a point of my meaning because of my phrasing, please do ask for clarification: I don't speak Portuguese or German at *all*. :-) ]

So, to capsulize my first recommendation:

1) It is important for the purpose of reducing support load on the people developing the underlying scripting engine (Nuno) and the mid-layer tools which use it (Peter), and those third layer people who will at some point write scripts to include specific apps into the builds created by people like me, up on the fourth layer, that:

a) each of those layers be pretty clearly specified, and probably formally named, even if that's just Layer 1 (builder) through layer 4 (a bootable PE disc), and

b ) each of the components at those layers have a formal proper name, and

c) those names need to be *searchably different* from earlier programs, even though the programs fill the same logical purpose, because they're written in a different language, and work a different way, and therefore create a completely separate support load, and that after that

d) both programs need to be versioned in a generally understandable fashion, so that the various authors can talk to each other and the public about what features and fixes are new in each version (looking up) and can complain intelligently about missing features (looking down); ie: "This version of PE7builder requires a version of Basebuilder >=0.37, because it needs feature XYZQ, which is not provided in earlier versions."

and

2) That the name of the end-produced component not be the same as that of the tool(s) that produced it. "A Win7PE Boot Disc" is not the same as "A Win7PE" or "Win7PE".

I generally recommend, for public releases, a three-component use of the protocol at:

http://en.wikipedia....sed_identifiers

with careful attention to the part about Change Significance.

For pre-release software, which both Nuno and Peter's programs are, right now, in my professional estimation as someone who tried to use them, and then used the older WinBuilder, the first component should be 0. That's not a dig at either of you, but the point of version numbers is *communication*; it's a message from the programmer to the users, and it has culturally imposed meaning, which you should work with, not against.

If either of you hasn't ever read that entire article, it would not be a waste of time to go do so... though, as a user, I do recommend avoiding some of the ... whackier versions of numbering mentioned therein. :-)

So that's my first set of thoughts; hopefully you find them comprehensible, rather than compost.

Edited by Baylink, 02 January 2014 - 04:59 PM.

[Baylink]

As a followup: "this is the easiest boot disc builder ever" is ... either overselling, or at least premature.

It certainly oversold me, which was part of the reason for my later attitude. :-} You might want to reconsider how you're positioning the New Hawtness while it is still warming up.