Has anybody managed to get ZapMJE to compile with the DDE?

The distribution rights are simple and complicated at the same time.

Basically it amounts to “you can, so long as it’s unmodified and you don’t charge for it”; however this was written in a time when there were actual maintainers who kept things up to date and, for good reason, didn’t want a dozen “pet builds” doing the rounds – all of which will be calling themselves Zap without actually being Zap.

The problem now is that the maintainers have moved on and nobody is looking after Zap. This leaves us with a build of Zap and several fixes. Since nobody is actively maintaining Zap, my compilation won’t be interfering with the official version (which doesn’t work correctly on new machines anyway).
Some clarity on the licence would be helpful (please, anything but GPL) but I believe this question was raised several years ago when looking for software to put on the SD image…and here we are asking the same question again. ;-)
Sticking to the absolute letter of the licence means distributing a broken application. I feel that what I’m doing here is more in the spirit of things, and is helping keep Zap alive. If anybody involved with Zap disagrees, y’all know my email address, right?

[I’m not interested in mailing lists and code repos and such, I’m only interesting in keeping my preferred editor working as ARM’s subtle changes bite us on the ass, and to make installation less chaotic for others; so if anybody else would rather run the show, let me know and I’ll back off]

8~(

Machine still working. !Zap appears in Apps on my hard drive, where I put it – the rest’s in the specified places in !Boot on my SD card: !ZapUser in Choices, zap_ultimate.zip doesn’t appear to have !!DeepKeys or !Country but I’ve got some old ones and put them in Predesk (hope they’re okay versions?)

Double click on !Zap and the machine appears to hang. Can still move the pointer around, but nothing else – for about five minutes, after which I get a report, “Error: Unexpected disconnection” and the machine is back, but still no !Zap on the icon bar.

Am I running Hi Vector? I’ve done nothing about it either way since installing RC15. Or what else might I have done wrong?

doesn’t appear to have !!DeepKeys or !Country

Ah. Thanks, I’ll drop those in later.

Double click on !Zap and the machine appears to hang.

Right. That’s not so good. I’ll see if I have an SD card spare to drop a plain RC15 on it. See if I can figure out what’s going wrong here.

Error: Unexpected disconnection

??? Well, okay, there’s something to look for, but, um… Not a message is have expected to see from Zap…!

Thanks for trying this.

My base version of Zap was the one on !Store. I wonder if it was missing stuff (certainly DeepKeys etc)?

My base version of Zap was the one on !Store. I wonder if it was missing stuff (certainly DeepKeys etc)?

As a non-user of Zap (I have tried it but various aspects put me off¹) it seems like it’s needed a lot of TLC for a long time.

My base version of Zap was the one on !Store. I wonder if it was missing stuff (certainly DeepKeys etc)?

and there’s a twist, the numerous not-quite-the-same versions. Even Zap users comment on it.

¹ The chief negative aspect is the fragmented install that has been around for a long time. I don’t think regular Zap users have noticed over the years how bad this is as they have a coherent base install and don’t need to search out and add the “extras” that are in fact “essentials”. So if someone could create a simple install/remove² it might get more use.
²It’s, erm, “entertaining” digging out the bits left behind when Zap is removed.

Ah – an oddity and a question: there are two copies of !ZapUser in zap_ultimate.zip: one in the top level, and one in !Boot.Choices. I assumed they were the same, but they’re not. I’ve actually installed the one from the top level without realizing they were different. On inspection, I see that the one I’ve installed has at least some files newer (& larger), but it’s also missing some sprite files. !Zap and !ZapUser are clearly finding their sprites somewhere though.

Not dug any deeper yet. Shall I try deleting my !ZapUser and putting the other one in?

Right tried zap on a Pi3 and on Omap5432 they are high vector versions 5.23 (4 aug 17) for the pi and ( 22 Jul 17) for the Omap same version from Ultimate zap version 1.48(17 jun 2017) no errors yet so nothing to report, tried basic files,text,data,code no problems.
Shows 11 non 32 modules in Zap folder using !ScanMod will keep testing.

Thanks. I’ve picked up a copy of ScanMod, so will give things a test later on.

Clive – do you still have Zap installed? Just out of interest, if you run it again, does the same thing happen?

Ah – just saw your other post. Not sure why there are two ZapUsers – that’s how Store provided it. What I did was install the base version, then copy the one within Zap over the top.

What I think I might do is throw away the Store version and see if I can build up an installation with the last proper version of Zap…

What I think I might do is throw away the Store version and see if I can build up an installation with the last proper version of Zap…

Good move. As I said “…the numerous not-quite-the-same versions. Even Zap users comment on it.”

People have papered over perceived cracks and band-aids until it works only by accident. It will probably take a bit of effort to trim down to a working base install as opposed to someones modified with sticking plaster version.

Certainly – the installation instructions are complex enough without the problem of patching together all the modifications in the last half decade or so…
…assuming one knows there are updates and where to find them.

Certainly – the installation instructions are complex enough without the problem of patching together all the modifications in the last half decade or so…

I think the install is complex because of twiddles and tweaks and sticking plaster that goes back to ye olde Acorn times. If someone were to volunteer to strip things back and rebuild I can’t see the original builders objecting to the new structure if it is the same in UI terms.

…assuming one knows there are updates and where to find them.

With a true rebuild the updates and additions and workarounds are not needed as they are part of the whole.

Aaarrrghhh!

Something has f***ed up my set-up; NetSurf now reports “No Domain Name Servers are configured, so only local browsing will be possible. Use Configure to set your name server(s).” Sadly, I don’t know how to do that…

But yes, the same thing happens to !Zap. But… if I hit Escape early in the process, I get control back with a beep, and sometimes with a window saying “Escape,” or the briefest flash of a window that looks as though it might be the same one but it’s gone before I can see it, or not even that: just a beep and my machine’s back. Other apps are working normally, apart from NetSurf who’s forgotten its configuration 8~(

However, I’m getting slightly suspicious of a hard disc problem: some directories are getting very slow to open, with the machine appearing to hang until they open – a couple of times for so long I gave up and rebooted, after which the offending directory opened no problem. The first couple of times this was the directory with !Zap in it, but I’ve now moved that into a subdirectory, and the parent was still dodgy to open. But NetSurf doesn’t live on the hard drive, and nor does its configuration.

Steve:

With a true rebuild the updates and additions and workarounds are not needed as they are part of the whole.

Yes… that’s what I’m heading towards.

Clive:

Something has f***ed up my set-up; NetSurf now reports “No Domain Name Servers are configured, so only local browsing will be possible.

Try *RMReInit Resolver. Does it work after that? I found a while back that some sort of interaction with NetTime destroys Resolver. It will work if Resolver is kicked in the ass once the Wimp has started (you can’t RMReInit before then, it’ll fail). How I wish we had source for Resolver so could fix its many quirks. :-/

some directories are getting very slow to open, with the machine appearing to hang until they open

That’s what my IDE drive did way long ago until one day it just died without warning. So – obvious question – do you have backups? Are they recent?

I can’t help thinking that the installation method is still a hangover from the days of floppy discs, or maybe the very early days of hard discs when you were unlikely to have more than 100 MB available. My !Zap directory contains just over 4 MB, which is nothing these days. It must be far easier to just install the whole lot, i.e. a full set of modules.

There are, I believe, still good reasons for !ZapUser to go in Choices, and for !ZapFonts to go in Resources.

So we can, in principle, get a full installation of Zap in three drags from the archive. I no longer see any reason to complicate it beyond that.

So we can, in principle, get a full installation of Zap in three drags from the archive. I no longer see any reason to complicate it beyond that.

This. This is how it should be.

Unfortunately there are resources (DeepKeys etc) that complicate things. It could be done in two drags if we package up an !Boot with our additions in the correct places. The user can just drag our !Boot over their own to me in the bits. Easy to install (two drags) but not so easy when it comes to uninstall.

Thoughts?

Excellent work – hopefully we’ll have a simple download and install process soon. That way new users can be directed to something they will understand and find easy to use.

Panic over. The whole trouble with !Zap was probably my hard drive, unless there’s something about the way it’s set up that doesn’t like being on a different filing system from !Boot. !Zap is now working, installed in Apps on the SD card.

And yes, I’ve been in this sort of business a long time…I have recent backups. Nine year old 160GB Lacie drive is now officially Pushing Up Daisies.

Now all I have to do is figure out how to reconfigure NetSurf. *RMREINIT Resolver didn’t fix it 8~(

Clive – is it NetSurf or the Resolver? Can you ping Google?

No idea. Now you know how very patchy my IT knowledge is – I’ve no idea how to ping anyone on any of the systems in this house!

How do I (try to) ping Google?

from the command line, I tried *ping http://www.google.com/ – which reports “unknown host” – is that what I should have tried?

now ping 8.8.8.8

Just the domain part is needed to ping, e.g. Not http://
So, F12
*ping www.google.com
And Escape to stop

which reports “unknown host”

Okay, it’s not NetSurf, it’s the Resolver.

First, check that you can ping your broadband router. It will probably have an IP address like 192.168.0.1 or 192.168.1.1.

*Ping 192.168.1.1
PING 192.168.1.1 (192.168.1.1): 56 data bytes
64 bytes from 192.168.1.1: icmp_seq=0 ttl=64 time=6.481 ms
64 bytes from 192.168.1.1: icmp_seq=1 ttl=64 time=4.624 ms
64 bytes from 192.168.1.1: icmp_seq=2 ttl=64 time=4.262 ms
64 bytes from 192.168.1.1: icmp_seq=3 ttl=64 time=4.019 ms

--- 192.168.1.1 ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 4.019/4.846/6.481 ms

If you have packet loss or no connection at all, then you’ll need to check your networking set up. Let us know if this is the case.

Hang on – are you static IP or DHCP? Maybe the DHCP isn’t working right? I don’t know, because my Pi boots in ~12 seconds, the Vonets in a little more, and the Livebox in ~3-4 minutes, there’s no point in me bothering with DHCP. My Pi is 192.168.1.10 and that’s that.

…

I will continue as if it’s Resolver acting up:

At a command line, try *Show *Resol*. Does anything come up? You’ll be looking for something like:

Inet$ResolverDelay : 5
Inet$ResolverRetries : 3
Inet$Resolvers : 192.168.1.1 192.168.1.1

The “resolvers” should be the same IP address as your broadband router.

If you have that, try *Route -n get default and see if the “gateway” value is the IP address of your broadband router, something like this:

   route to: default
destination: default
       mask: default
    gateway: 192.168.1.1
  interface: ej0
      flags: <UP,GATEWAY,DONE,STATIC,PRCLONING>
 recvpipe  sendpipe  ssthresh  rtt,msec    rttvar  hopcount      mtu     expire
   16384     16384         0         0         0         0      1500         0 

It is my understanding that NetSurf uses its own resolver, and not the OS version.

ping 8.8.8.8 reports “no route to host”

ping www.google.com reports “unknown host”

I’m DCHP – and it’s working fine on this Mac that’s plugged into the same repeater, and through which the Pi was talking to you yesterday evening.