Quote:

Abi V1 (17/02/14) seems to have the same issue.
Abi V0, also from 17/02/14 (the one with unarc instead of zunearc) works correctly -> Shows an icon and when double clicking, opens the lha file in Unarc.

Magorium. Absolutely you can test it through live mode. Just boot with the iso to test!

Sorry not really a reply, just an additional 'issue'.

BSzili: i did not test this myself, but someone on irc described that whenever on aros a not recognized disk/filesystem is inserted that magellan reports it as being displayed as f.e. dh0:sodn (or similar, he could not remember exactly).

This seems a harmless endianness problem, but thought i mention it.

Do not know if this issue is known, if so just ingnore it, if not then take a bite....


This happens to me (all the time I do this - until now at least)

Freshly started AROS with Magellan, the vertical named buttonbar which I place in the right side of the screen, below the screen dragbar.

I select RMB on Edit... sometimes the edit button stays down blinking, second time I chose the same button with RMB the whole (vertical) button bank - dissapear... Sometimes it dissapears when I RMB->Up the first time on Edit (showing New File - then *pooof* gone)

ArosX - WIP 5

paolone wrote:
Quote:
I noticed also others. try adding an icon with dopus4 on 'recognized' filetypes. it creates a default icon with tool multiview. mp3's, lha's etc. all behave the same :/

Quote:
None of the default icons seem to have a (proper) tool attached. I have some ignorance towards this subject, the default icon _should_ contain a tool, no ?
red: my fault, i simply overlooked it as i expected it be on the bottom of the file (and not in between two pictures).


Quote:
I can do that for you (not right now), but will take some time (i have to re-install). Could i test this in live mode ? (as that would certainly speed up things for me).

Quote:
afaik you are correct, however there seems to be regression (as stated above).

Quote:
This is as far is i could get atm

red: this 'faulty' behaviour is consistent with that from regular 1.5.2. There is no regression from my pov, so not a priority. When did you introduce 'working' behaviour ?

fwiw: i can confirm the 'broken' arexx behaviour. i can't seem to get a grip as of why it fails.

The arexx tests (in test directroy) simply works, port can be written to with a simple arexx script so it seems to execute properly. However, trying 'simple' things like SAY command or display the ports (SHOW) seems to fail (at least it doesn't return anything visible).

red: bugreport added to sheet.

Hi azvareth,
Just tried the rexx issue on WIP5 and the same is happening for me - no say output. This is with Wanderer as the desktop.

Nigel.

Edit: Just tried this with some other rexx programs; this is not just the 'say' command, it seems a rexx program is not starting up at all. I've checked with Scout, and RexxMast is running. However, the 'Regina Helper' task, which only appears on regular Icaros whilst the rexx prgram is running, but in WIP5 I saw serveal occurances of it still hanging round in memory - one for each failed run of rx. So whilst after running 'rx' control immediately returns to the shell prompt, I'd guess the Regina Helper task is locking up somehow so the rexx script itself does nothing.

I have added this misbehavior of LHA and ZIP files in icaros' bug tracking spreadsheet. Unluckily in the latest weeks I've spent all my time resources to make Magellan work as a real wanderer replacement, but I couldn't care about Wanderer at the same time. So I can't say exactly when LHA and ZIP files stopped working as they should.

What I can say, is that the def_Archive.info file in envarc:SYS has not changed a bit between regular Icaros and WIP 5. I have tried replacing it and the files in devs/datatype both with regular 1.5.2 and with latest build from Deadwood, but without success. In any case, LHA and ZIP stopped working.

If someone has kept all WIP ISOs (I don't have older ones anymore, sorry), maybe he can be helpful finding when the issue arose.

The only thing I changed for LHA and ZIP is the way MAGELLAN handles them, creating apposite filetypes. But AFAIK Magellan filetypes have nothing in common with Wanderer's datatypes, hence my serenity about the fact that those changes can't be anyhow related to this regression.

Any idea?

Hmm, that means it wasn't a Magellan issue after all.

I didn't use magellan (this time), just wanderer.