Re: Trouble with Remote Apple Events (eppc) and Keychain
Re: Trouble with Remote Apple Events (eppc) and Keychain
- Subject: Re: Trouble with Remote Apple Events (eppc) and Keychain
- From: Axel Luttgens <email@hidden>
- Date: Tue, 09 Oct 2007 10:33:57 +0200
On 8/10/07 22:13, Maximilian Schirmer wrote:
I did some more testing today and received results from people
I asked to run the sample code. Here is what I came up with:
8 (out of 8) PPC machines (PowerBooks, iMac G5, PowerMac G5 and
a Mac mini G4) with a variety of 10.4 versions (from 10.4.4 to
10.4.10) ran the code without ever crashing. They also stored the
protocol identifier correctly (eppc) instead of reversing it (cppe).
Fine.
This, in conjunction with your trials on Intel boxes, tells us that the
Intel version of some library has a slight bug wrt byte ordering.
Have you already filed a bug report?
It made no difference at all if they were connecting to another PPC
or Intel machine, or to themselves using the loopback interface.
This was to be expected from a logical perspective: in the present case,
the value of concern in the keychain is a strictly local matter, in the
sense that the "eppc://..." string is about how to access a remote box,
not about some reply from that remote box.
But reassuring anyway: logics seems to be preserved... ;-)
2 (out of 2) Intel machines (MacBooks, 10.4.10) corrupted the protocol
identifier and crashed everytime the code was executed. No difference
in connecting to PPC, Intel or themselves either.
So it clearly looks like the problem is limited to Intel machines.
But I don't understand why Axel did not encounter the bug on his
Intel box...
Nor do I... :-(
Should I have the opportunity, I'll check on another Intel box.
Perhaps it is also worth to mention that it doesn't matter if the
event is triggered from a regular AppleScript in ScriptEditor or
through my ObjectiveC-Code (I use NSAppleScript to compile the
script string and create and NSAppleEventDescriptor from the
executed script to check for execution errors, which should boil
down to the same low-level event handling mechanisms).
A little bit at a loss and still looking for advice,
Now, the fact that I experience the reversal of the protocol identifier
too, but without crashing (the Intel version thus just mistakenly reads
what it mistakenly wrote), could also lead to suspect some third party
software.
In my case, I believe to be able to say that I've a rather "clean" box
(no installs involving exotic extensions or "invasive" software).
HTH,
Axel
_______________________________________________
Do not post admin requests to the list. They will be ignored.
AppleScript-Users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
Archives: http://lists.apple.com/archives/applescript-users
This email sent to email@hidden