Bifroest Mr. Tines MacCTC CTCjava Manual Pages
Bifroest Mr. Tines CTC Home CTClib MacCTC CTCjava Manual


Release Notes: MacCTC 2.1

MacCTC2.1 is built with CTClib2.1, and hence incorporates the bug-fixes and improvements of this library release. See the CTClib2.1 release notes for details.

Testing: I regret that due to the demise of my 68k lap-top, I no longer have a 68k machine for native mode testing this variant. Accordingly the 68k variant is seriously undertested.

Changes

Key Management

The most significant change is the addition of some new key management windows that allow the examination of the details of keys, and a number of manipulations. This includes the addition of a number of extra windows. All the changes are accessed via the Keys menu. Formerly this menu had several sub-menus each for a different operation, Extract, Enable, Disable, Discard, etc. At 2.1, this menu is simplified and most of the functionality is moved into the new key editing window. This is accessed (for any key) from the Edit Key... window. Only selecting Signing Key and Extract Key remain as separate operations in the Keys menu. For details see the relevant section of the MacCTC manual.

Text Window handling

Text (edit) windows are not handled slightly differently, especially with respect to saving them. Previously the handling was awkward. It was not possible to "save" a window with no associated file, you had to use "Save As..." to store it. The new behaviour is more intuitive and standard.

Window Resizing

A number of windows have been resized or repositioned. This is to reduce overlap and give a more comprehensible user interface, especially on small screens.

Bug Fixes

The following problems were in 2.0 and are fixed in 2.1.

Wrong Clock

MacCTC used to generate time-stamps in localtime not UTC.

UUencode Armouring did not work

Selecting UUencode format ASCII-armour produces not output. (This is a simple coding error handling the menu. The UUencode item is return ARM_MIME, not ARM_UUENCODE and as MIME isn't implemented, there is no output.)
[This did not apply to the 68k version.]

Changing key-ring was not immediate

A change of key-ring does not seem to be immediately effective. You needed to restart the program.

Line-wrapping breaks lines

When MacCTC performed its automatic line wrap before encryption it could split words in long paragraphs.

Ignoring unsaved changes

If a text window had a disc file associated with it, and had unsaved changes in the window, then "encrypt window" operations could take the data from the disc file thereby ignoring the unsaved changes.

This is in addition to in CTClib2.1. The CTClib release notes includes a technical description of these. However for non-programmer users for MacCTC the user visible effects are:-

3DES conventional encryption

This is now interoperable with PGP5.0 and other OpenPGP implementations. Previously it was not.
N.B. This means that CTC2.0 is not inter-operable with CTC2.1 on this class of encryption only. Anyone who has been using CTC2.0 for 3-DES conventional encryption must decrypt with CTC2.0 and re-encrypt with CTC2.1

CTC2.1 handles Key-Recovery Keys

CTC will now read key-rings and files containing Key-Recovery Keys. It discards any such keys. CTC is not big-brother friendly. (Previously CTC suffered a detected buffer overflow and refused to process the Key-Recovery key or anything following it in the file.)

Eudora Interface Fixes:

There have been a number of improvement in the robustness of the Eudora interface.

All messages parsed
Earlier versions were inclined to fail to decrypt some of the newly-arrived messages in the In box, when there was more than one of them. 2.1 decrypts them all.
No Queueing without recipients
At 2.0, if a message is sent to Eudora with recipients (not possible for normal encryption; but is possible for password encryption or plain-signed messages), then it attempts to queue the message. This causes an Eudora error.
No launching if already running.
MacCTC 2.0 was inclined to attempt to launch Eudora even if it was already running. This had the effect with some variants of Eudora (generally with -Pro but not with -Lite) to generate a new out-going message with the Eudora Settings file as an attachment, but otherwise filled in. This was especially confusing as the encrypted message itself was typically queued and hidden in the Out box. It gave the impression that this spurious message had been created instead, rather than as well.
2.1 checks for the running application and only sends the launch command if it is necessary.
Reduced AppleEvent Time-out
MacCTC 2.0 used the AppleEvent Manager default time-out. If the command sent by Eudora generated an error, this seems to lock the whole machine for the duration of the time-out (on MacOS 7.1&7.5 at least). (I don't understand that is happening here. It seems to intrinsic to sending AppleEvents synchronously.) This would be acceptable with a reasonable time-out. However on my machine the time-out is two minutes. (I don't normally wait 2 minutes before doing something drastic so I only recently discovered this was a time-out.) I suspect handling this elegantly requires asynchronous handling of AppleEvents, which may introduce all sorts of other problems. For the moment I have reduced the time-out to 5 seconds.


webmaster@bifroest.demon.co.uk