Mudlet 2.0 final

[Manni]

This issue started in test7 for me. I was in test4 before I ran into this issue, but there seems to be an issue with display(). I'm not sure how to explain this so I'll just paste my error here.

<[string "-------------------------------------------..."]:14: attempt to call global 'printable' (a nil value)>


This was the error in every attempted display be it a string, number or variable value.

[Vadi]

The issue is with a 3rd party package that you've installed - Coding pack. Remove or upgrade it, restart Mudlet, and it'll be fine.

[Phoenix]

This issue started in test7 for me. I was in test4 before I ran into this issue, but there seems to be an issue with display(). I'm not sure how to explain this so I'll just paste my error here.

<[string "-------------------------------------------..."]:14: attempt to call global 'printable' (a nil value)>


This was the error in every attempted display be it a string, number or variable value.

The quickest fix for that particular one is to delete the display() script out of the coding pack - that's all that's needed, with a restart of mudlet, you'd be fine. Also, when you've got a script downloaded that has a function rewritten? It's prolly in that script, not the mudlet build.

[Phoenix]

Mac bug - unsure when this started. In miniconsoles, at certain font-sizes (In fact, any font size other than 10), this bug is heavily apparent. The 'spacing' between letters is incorrect, unless highlighted... Thus, when highlighting, you will witness the font 'contracting' or 'expanding', depending on the particular font size - I've shot a quick screen-cap video of a miniconsole at size, where you will witness the font contract as I highlight it. I've also included several screen shots of the bug. This is a Mac Unique bug, to the best of my knowledge... and I'm also pretty sure that it started back around the rc8, though I can't pinpoint that exactly.

Mac 10.6, running 2.0 Release

https://dl.dropbox.com/u/22569276/Mudlet%20Bug.mov - movie of bug. (download it, don't play it in the browser. And ignore the initial green-screen!)

Willing to give more details or info, shoot me a message or email.

Edit: More info -

You may need to 'switch tabs' if you're using the tabbed chat to reproduce this, in order to see it - it doesnt' seem to happen if the window is just 'normally' up. I'm unsure what code YATCO uses to switch tabs, I can look that up... I may end up making a profile that will reproduce this issue on mac for whomever is trying.

As well, if you CAN reproduce this error, highlighting down past the end of the miniconsole will 'fix' the entire issue for that miniconsole, until you switch off that tab and back onto it.

-Phoenix

[Phoenix]

Last post - This started in test 6 on Mac... When something is disabled (IE, the checkbox is 'open'), it appears to be a 'low quality' image that I'm seeing, as opposed to just an open checkbox. I'm unsure who all this affects, or if there's something I've gone and installed wrong that's causing issues. A screenshot of various 'types of boxes' is below - note, that if it's 'disabled', it has flak in the box, and if it's disabled within a disabled class, there's flak (in a different way) in the box... but under no other circumstances, does the flak appear.

Disabled Bugs.png (36.69 KiB) Viewed 8970 times

[Phoenix]

The new display has a bug, and a 'derp' feature. Bug first.

When I use 'display()', it adds a 'nil' after the correct output. This is happening on a new blank profile. This happens regardless of whether or not I am using it in scripts, alias, or from a lua alias on the command line. This happens when displaying strings, number, or nil... but not when displaying a table.

To reproduce, simply display(42), or display("The Ultimate Answer") or some such. I've heard of this happening on Linux, and I'm having this issue on Mac OSX 10.6.


'Derp' feature is that display() does not differentiate in it's output between a number and a string. For instance, display(42) and display("42") will show the same output, although the difference between a string and a number can mean the difference between the right answer and the wrong answer. This needs to revert to the old output of a string inside singlequotes, and a numeral not in quotes, or some equally easy to understand way to differentiate between the two.

[demonnic]

I have found the issue in prettywrite() which was causing display to show the extra line with nil. I also adjusted prettywrite() so that it would return a string in doublequotes, as this is in keeping with how it displays strings within tables. I have submitted a pull request to Beliaar's github repository as of just a few minutes ago.

[Vadi]

Thanks, it'll be in the next update.

[Phoenix]

Another issue with display() is that it doesn't display empty tables on a single line... this isn't a major issue, however it'd be nice if an empty table displayed as {} instead of {
}

[Vadi]

I don't think it's an issue, it just returned {\n} (not \n{\n})- but I've improved it to return just {} on empty tables.