@require Instructions

in
11 posts, 3 voices

Sep 5, 2009 10:25pm sizzlemctwizzle Script's Author	The @require instructions have now been posted in this guide.

Jan 2, 2010 3:03pm sizzlemctwizzle Script's Author	Update Just implement a completely new version comparison method. Don't worry, it is completely backwards compatible with my older version scheme(which just stripped out periods), but it is now way more flexible. You no longer have to stop at single digits between periods. My updater would evaluate all these as true: 2.10.0 < 21.0.0 1.3 < 1.29 7 > 6.10.2 So what good is all this? Well now your version numbers can actually mean something. Rather than arbitrarily increasing the number, you can use your version number to symbolically denote how much is changed in an update. The first number could mean major version(like you basically rewrote the entire script). The second could be minor version(just general improvements), and the third could denote a tweak.

Jan 3, 2010 1:16am JoeSimmons Scriptwright	sizzlemctwizzle wrote: My updater would evaluate all these as true: sizzlemctwizzle wrote: 1.3 < 1.29 confused Isn't it more?

Jan 3, 2010 2:23am sizzlemctwizzle Script's Author	JoeSimmons wrote: confused Isn't it more? That's what I thought for awhile, but your thinking that they're floats when they're not. The first number represents the major version(one in this case). The second number is a minor version number and remember that ~~3 > 29~~ 3 < 29. If we weren't using the period, which just happens to be the decimal point, you wouldn't be confused. Consider which of these is larger: 1.34.5, 1.5.10 My updater evaluates: 1.34.5 > 1.5.10

Mar 19, 2010 12:44am Marti Scriptwright	sizzlemctwizzle wrote: The second number is a minor version number and remember that 3 > 29. This is only true if you are padding zeroes to the right on the 3. (e.g. `30 > 29` ) sizzlemctwizzle wrote: My updater evaluates: 1.34.5 > 1.5.10 And if you apply the same rule of padding zeroes to the right this is not true. (e.g. `34 > 50` evaluates to false)

Mar 19, 2010 5:07am sizzlemctwizzle Script's Author	Marti wrote: This is only true if you are padding zeroes to the right on the 3 No you can't pad zeros to the right(but you can pad to the left).

Mar 19, 2010 6:45am Marti Scriptwright	sizzlemctwizzle wrote: No you can't pad zeros to the right(but you can pad to the left). Sure you can ;) Point being is that symmetry with this portion of the updater isn't congruent... even if it works (which I haven't confirmed in depth yet). Padding is more of a logical thing then a physical thing... it's up to the interpretation of the developer. This is why I currently choose not to ever rely upon @version and I'm grateful you have an override. :) In my book `1.3.0 < 1.29.0 == true`. It's more important for uso - installWith because a user might run into this case (especially on my scripts as I treat each item as a whole, individual number)

Mar 19, 2010 7:53am sizzlemctwizzle Script's Author	Marti wrote: `1.3.0 < 1.29.0 == true` That was a typo in my post(3 is less than 29). My updater does evaluate the quoted statement as true. My updater doesn't pad zeroes to the right. The symmetry in the way my updater parses version numbers is congruent. Sorry for the confusion.

Mar 19, 2010 8:29am Marti Scriptwright	sizzlemctwizzle wrote: typo ahhh done a few of those myself... thanks for the clarification. :) Some users don't use major.minor.build too so I'm grateful that you have an override... useful with uso - installWith.

Mar 19, 2010 8:32am sizzlemctwizzle Script's Author	Marti wrote: Some users don't use major.minor.build too so I'm grateful that you have an override... I also fallback to @uso:version when @version isn't present. Using the @uso:verison for installWith makes more sense because its targeted at users, whereas a scriptwright might want to keep control of the version number.

Mar 19, 2010 8:36am Marti Scriptwright	sizzlemctwizzle wrote: I also fallback to @uso:version when @version isn't present. Yup... that's what I pm'd and canceled you about earlier because I didn't know that... but I figured it out eventually.