Changes to @REPEAT

[Joe Caverly]

As of and prior to TCC 33.00.12, the following worked;

TCC  32.10.21 x64   Windows 10 [Version 10.0.19045.5011]
Copyright 2023 JP Software Inc.  All Rights Reserved

C:\...\TCMD32>echo %@repeat[#,0]
ECHO is OFF

Now, with TCC 33.00.14, it returns;

C:\...\TCMD33>echo %@repeat[#,0]
TCC: (Sys) The parameter is incorrect.
 "%@repeat[#,0]"

This has affected my PROMPT that uses @REPEAT;

prompt `$e[9999E$w$g%@repeat[#,0%localcount]%@repeat[@,0%_shell]`

From tcmdupdate.aiu;

BugFix1 = 33.0.14 @REPEAT - Fixed RTL problem when entering invalid arguments (count < 0)

Ref: How to? - SETLOCAL Level

Joe

 

[vefatica]

Just a curiosity ... I issued echo %@repeat[x,-1000000000000] and, after not too long a wait, got 727,379,968 x's.

Any ideas where 727,379,968 came from?

 

[Charles Dye]

Just a curiosity ... I issued echo %@repeat[x,-1000000000000] and, after not too long a wait, got 727,379,968 x's.

Any ideas where 727,379,968 came from?

C:\>echo %@eval[-1000000000000 =x]
0xFFFFFF172B5AF000

C:\>echo %@eval[-1000000000000 & 0xffffffff =x]
0x2B5AF000

C:\>echo %@eval[-1000000000000 & 0xffffffff]
727379968

C:\>

The repeat count is a 32-bit integer.

 

[Charles Dye]

A 32-bit unsigned integer; negative values are meaningless in this context. So everyone please stop flippin' doing that.

 

[rconn]

TCC v33.0.15 (already uploaded) allows the (useless) 0 argument for the count.

 

[rconn]

This has affected my PROMPT that uses @REPEAT;

prompt `$e[9999E$w$g%@repeat[#,0%localcount]%@repeat[@,0%_shell]`

%_shell has been mostly meaningless for the last 20 years or so ...

 

[vefatica]

I'd expect

v:\> echo %@repeat[x,0]
ECHO is OFF

That's what I asked for. :-)

 

[Claire_CJS]

A 32-bit unsigned integer; negative values are meaningless in this context. So everyone please stop flippin' doing that.

Easier said than done when some values are the results of calculations, which usually take a couple tries. Thus, the not-dying-gracefully bug cost me 30+ minutes. Sad that the fix messed somebody else up but yea stop doing that lol. At least my negative values were column differentials ;) i just was too lazy to absolute-value the value.