Error when using command as parameter for an other command

I want to use 2 seperate commands to first execute a calculation on a number and then displaying the number with plus or minus using following code.

When trying to run it I get various errors. I think that propably FPeval is causing problems.

FPeval

documentclassscrbook usepackagefp usepackageifthen newcommandMyHalf[1]FPevalresulttrunc(#1 / 2, 0)result newcommandMyDisplaySign[1]ifthenelse#1<0$#1$$+#1$ begindocument MyHalf12 MyDisplaySignMyHalf12 enddocument

FPeval is not expandable, it assigns the result to result, so you need to do the evaluation first and then pass result to the expansion context in the numeric test.

– David Carlisle
Sep 8 '18 at 18:38

FPeval

result

2 Answers
2

The problem is that for the ifthenelse test compare #1<0, #1 has to be a number. But if you pass a macro that contains a number (or, in some way, expands to a number), that macro has to be expandable, so that ifthenelse can performing the test. The problem, as David said in the comment, is that FPeval is not expandable because it assigns the result to result.

ifthenelse

#1<0

#1

ifthenelse

FPeval

result

But fp_eval:n is expandable :)

fp_eval:n

documentclassscrbook usepackagexparse ExplSyntaxOn NewExpandableDocumentCommandMyHalf m fp_eval:n trunc ( #1 / 2, 0 ) NewExpandableDocumentCommandMyDisplaySign m int_compare:nTF #1 <= c_zero_int $ #1 $ $ + #1 $ ExplSyntaxOff begindocument MyHalf12 % Prints 6 MyDisplaySignMyHalf12 % Prints +6 MyDisplaySignMyHalf-12 % Prints -6 MyDisplaySignMyHalf0 % Doesn't print "+0" because 0 isn't larger than c_zero_int, so the conditional evaluates to true MyDisplaySignMyHalf-0 % Prints -0 because of the leading sign enddocument

The only issue is with 0. With the code above you don't get the leading + sign because 0=0 is true, so the first (for negative numbers) is taken. If you want the leading + sign then you have to print the sign if the value is equal to zero too, but then with negative zero you get +-0, because -0=+0 and the sign is added too:

0

+

0=0

+

+-0

-0=+0

documentclassscrbook usepackagexparse ExplSyntaxOn NewExpandableDocumentCommandMyHalf m fp_eval:n trunc ( #1 / 2, 0 ) NewExpandableDocumentCommandMyDisplaySign m int_compare:nTF #1 >= c_zero_int $ + #1 $ $ #1 $ ExplSyntaxOff begindocument MyHalf12 % Prints 6 MyDisplaySignMyHalf12 % Prints +6 MyDisplaySignMyHalf-12 % Prints -6 MyDisplaySignMyHalf0 % Prints +0 MyDisplaySignMyHalf-0 % Prints +-0 enddocument

The isn't, as far as I know, a way to get -0 and +0 correct without doing extra tests for this specific scenario.

-0

+0

ifthenelse does expand the argument in the test, the problem is rather that FPeval is not expandable. (but using fp_eval is better, I agree:-)

– David Carlisle
Sep 8 '18 at 18:40

ifthenelse

FPeval

@DavidCarlisle Ooh. Thanks, I'll fix it :)

– Phelype Oleinik
Sep 8 '18 at 18:41

This would print +0 With int_compare:nTF #1 <= 0 you'd solve the issue. Or reverse the test: int_compare:nNnTF #1 > 0 $+#1$ $#1$

– egreg
Sep 8 '18 at 21:25

+0

int_compare:nTF #1 <= 0

int_compare:nNnTF #1 > 0 $+#1$ $#1$

@egreg Well pointed. Thanks :)

– Phelype Oleinik
Sep 8 '18 at 21:49

The changed version still prints +0.

– egreg
Sep 9 '18 at 16:34

documentclassscrbook usepackagefp usepackageifthen newcommandMyHalf[1]FPevalresulttrunc(#1 / 2, 0) newcommandMyDisplaySign[1]ifthenelse#1<0$#1$$+#1$ begindocument MyHalf12% % Now the result is stored in the macro result. % Let's see the result: result % Now let's see the result with preceding algebraic sign: MyDisplaySignresult enddocument

Thanks for contributing an answer to TeX - LaTeX Stack Exchange!

But avoid …

To learn more, see our tips on writing great answers.

Required, but never shown

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

搜尋此網誌

Dfyjkt