I need a behaviour provided by:
x = 2;
Block[{x = HoldForm[x]}, x]
x
FrontEnd marks this syntax as not quite correct.
It has not caused me any troubles so far so my question is: could it be a problem?
$\begingroup$
$\endgroup$
2
1 Answer
$\begingroup$
$\endgroup$
6
The FE only looks at the structure of your code for colouring. It doesn't evaluate anything. This means two things: (1) it can only guess that there might be a problem, because by looking at the structure, it doesn't know whether your code really evaluates to something you might not want. (2) You can easily trick the FE by changing the structure into something equivalent which looks different. With this you can easily trick the FE into whatever colouring you prefer. Here are only some examples of exactly the same code:
answered May 24, 2014 at 2:58
-
1
-
-
$\begingroup$ @halirutan So do you think it is only artifact/preventive behaviour and not an idication of any subtle cases that I may face one day? $\endgroup$– Kuba ♦May 24, 2014 at 8:36
-
2$\begingroup$ There is one issue: It seems you cannot remove the
HoldFormwrapper inside theBlock. If you usex=HoldForm[a]instead, everything works as expected. Look at the output of this and then replaceabyx. $\endgroup$– halirutan ♦May 25, 2014 at 1:28 -
$\begingroup$ I stumbled over this when I was trying to mess with the scoping by releasing the hold of
x. I wanted to see what happens because thexinside theHoldFormis the globalx. Obviously they prevented it. What consequences this has for your code, I don't know. $\endgroup$– halirutan ♦May 25, 2014 at 1:30
xwill go out of scope before being used. I must admit I don't understand what this even means forHoldForm.) $\endgroup$