Hi peppiv,

Like everyone else, at some time or other, I have been bitten by this. I have noticed lately around the monastery some people using the idiom

if ( 1 == $row ){ ... }
[download]

the reason being that if ( 1 = $row ){ ... } would yield an error. ( Can't modify constant item in scalar assignment )

My first impression was that the construct was ugly, and counter-intuitive, but I feel my opinion changing. I haven't included it in any code, yet, but who knows what the future will bring? :-)

Does anyone have any thoughts on whether the use of this construct should be encouraged?

thinker

I'm a big fan of the if ( 2 == $whatever ) {...} construct myself, and try to use it whenever possible to avoid the problem you describe. After a while you get used to it and it looks normal.

Maybe I haven't been reading closely enough, but I've never seen that before. I must say that I find that counter-intuitive, as you did initially. I don't see the need for additional code that exists only to validate earlier code where you want to guard against making a coding mistake - does that make sense? Just bite the bullet and learn that "==" is to compare numbers. peppiv did the right thing posting his question here - someone will see the error, even if you can't ;-)

It's not additional code. It's using

if (1 == $x) { ... }
[download]

instead of

if ($x == 1) { ... }
[download]

because

if (1 = $x) { ... }
[download]

is a syntax error, but

if ($x = 1) { ... }
[download]

is not.

It's easy enough to learn that "==" is to compare numbers, "eq" is to compare strings, and "=" is for assignment in Perl, but peppiv is using Perl and SQL at the same time, and in SQL "=" is for all exact comparison and ":=" is for assignment. I (fortunately) tend to confuse the two to the detriment of my SQL, which almost always causes a syntax error instead of a logical error. Syntax errors are easier to fix than logical ones - I like "1 == $x" style for that reason (although I never remember to do it).