Length of float using printf

vinaynp has asked for the wisdom of the Perl Monks concerning the following question:

For instance, I can control the length for a string in the example below and get a print as "Vina"

my $string = "Vinay";
printf('%-4.4s', $string);
print "\n";
[download]

The below set of statements prints 40000. How can I control the length of the float using printf to print 4000 even though the value of num is 40000?

my $num = 40000;
printf('%4.0f', $num);
print "\n";
[download]

Comment on Length of float using printf Select or Download Code

Replies are listed 'Best First'.
Re: Length of float using printf by bart (Canon) on Jun 02, 2009 at 18:35 UTC
You seem to want the equivalent of BASIC's Left$. In Perl, we have substr, though we can also get by with pack/unpack, with template `"A4"`, that behave differently on edge cases: `$out = pack "A4", $string;` [download] will always produce a string of 4 characters, padded with spaces if shorter `$out = unpack "A4", $string;` [download] will break off a string after 4 characters, and remove trailing spaces. (Bytes, actually...) `$out = substr $string, 0, 4;` [download] The easiest approach, cutting off a substring of at most 4 characters (yes, characters).	[reply] [d/l] [select]
Re^2: Length of float using printf by ikegami (Patriarch) on Jun 02, 2009 at 18:43 UTC
"A" works with characters since 5.10. (It still works with bytes, being a subset of characters.) `>perl588\bin\perl -MTest::More=no_plan -e"$s=join '', map chr, 0x2660. +.0x2666; is(unpack('A4',$s), substr($s,0,4))" not ok 1 [...] 1..1 # Looks like you failed 1 test of 1. >perl589\bin\perl -MTest::More=no_plan -e"$s=join '', map chr, 0x2660. +.0x2666; is(unpack('A4',$s), substr($s,0,4))" not ok 1 [...] 1..1 # Looks like you failed 1 test of 1. >perl5100\bin\perl -MTest::More=no_plan -e"$s=join '', map chr, 0x2660 +..0x2666; is(unpack('A4',$s), substr($s,0,4))" ok 1 1..1` [download]	[reply] [d/l]
Re^3: Length of float using printf by vinaynp (Initiate) on Jun 02, 2009 at 19:10 UTC
So, I have decided to implement the substr and change the way my application uses it. Thanks again for all the inputs.	[reply]
Re^2: Length of float using printf by vinaynp (Initiate) on Jun 02, 2009 at 18:42 UTC
I want to use printf only as my application uses printf and stores the format type in the object. The reason I want to represent 40000 as 4000 as it is incorrect if somebody is passing 40000. My object can't figure that out. It is for error handling.	[reply]
Re^3: Length of float using printf by ikegami (Patriarch) on Jun 02, 2009 at 18:47 UTC
`printf` doesn't chop anything except the decimal portion. It makes sense since 40000 means something very different than 4000 (left-chopped) or 0000 (right-chopped), while 5.123 and 5.12 are similar. You can still use `printf`, but you first need to chop it using `substr`, `unpack` or some other means.	[reply] [d/l] [select]