In if condition I used to take one value from log file after matching the particular pattern. That pattern is matched two times in log file. While matching the pattern first time that value is 0 and second time value is 48. It may be also reverse. First value may contain 48 and second value may contain 0. I need to calculate the exact value. So I planned to add these two values. but after adding these two values also while printing the total value in if condition I used to get the two values separately. But I need single value only.
Please give me solution to solve this issue.
Thanks in advance.
Do you mean something like this:
my $entry = "First is 10, seconds is 48";
if(my ($a,$b) = $entry =~ /(\d+)/g) {
print $a + $b,"\n"; # 58
}
But without actual code it is hard to see what your problem really is.
Related
I'd like to understand how unpack is returning the sum in the given perl one-liner.
I've looked at pack man page and mostly understood that it is simply formatting the given array into a scalar of ten doubles.
However, I couldn't find proper documentation for unpack with %123. Looking for help here.
print unpack "%123d*" , pack( "d*", (1..10));
This line correctly outputs 55 which is 1+2+3+...+10.
From perldoc -f unpack:
In addition to fields allowed in pack(), you may prefix a field with a % to indicate that you want a <number>-bit checksum of the items instead of the items themselves.
Thus %123d* means to add all the input integers 1..10 and then take the first 123 bit of this result in order to construct the "<number>-bit checksum". Note that %8d* or just %d* (which is equivalent to %16d*) would suffice too given that the sum is small enough.
I am still learning perl and have all most got a program written. My question, as simple as it may be, is if I want to hardcode a string to a field would the below do that? Thank you :).
$out[45]="VUS";
In the other lines I use the below to define the values that are passed into the `$[out], but the one in question is hardcoded and the others come from a split.
my #vals = split/\t/; # this splits the line at tabs
my #mutations=split/,/,$vals[9]; # splits on comma to create an array of mutations
my ($gene,$transcript,$exon,$coding,$aa);
for (#mutations)
{
($gene,$transcript,$exon,$coding,$aa) = split/\:/; # this takes col AB and splits it at colons
grep {$transcript eq $_} keys %nms or next;
}
my #out=($.,#colsleft,$_,#colsright);
$out[2]=$gene;
$out[3]=$nms{$transcript};
$out[4]=$transcript;
$out[15]=$coding;
$out[17]=$aa;
Your line of code: $out[45]="VUS"; is correct in that it is defining that 46th element of the array #out to the string, "VUS". I am trying to understand from your code, however why you would want to do that? Usually, it is better practice to not hardcode if at all possible. You want to make it your goal to make your program as dynamic as possible.
I use an ActivePerl script to take in CSV files and create XML files that I load into a database. These are userid database entries, name, address, etc. We've always used the home phone number field to generate an initial password (which we encourage the users to change immediately!). The proliferation of cellphones means I have a bunch of people with no home phone, so I want to use the cell phone field when the home phone field is empty.
My input fields look like this:
# 0 Firstname
# 1 Lastname
# 2 VP (voicepart)
# 3 Address
# 4 City
# 5 State
# 6 Zip
# 7 Phone
# 8 Mobile
# 9 Email
Here's the Perl code I've worked up to create the password - the create_password subroutine is working when there's a value in field 7:
my $pass_word = '';
my $pass_word = create_password($fields[7]);
if (my $pass_word = '') {
print "Use the cell phone number \n";
my $pass_word = create_password($fields[8]);
}
The "print" statement is to tell me what it thinks it's doing.
This looks to me like it should work, but the "if" statment never fires. The Print statement doesn't print, and nobody with a value only in field 8 ever gets a password generated. There must be something wrong with the way I'm testing the value of $pass_word but I can't see it. Should I be testing the values of $fields[7] and $fields[8] instead of the variable value? How DO you test a Perl variable for null value if this doesn't work?
You have several problems in your code.
First of all, after you declared a variable using my, you don't need to add my before the variable when you use it;
Secondly, for this line:
if (my $pass_word = '')
I think you meant
if ($pass_word == '')
(my is removed, as talked in the first point)
= means assignment, which returns the value you assigned to $pass_word, which is '' here, that's why this condition always return false.
But still, == is not correct here. In perl, we use eq to compare two strings. == is used to compare numbers.
So, remove all the my except the first one, and use eq to compare your strings.
You've got two major problems in here.
First one is your string equality test. In Perl, strings are compared for equality using operator eq (as in $string eq 'something'). = is the assignment operator.
Second one is your (ab)use of my. Each my declares a new variable that “hides” the previous one, so in effect you can never re-use its value, you're confronted to undef every time.
Replace = with eq in your if clause; remove all but the first uses of my, and you should be set!
my declares a new variable which hides the variable with the same name in the surrounding scope. Remove the excessive use of my.
I have a file that I am reading in. I'm using perl to reformat the date. It is a comma seperated file. In one of the files, I know that element.0 is a zipcode and element.1 is a counter. Each row can have 1-n number of cities. I need to know the number of elements from element.3 to the end of the line so that I can reformat them properly. I was wanting to use a foreach loop starting at element.3 to format the other elements into a single string.
Any help would be appreciated. Basically I am trying to read in a csv file and create a cpp file that can then be compiled on another platform as a plug-in for that platform.
Best Regards
Michael Gould
you can do something like this to get the fields from a line:
my #fields = split /,/, $line;
To access all elements from 3 to the end, do this:
foreach my $city (#fields[3..$#fields])
{
#do stuff
}
(Note, based on your question I assume you are using zero-based indexing. Thus "element 3" is the 4th element).
Alternatively, consider Text::CSV to read your CSV file, especially if you have things like escaped delimiters.
Well if your line is being read into an array, you can get the number of elements in the array by evaluating it in scalar context, for example
my $elems = #line;
or to be really sure
my $elems = scalar(#line);
Although in that case the scalar is redundant, it's handy for forcing scalar context where it would otherwise be list context. You can also find the index of the last element of the array with $#line.
After that, if you want to get everything from element 3 onwards you can use an array slice:
my #threeonwards = #line[3 .. $#line];
I need an algorithm that identifies all possible combinations of a set of numbers that sum to some other number.
For example, given the set {2,3,4,7}, I need to know all possible subsets that sum to x. If x == 12, the answer is {2,3,7}; if x ==7 the answer is {{3,4},{7}} (ie, two possible answers); and if x==8 there is no answer. Note that, as these example imply, numbers in the set cannot be reused.
This question was asked on this site a couple years ago but the answer is in C# and I need to do it in Perl and don't know enough to translate the answer.
I know that this problem is hard (see other post for discussion), but I just need a brute-force solution because I am dealing with fairly small sets.
sub Solve
{
my ($goal, $elements) = #_;
# For extra speed, you can remove this next line
# if #$elements is guaranteed to be already sorted:
$elements = [ sort { $a <=> $b } #$elements ];
my (#results, $RecursiveSolve, $nextValue);
$RecursiveSolve = sub {
my ($currentGoal, $included, $index) = #_;
for ( ; $index < #$elements; ++$index) {
$nextValue = $elements->[$index];
# Since elements are sorted, there's no point in trying a
# non-final element unless it's less than goal/2:
if ($currentGoal > 2 * $nextValue) {
$RecursiveSolve->($currentGoal - $nextValue,
[ #$included, $nextValue ],
$index + 1);
} else {
push #results, [ #$included, $nextValue ]
if $currentGoal == $nextValue;
return if $nextValue >= $currentGoal;
}
} # end for
}; # end $RecursiveSolve
$RecursiveSolve->($goal, [], 0);
undef $RecursiveSolve; # Avoid memory leak from circular reference
return #results;
} # end Solve
my #results = Solve(7, [2,3,4,7]);
print "#$_\n" for #results;
This started as a fairly direct translation of the C# version from the question you linked, but I simplified it a bit (and now a bit more, and also removed some unnecessary variable allocations, added some optimizations based on the list of elements being sorted, and rearranged the conditions to be slightly more efficient).
I've also now added another significant optimization. When considering whether to try using an element that doesn't complete the sum, there's no point if the element is greater than or equal to half the current goal. (The next number we add will be even bigger.) Depending on the set you're trying, this can short-circuit quite a bit more. (You could also try adding the next element instead of multiplying by 2, but then you have to worry about running off the end of the list.)
The rough algorithm is as follows:
have a "solve" function that takes in a list of numbers already included and a list of those not yet included.
This function will loop through all the numbers not yet included.
If adding that number in hits the goal then record that set of numbers and move on,
if it is less than the target recursively call the function with the included/exluded lists modified with the number you are looking at.
else just go to the next step in the loop (since if you are over there is no point trying to add more numbers unless you allow negative ones)
You call this function initially with your included list empty and your yet to be included list with your full list of numbers.
There are optimisations you can do with this such as passing the sum around rather than recalculating each time. Also if you sort your list initially you can do optimisations based on the fact that if adding number k in the list makes you go over target then adding k+1 will also send you over target.
Hopefully that will give you a good enough start. My perl is unfortuantely quite rusty.
Pretty much though this is a brute force algorithm with a few shortcuts in it so its never going to be that efficient.
You can make use of the Data::PowerSet module which generates all subsets of a list of elements:
Use Algorithm::Combinatorics. That way, you can decide ahead of time what size subsets you want to consider and keep memory use to a minimum. Apply some heuristics to return early.
#!/usr/bin/perl
use strict; use warnings;
use List::Util qw( sum );
use Algorithm::Combinatorics qw( combinations );
my #x = (1 .. 10);
my $target_sum = 12;
{
use integer;
for my $n ( 1 .. #x ) {
my $iter = combinations(\#x, $n);
while ( my $set = $iter->next ) {
print "#$set\n" if $target_sum == sum #$set;
}
}
}
The numbers do blow up fairly rapidly: It would take thousands of days to go through all subsets of a 40 element set. So, you should decide on the interesting sizes of subsets.
Is this a 'do my homework for me' question?
To do this deterministically would need an algorithm of order N! (i.e. (N-0) * (N-1) * (N-2)...) which is going to be very slow with large sets of inputs. But the algorithm is very simple: work out each possible sequence of the inputs in the set and try adding up the inputs in the sequence. If at any point the sum matches, you've got one of the answers, save the result and move on to the next sequence. If at any point the sum is greater than the target, abandon the current sequence and move on to the next.
You could optimize this a little by deleting any of the inputs greater than the target. Another approach for optimization would be to to take the first input I in the sequence and create a new sequence S1, deduct I from the target T to get a new target T1, then check if T exists in S1, if it does then you've got a match, otherwise repeat the process with S1 and T1. The order is still N! though.
If you needed to do this with a very large set of numbers then I'd suggest reading up on genetic algorithms.
C.
Someone posted a similar question a while ago and another person showed a neat shell trick to answer it. Here is a shell technique, but I don't think it is as neat a solution as the one I saw before (so I'm not taking credit for this approach). It's cute because it takes advantage of shell expansion:
for i in 0{,+2}{,+3}{,+4}{,+7}; do
y=$(( $i )); # evaluate expression
if [ $y -eq 7 ]; then
echo $i = $y;
fi;
done
Outputs:
0+7 = 7
0+3+4 = 7