Template: Matching an entire list

[ownedbycats]

I want one of the clauses of a gpm template to return true iif the only available format is epub. I tested with this:

program:
	f = $formats;

	if 
		list_contains(f, ',', '^epub$', '1', '') 
	then 'foobar' 
	fi

However, it returns 'foobar' for every book that has epub, even if there's other formats.

What should I do?

EDIT: Also, there are other times I might want to match the entire list contents. I am not sure how to do that though.

[davidfor]

My first thought is to count the items in the list:

program:
	f = $formats;
	if 
		(count(f, ',') == 1) && (list_contains(f, ',', '^epub$', '1', '') )
	then 'foobar' 
	fi

But, you can just do:

program:
	f = $formats;
	if 
		contains(f, '^epub$', '1', '')
	then 'foobar' 
	fi

[capink]

If you want to extend it so that it is able to test multiple formats also:

program:
    mandatory_formats = 'epub,mobi';
    if list_equals($formats, ',', mandatory_formats, ',', 1, '') then
        'foobar'
    fi

You can also modify this to be a stored template and use the arguments function so that you are able to pass the mandatory formats to the stored template.

[chaley]

This template solves the original problem, to test if a book has epub as its only format:

program:
	if $formats == 'epub' then
		'foobar'
	fi

This works because a list of 1 item has no separating commas.

[ownedbycats]

Thanks. I will save these

[ownedbycats]

Related to this: I am trying to remove "Crossover" from fanfics that technically aren't crossovers:

program:
	f = $#fanficcat;
	if 
		contains(
			f, '^Crossover, Mass Effect Trilogy, Mass Effect: Andromeda$', '1', ''
		)
	then 
		'Mass Effect Trilogy, Mass Effect: Andromeda' 
	elif
		contains(
			f, '^Crossover, Half-Life, Portal$', '1', ''
		)	
		then 'Half-Life, Portal' 
	else
		$#fanficcat;
	fi

However, this fails if the tags aren't in that exact order. Since it appears sorted in book details it's hard to tell. Is there a better way to do this?

[chaley]

Originally Posted by ownedbycats

Related to this: I am trying to remove "Crossover" from fanfics that technically aren't crossovers:

program:
	f = $#fanficcat;
	if 
		contains(
			f, '^Crossover, Mass Effect Trilogy, Mass Effect: Andromeda$', '1', ''
		)
	then 
		'Mass Effect Trilogy, Mass Effect: Andromeda' 
	elif
		contains(
			f, '^Crossover, Half-Life, Portal$', '1', ''
		)	
		then 'Half-Life, Portal' 
	else
		$#fanficcat;
	fi

However, this fails on two fronts:

1. In template tester, it only works if the tags are in that exact order. Since it appears sorted in book details it's hard to tell.
2. Putting this in Action Chains' single-field edit on a book that tested fine just results in the column becoming empty.

Any idea how to do this?

The function list_difference() removes items, but it isn't clear what you want to do. Are you asking to remove 'Crossover' if #fanficcat contains some other items, and possibly only those items? What exactly are the rules?

EDIT: writing down the rules can help with formulating the correct expressions. If you run into situations where the answer is "It depends" then you aren't done.

[ownedbycats]

The blanking out issue I fixed; it was a template error that I fixed in the post before testing in the chain itself.

If the only three items on the list are Crossover, Mass Effect Trilogy, Mass Effect: Andromeda then I want to remove Crossover.

If it's Crossover, foobar, Mass Effect Trilogy, Mass Effect: Andromeda then leave Crossover in.

[chaley]

Originally Posted by ownedbycats

The blanking out issue I fixed; it was a template error that I fixed in the post before testing in the chain itself.

If the only three items on the list are Crossover, Mass Effect Trilogy, Mass Effect: Andromeda then I want to remove Crossover.

If it's Crossover, foobar, Mass Effect Trilogy, Mass Effect: Andromeda then leave Crossover in.

This is an incredibly specific task that I wonder if it is best done with search/replace. That said, this template or something close to it should do the job

program:
	f = $#fanficcat;
	if list_count(f, ',') ==# 3 &&
		'^Crossover$' inlist f &&
		'^Mass Effect Trilogy$' inlist f &&
		'^Mass Effect: Andromeda$' inlist f then
			f = 'Mass Effect Trilogy, Mass Effect: Andromeda'
	fi;
	f

[ownedbycats]

it is pretty specific... the other alternative I thought of is that if I could sort the list before, it would definitely match Crossover, Mass Effect Trilogy, Mass Effect: Andromeda and I could use my original template. Or just checking all nine variations...

[chaley]

Originally Posted by ownedbycats

it is pretty specific... the other alternative I thought of is that if I could sort the list before, it would definitely match Crossover, Mass Effect Trilogy, Mass Effect: Andromeda and I could use my original template. Or just checking all nine variations...

You should be able to use list_sort() to do it this way. EDIT: But you need to be careful about the spacing. I think that the list will be separated by "comma space", but I can't guarantee that.

[ownedbycats]

program:
	f = list_sort($#fanficcat,0,',');

	if 
		contains(
			f, '^Crossover, Mass Effect Trilogy, Mass Effect: Andromeda$', '1', ''
		)
	then 
		'Mass Effect Trilogy, Mass Effect: Andromeda' 
	elif
		contains(
			f, '^Crossover, Half-Life, Portal$', '1', ''
		)	
		then 'Half-Life, Portal' 
	else
		$#fanficcat;
	fi

This gave me a EXCEPTION: Formatter: Expected an expression, found 'fi' near Unknown on line 17.

[chaley]

Remove the semicolon on line 16

$#fanficcat;

A semicolon is a statement separator, not a statement terminator. That means that if a semicolon is found then it must be followed by another expression.

The exception is a semicolon at the end of the program. That is accepted because a) it can be detected and b) so many people put one there that I got tired of the questions.

[ownedbycats]

I'm not even sure how that semicolon got there. thank you

[chaley]

Originally Posted by ownedbycats

I'm not even sure how that semicolon got there. thank you

This isn't intended to be criticism ...

When I was first studying computing back in 1971 we had to submit "card decks", a pile of punched computer cards, to run. If *anything* was wrong it cost us hours, so we learned to carefully check the cards for spelling, logic, and syntactic errors before submitting them. Once submitted we would go to the local all-night pizza or donut shop to wait, only to come back to see the error message "You have too many errors. You must be stoned" (yes, this was a real message) and to do it all again.

Today the time penalty for not checking for errors is near zero, so people don't bother. The process is "hack at it until it seems to work." The downside: people have less understanding of the basic details.

On the other hand, today people are using very complex "APIs" (Application Programming Interfaces) that do an incredible amount of work for you. Problem: these APIs aren't sufficiently documented so the only way to determine what they do is to try using them, over and over until one is confident that the behavior is understood. But of course the behavior isn't understood because the edge conditions aren't tested. Sigh...