Muriel | An electron app to binge-watch anything | Runtime Evironment library

The easiest way is to use pandas directly:

We can filter clients which have both the conditions satisfied.

You have a problem in which you need to find an equilibrium based on constraints.

I propose here a solution in which we do not work with pandas, but rather pure python lists because pandas' overhead would only slow the solution down.

First of all, there's an obvious possiblity of an infinite loop. As such, you have to keep track of past arrangement configurations so as to know when you have made a circle and began a loop.

I have created a custom error InfeasableError (for readability) which is raised whenever this happens. The rest of the code should be straightforward.

Let's do this interactively (in SWI Prolog) instead of in a script which prints the answers at the end using format/2.

We want all possible ancestors of eve in a list.

So we have to

1. query the Prolog program for all possible solutions to the goal ancestor(A,eve)
2. and then collect them into a list

This is done using one of the predicates bagof/3, setof/3 or findall/3, which backtrack over answers to a goal and unify a variable with a list containing all the answers (with duplicate answers for bagof/3, without duplicate answers for setof/3, and with "no possible answer" yielding [] instead of failure for findall/3).

So we just need to make sure the goal to find any ancestor is correct.

We can state that A is an ancestor of C if

• A is a parent of C or
• A is a parent of some D, and D is an ancestor of C

(Note: just 'if', not 'if an only if'. However, it is assumed there are no other ways in which A could possibly be an ancestor of C ... a reasonable "closed world assumption")

The above formulation is well adapted to the search strategy of Prolog, which attempts to resolve a leftmost sub-goal in the body first:

It seems that invoked with a file name, gprolog validates it but does not load its content into the interpreter. In order to load the file content you have to use the --consult-file command line option:

Bit complex data structure, you should clean. However, solved what expected.

I want the sum of the "speech share"

There is no "speech share" in the XML you show. I am assuming you meant the total time of each contributor's segments. This is not trivial to do in XSLT 1.0.

First, you need to compute the individual pt:segment durations. For this, you need to convert the start and end times to seconds (or some other common unit) so that you can subtract them. And you need to convert the results to a node-set so that they can be summed.

Try something like the following:

XSLT 1.0 (+ EXSLT)

Consider the following example:

XSLT 1.0