Generate non-overlapping (gapped) sequences. Partitioned sequences auto-incrementing numbers. generate-non-overlapping-sequences.js

generate-non-overlapping-sequences.js

/**
 * Quick no doco utility fn. Not sure what the SEO for searchability of this concept is. 
 * Reminds me of a hacky workaround for distributed database reconciliation of separately incrementing distinct IDs
 * Can probably shift some parameters to different places.
 * Functional! FP
 * - TODO: Rename to "get____" function name convention. ref transp
 */
function generateSteppedSequence(
    stepNumberOfSequences,
    originalBaseLength,
    baseOffsetStartingNumber,
    sequenceIdentifier
) {
    // Valid range sequenceIdentifier should be 1 2 3 if base length is 3
    const sequenceOffsetForNoOverlaps = sequenceIdentifier - 1;

    const individualPartitionedSteppedLength = Math.ceil(originalBaseLength / stepNumberOfSequences);

    const separatedNumbers = Array.from(
        {
            length: individualPartitionedSteppedLength,
        },
        (empty, index) => {
            const calcJump =
                index * stepNumberOfSequences +
                sequenceOffsetForNoOverlaps +
                baseOffsetStartingNumber;

            return calcJump;
        }
    );

    return separatedNumbers;
}


function generateNonOverlappingSequences(
    numSequences,
    totalBaseLength,
    baseOffset
) {
    const parentArrayOfSequences = Array.from(
        { length: numSequences },

        (empty, index) =>
            generateSteppedSequence(
                numSequences,
                totalBaseLength,
                baseOffset,
                index
            )
    );

    return parentArrayOfSequences;
}


/**
 * Building word list
 * @param listOfSequences {Array<number[]>} the sequences we generated
 * naming things is hard
 */
function getListOfTextNewLinedStringifiedSequencesFromListOfSequences(listOfSequences) {
    const listOfInterspersedNewLinedSequences = listOfSequences.map(sequence => sequence.join("\n"));
    

    return listOfInterspersedNewLinedSequences;
}

/**
 * Quick fn
 */
function directMapListOfUrlsFromTokenIDs(sequence, strPrefix="https://.../?&blah=") {
    return sequence
        .map(
            (num) =>
                `${strPrefix}${num}`
        )
        // New line accumulated into single string
        .join("\n");
}

oof 16 million characters for one of the splits? whoops!

nothing to see here
not generating wordlists 👀