axyseo 2.0.0-alpha.0.0.4 → 2.0.0-alpha.0.0.5

package/src/parsedPaper/build/tree/cleanup/getElementContent.js

@@ -1,21 +0,0 @@
-/**
- * Gets the content of an element (the part _between_ the opening and closing tag) from the HTML source code.
- *
- * @param {module:parsedPaper/structure.Node|module:parsedPaper/structure.FormattingElement} element The element to parse the contents of
- * @param {string} html                                                                The source code to parse the contents from
- *
- * @returns {string} The element's contents.
- *
- * @private
- */
-const getElementContent = function( element, html ) {
-	const location = element.location;
-	if ( location ) {
-		const start = location.startTag ? location.startTag.endOffset : location.startOffset;
-		const end = location.endTag ? location.endTag.startOffset : location.endOffset;
-		return html.slice( start, end );
-	}
-	return "";
-};
-
-export default getElementContent;

package/src/parsedPaper/build/tree/cleanup/postParsing.js

@@ -1,37 +0,0 @@
-import { LeafNode } from "../../../structure/tree";
-import calculateTextIndices from "./calculateTextIndices";
-
-/**
- * Cleans up a node in the tree.
- *
- * @param {module:parsedPaper/structure.Node} node The node that needs to be cleaned.
- *
- * @returns {module:parsedPaper/structure.Node} The cleaned up node.
- *
- * @private
- */
-const cleanUpNode = function( node ) {
-	// Clean up formatting elements in headings and paragraphs.
-	if ( node instanceof LeafNode ) {
-		// Start and end position in leaf node's (header's or paragraph's) text without formatting.
-		calculateTextIndices( node );
-	}
-	return node;
-};
-
-/**
- * Cleans up the given tree after parsing of the HTML source code
- * by setting the start and end index of each formatting element in a leaf node's text.
- *
- * @param {module:parsedPaper/structure.Node} tree The tree structure to be cleaned.
- *
- * @returns {module:parsedPaper/structure.Node} The cleaned up tree.
- *
- * @private
- */
-const cleanUpTree = function( tree ) {
-	tree.map( node => cleanUpNode( node ) );
-	return tree;
-};
-
-export default cleanUpTree;

package/src/parsedPaper/build/tree/html/HTMLTreeConverter.js

@@ -1,230 +0,0 @@
-import { FormattingElement, Heading, Paragraph, StructuredNode, List, ListItem } from "../../../structure/tree";
-import LeafNode from "../../../structure/tree/nodes/LeafNode";
-import { formattingElements, headings, ignoredHtmlElements } from "./htmlConstants";
-
-/**
- * Converts a parse5 tree to a tree that can be analyzed.
- */
-class HTMLTreeConverter {
-	/**
-	 * Converts the parse5 tree to a Yoast tree.
-	 *
-	 * @param {Object} parse5Tree The parse5 tree to convert.
-	 *
-	 * @returns {module:parsedPaper/structure.Node} The converted tree.
-	 */
-	convert( parse5Tree ) {
-		const root = new StructuredNode( "root", null );
-		this._convert( parse5Tree, root );
-		return root;
-	}
-
-	/**
-	 * Converts the tree from a parse5 implementation to a Yoast tree.
-	 *
-	 * @param {Object}                                parse5Tree    The parse5 tree to convert.
-	 * @param {module:parsedPaper/structure.Node}     parent        The tree in progress.
-	 *
-	 * @returns {void}
-	 *
-	 * @private
-	 */
-	_convert( parse5Tree, parent ) {
-		if ( ignoredHtmlElements.includes( parse5Tree.nodeName ) ) {
-			return;
-		}
-		if ( parse5Tree.childNodes ) {
-			for ( const node of parse5Tree.childNodes ) {
-				const child = this._createChild( node, parent );
-
-				if ( child ) {
-					parent.addChild( child );
-					this._convert( node, child );
-				} else {
-					this._convert( node, parent );
-				}
-			}
-		}
-	}
-
-	/**
-	 * Creates a new node for in the structured tree from the given parse 5 node.
-	 * Returns `null` if no node should be added to the tree.
-	 *
-	 * @param {Object}                            parse5Node The parse5 node that should be parsed.
-	 * @param {module:parsedPaper/structure.Node} parentNode The parent node.
-	 *
-	 * @returns {module:parsedPaper/structure.Node|null} The node that should be added to the tree, or `null` if no node should be added.
-	 *
-	 * @private
-	 */
-	_createChild( parse5Node, parentNode ) {
-		let child = null;
-
-		const nodeType = parse5Node.nodeName;
-
-		if ( ignoredHtmlElements.includes( nodeType ) ) {
-			const formatting = new FormattingElement( nodeType, parse5Node.sourceCodeLocation, this._parseAttributes( parse5Node.attrs ) );
-			this._addLeafNodeContent( formatting, this._addFormatting, parentNode, parse5Node.sourceCodeLocation );
-		} else if ( nodeType === "p" ) {
-			// Paragraph.
-			child = new Paragraph( parse5Node.sourceCodeLocation );
-		} else if ( headings.includes( nodeType ) ) {
-			// Heading.
-			child = new Heading( parseInt( nodeType[ 1 ], 10 ), parse5Node.sourceCodeLocation );
-		} else if ( nodeType === "li" ) {
-			// List item.
-			child = new ListItem( parse5Node.sourceCodeLocation );
-		} else if ( nodeType === "ol" || nodeType === "ul" ) {
-			// List node.
-			child = new List( nodeType === "ol", parse5Node.sourceCodeLocation );
-		} else if ( nodeType === "#text" ) {
-			// Text (outside of an ignored element).
-			child = this._addLeafNodeContent( parse5Node.value, this._addText, parentNode, parse5Node.sourceCodeLocation );
-		} else if ( formattingElements.includes( nodeType ) ) {
-			// Formatting element.
-			const formatting = new FormattingElement( nodeType, parse5Node.sourceCodeLocation, this._parseAttributes( parse5Node.attrs ) );
-			child = this._addLeafNodeContent( formatting, this._addFormatting, parentNode, parse5Node.sourceCodeLocation );
-		} else {
-			// Other element (`div`, `section`, `article`, etc.).
-			child = new StructuredNode( nodeType, parse5Node.sourceCodeLocation );
-		}
-
-		return child;
-	}
-
-	/**
-	 * Adds leaf node content (text or formatting) to the tree according to a predefined strategy:
-	 *
-	 * If the content has a leaf node (paragraph, heading, list item) parent or ancestor: add it to this leaf node.
-	 * Else: wrap the content in an implicit paragraph, and add the content to it in such a way that runs
-	 * of phrasing content make up one implicit paragraph.
-	 *
-	 * @param {string|module:parsedPaper/structure.FormattingElement} contentToAdd The content to add.
-	 * @param {function}                                              add          A function that adds the content to the specified leaf node.
-	 * @param {module:parsedPaper/structure.Node}                     parent       The parent to which to try to add the content.
-	 * @param {Object}                                                location     The location of the content as parsed by parse5.
-	 *
-	 * @returns {null|module:parsedPaper/structure.Paragraph|null} The implicit paragraph, if one was created.
-	 *
-	 * @private
-	 */
-	_addLeafNodeContent( contentToAdd, add, parent, location ) {
-		if ( parent instanceof LeafNode ) {
-			add( parent, contentToAdd );
-			return null;
-		}
-
-		const previousChild = this._previousChild( parent );
-		const leafNodeAncestor = this._leafNodeAncestor( parent );
-
-		if ( leafNodeAncestor instanceof Paragraph ) {
-			add( leafNodeAncestor, contentToAdd );
-			return null;
-		} else if ( previousChild && previousChild instanceof Paragraph && previousChild.isImplicit ) {
-			add( previousChild, contentToAdd );
-			return null;
-		}
-
-		// Don't create an implicit paragraph when the content is whitespace only.
-		if ( /^\s*$/.exec( contentToAdd ) ) {
-			return null;
-		}
-
-		const child = new Paragraph( location, true );
-		add( child, contentToAdd );
-		return child;
-	}
-
-	/**
-	 * Add text to the parent leaf node.
-	 *
-	 * @param {module:parsedPaper/structure.LeafNode} parent The leaf node to which to add the text.
-	 * @param {string}                                text   The text to add to the node.
-	 *
-	 * @returns {void}
-	 *
-	 * @private
-	 */
-	_addText( parent, text ) {
-		parent.appendText( text );
-	}
-
-	/**
-	 * Add a formatting element to the parent leaf node.
-	 *
-	 * @param {module:parsedPaper/structure.LeafNode}          parent     The leaf node to which to add the formatting.
-	 * @param {module:parsedPaper/structure.FormattingElement} formatting The formatting element to add.
-	 *
-	 * @returns {void}
-	 *
-	 * @private
-	 */
-	_addFormatting( parent, formatting ) {
-		parent.addFormatting( formatting );
-	}
-
-	/**
-	 * Parses the HTML element attributes from parse5's format to a plain JS object.
-	 *
-	 * @example
-	 *
-	 * const attributes = _parseAttributes( { name: "id", value: "an-id" } ) // becomes { id: "an-id" }.
-	 *
-	 * @param {Array<{ name: string, value: string }>} parse5attributes The attributes as parsed by parse5.
-	 *
-	 * @returns {Object} The parsed attributes.
-	 *
-	 * @private
-	 */
-	_parseAttributes( parse5attributes ) {
-		if ( parse5attributes && parse5attributes.length > 0 ) {
-			return parse5attributes.reduce( ( attributes, attribute ) => {
-				attributes[ attribute.name ] = attribute.value;
-				return attributes;
-			}, {} );
-		}
-		return null;
-	}
-
-	/**
-	 * Returns the last child of the given parent.
-	 *
-	 * @param {module:parsedPaper/structure.StructuredNode} parent The parent.
-	 *
-	 * @returns {module:parsedPaper/structure.Node} The parent's last child.
-	 *
-	 * @private
-	 */
-	_previousChild( parent ) {
-		if ( parent.children ) {
-			return parent.children[ parent.children.length - 1 ];
-		}
-		return null;
-	}
-
-	/**
-	 * Try to find the leaf node ancestor of this node. Returns `null` if the node does not have one.
-	 *
-	 * @param {module:parsedPaper/structure.Node} element The node for which to find the leaf node ancestor.
-	 *
-	 * @returns {null|module:parsedPaper/structure.Node} The leaf node ancestor, if it has one, `null` if not.
-	 *
-	 * @private
-	 */
-	_leafNodeAncestor( element ) {
-		const parent = element.parent;
-
-		if ( ! parent ) {
-			return null;
-		}
-
-		if ( element instanceof LeafNode ) {
-			return parent;
-		}
-
-		return this._leafNodeAncestor( parent );
-	}
-}
-
-export default HTMLTreeConverter;

package/src/parsedPaper/build/tree/html/buildTree.js

@@ -1,31 +0,0 @@
-import { parseFragment } from "parse5";
-
-import HTMLTreeConverter from "./HTMLTreeConverter";
-import cleanUpTree from "../cleanup/postParsing";
-
-
-/**
- * Parses the given html-string to a tree, to be used in further analysis.
- *
- * @param {string} html The html-string that should be parsed.
- *
- * @returns {module:parsedPaper/structure.Node} The build tree.
- *
- * @memberOf module:parsedPaper/builder
- */
-const buildTree = function( html ) {
-	/*
-	 * Parsing of a HTML article takes on average 19ms
-	 * (based on the fullTexts in the specs (n=24), measured using `console.time`).
-	 */
-	const parse5Tree = parseFragment( html, { sourceCodeLocationInfo: true } );
-
-	const htmlTreeConverter = new HTMLTreeConverter();
-	let tree = htmlTreeConverter.convert( parse5Tree );
-
-	// Cleanup takes < 2ms.
-	tree = cleanUpTree( tree );
-	return tree;
-};
-
-export default buildTree;

package/src/parsedPaper/build/tree/html/htmlConstants.js

@@ -1,37 +0,0 @@
-/**
- * All HTML element tags that we treat as formatting / phrasing content.
- * @type {string[]}
- * @const
- *
- * @see https://developer.mozilla.org/en-US/docs/Web/Guide/HTML/Content_categories#Phrasing_content
- *
- * @memberOf module:parsedPaper/builder
- */
-const formattingElements = [ "a", "abbr", "audio", "b", "bdo", "button", "canvas", "cite", "command",
-	"data", "datalist", "dfn", "del", "em", "embed", "i", "iframe", "img", "input", "kbd", "keygen", "label", "mark",
-	"math", "meter", "noscript", "object", "output", "progress", "q", "ruby", "samp", "select", "small",
-	"span", "strong", "sub", "sup", "svg", "textarea", "time", "var", "video", "wbr" ];
-
-/**
- * All tags of the HTML element types that are ignored in our analysis.
- * @type {string[]}
- * @const
- *
- * @memberOf module:parsedPaper/builder
- */
-const ignoredHtmlElements = [ "script", "style", "pre", "#comment", "code", "br" ];
-
-/**
- * All HTML heading element tags, from `h1` to `h6`.
- * @type {string[]}
- * @const
- *
- * @memberOf module:parsedPaper/builder
- */
-const headings = [ "h1", "h2", "h3", "h4", "h5", "h6" ];
-
-export {
-	formattingElements,
-	ignoredHtmlElements,
-	headings,
-};

package/src/parsedPaper/build/tree/index.js

@@ -1,14 +0,0 @@
-import TreeBuilder from "./TreeBuilder";
-
-/**
- * This module contains the main logic to transform a formatted text (for now only HTML is supported)
- * to a tree representation.
- *
- * This tree structure in turn can be used for further textual analysis.
- *
- * @module parsedPaper/builder
- *
- * @see module:parsedPaper/structure
- */
-
-export default TreeBuilder;

package/src/parsedPaper/build/tree/metadata/buildTree.js

@@ -1,32 +0,0 @@
-import parseSynonyms from "../../../../languageProcessing/helpers/sanitize/parseSynonyms";
-
-import { MetadataMiscellaneous, MetadataText, StructuredNode } from "../../../structure/tree/nodes";
-
-/**
- * Parses the given paper attributes to a tree.
- *
- * @param {Paper} paper The Paper to parse.
- *
- * @memberOf module:parsedPaper/builder
- *
- * @returns {module:parsedPaper/structure.Node} The build tree.
- */
-const buildTree = function buildTree( paper ) {
-	const metadata = new StructuredNode( "metadata", null );
-
-	// Add text types.
-	metadata.addChild( new MetadataText( "title", paper.getTitle() ) );
-	metadata.addChild( new MetadataText( "description", paper.getDescription() ) );
-
-	// Add miscellaneous types.
-	metadata.addChild( new MetadataMiscellaneous( "keyphrase", paper.getKeyword() ) );
-	metadata.addChild( new MetadataMiscellaneous( "synonyms", parseSynonyms( paper.getSynonyms() ) ) );
-	metadata.addChild( new MetadataMiscellaneous( "slug", paper.getSlug() ) );
-	metadata.addChild( new MetadataMiscellaneous( "titleWidth", paper.getTitleWidth() ) );
-	metadata.addChild( new MetadataMiscellaneous( "permalink", paper.getPermalink() ) );
-	metadata.addChild( new MetadataMiscellaneous( "locale", paper.getLocale() ) );
-
-	return metadata;
-};
-
-export default buildTree;

package/src/parsedPaper/research/TreeResearcher.js

@@ -1,134 +0,0 @@
-import { get, has } from "lodash";
-
-/**
- * This contains all possible, default researches
- * and logic to apply these researches to a formatted text,
- * represented as a tree structure.
- *
- * @memberOf module:parsedPaper/research
- */
-class TreeResearcher {
-	/**
-	 * Makes a new TreeResearcher.
-	 */
-	constructor() {
-		this._researches = {};
-		this._data = {};
-	}
-
-	/**
-	 * Adds or overwrites a research to the list of available researches.
-	 *
-	 * **Note**: When a research is already known under the given name,
-	 * the previous research with this name gets overwritten!
-	 *
-	 * @param {string} name       The ID to which to map the research to.
-	 * @param {Research} research The research to add.
-	 *
-	 * @returns {void}
-	 */
-	addResearch( name, research ) {
-		this._researches[ name ] = research;
-	}
-
-	/**
-	 * Returns all available researches.
-	 *
-	 * @returns {Object} An object containing all available researches.
-	 */
-	getResearches() {
-		return this._researches;
-	}
-
-	/**
-	 * Returns whether a research is known under this name.
-	 *
-	 * @param {string} name The name to get the research from.
-	 *
-	 * @returns {boolean} If a research is known under this name.
-	 */
-	hasResearch( name ) {
-		return has( this._researches, name );
-	}
-
-	/**
-	 * Gets the research with the given name.
-	 * If a research is not known under this name, false is returned instead.
-	 *
-	 * @throws {Error} When a research is not known under the given name.
-	 *
-	 * @param {string} name The name of the research to get.
-	 *
-	 * @returns {Research} The research stored under the given name.
-	 */
-	getResearch( name ) {
-		if ( this.hasResearch( name ) ) {
-			return this._researches[ name ];
-		}
-		throw new Error( `'${name}' research does not exist.` );
-	}
-
-	/**
-	 * Applies the research with the given name to the node and its descendants.
-	 *
-	 * @param {string} name The name of the research to apply to the node.
-	 * @param {module:parsedPaper/structure.Node} node The node to compute the research of.
-	 * @param {module:parsedPaper/structure.StructuredNode} metadata The node that holds the paper metadata.
-	 * @param {boolean} [bustCache=false] If we should force the results, as cached on each node, to be recomputed.
-	 *
-	 * @returns {Promise<*>} A promising research result.
-	 */
-	async doResearch( name, node, metadata, bustCache = false ) {
-		const research = this.getResearch( name );
-		let researchResult = Promise.resolve();
-
-		if ( research.isLeafNode( node ) ) {
-			/*
-			  Compute research results for this node, or use the cached results when available.
-			  Always compute it when we need to bust the cache.
-			 */
-			if ( ! node.hasResearchResult( name ) || bustCache ) {
-				node.setResearchResult( name, await research.calculateFor( node, metadata ) );
-			}
-			researchResult = node.getResearchResult( name );
-		} else {
-			const children = node.children;
-
-			// Heading and paragraph nodes do not have children.
-			if ( children ) {
-				const resultsForChildren = await Promise.all( children.map( ( child ) => {
-					return this.doResearch( name, child, metadata );
-				} ) );
-
-				researchResult = research.mergeChildrenResults( resultsForChildren );
-			}
-		}
-
-		return researchResult;
-	}
-
-	/**
-	 * Add research data to the researcher by the research name.
-	 *
-	 * @param {string} researchName The identifier of the research.
-	 * @param {Object} data         The data object.
-	 *
-	 * @returns {void}.
-	 */
-	addResearchData( researchName, data ) {
-		this._data[ researchName ] = data;
-	}
-
-	/**
-	 * Return the research data from a research data provider by research name.
-	 *
-	 * @param {string} researchName The identifier of the research.
-	 *
-	 * @returns {Object|boolean} The data provided by the provider, false if the data do not exist
-	 */
-	getData( researchName ) {
-		return get( this._data, researchName, false );
-	}
-}
-
-export default TreeResearcher;

package/src/parsedPaper/research/index.js

@@ -1,13 +0,0 @@
-import TreeResearcher from "./TreeResearcher";
-
-/**
- * Module for applying researches to text, represented as a tree structure.
- *
- * @see module:parsedPaper/structure
- *
- * @module parsedPaper/research
- */
-
-export {
-	TreeResearcher,
-};

package/src/parsedPaper/research/researches/Headings.js

@@ -1,20 +0,0 @@
-import { Heading } from "../../structure/tree";
-import Research from "./Research";
-
-/**
- * A research giving back the headings located in a text.
- */
-class Headings extends Research {
-	/**
-	 * Calculates the result of the research for the given Node.
-	 *
-	 * @param {module:parsedPaper/structure.Node} node The node to do the research on.
-	 *
-	 * @returns {Promise<module:parsedPaper/structure.Heading[]|[]>} The result of the research.
-	 */
-	calculateFor( node ) {
-		return node instanceof Heading ? Promise.resolve( [ node ] ) : Promise.resolve( [] );
-	}
-}
-
-export default Headings;