word-aligner 0.4.0 → 1.0.1-alpha

package/lib/js/utils/MorphUtils.js

@@ -3,7 +3,7 @@
 Object.defineProperty(exports, "__esModule", {
   value: true
 });
-exports.getMorphLocalizationKeysGreek = exports.getMorphLocalizationKeysHebrewAramaic = exports.getMorphLocalizationKeys = undefined;
+exports.getMorphLocalizationKeysGreekSR = exports.getMorphLocalizationKeysGreek = exports.getMorphLocalizationKeysHebrewAramaic = exports.getMorphLocalizationKeys = undefined;
 
 var _getIterator2 = require('babel-runtime/core-js/get-iterator');
 
@@ -29,6 +29,9 @@ var getMorphLocalizationKeys = exports.getMorphLocalizationKeys = function getMo
 
     case 'gr':
     default:
+      if (morph && morph.length === 12) {
+        return getMorphLocalizationKeysGreekSR(morph);
+      }
       return getMorphLocalizationKeysGreek(morph);
   }
 };
@@ -151,4 +154,53 @@ var getMorphLocalizationKeysGreek = exports.getMorphLocalizationKeysGreek = func
     } // unknown code, prefixing with '*'
   });
   return morphKeys;
+};
+/**
+ * @description - Get a list of all the localization keys for a morph string in Greek
+ * @param {String} morph - the morph string, e.g. Gr,N,,,,,GMS,
+ * @return {Array} - List of localization keys (unknown codes are prefixed with `*`)
+ */
+var getMorphLocalizationKeysGreekSR = exports.getMorphLocalizationKeysGreekSR = function getMorphLocalizationKeysGreekSR(morph) {
+  if (!morph || typeof morph !== 'string' || !morph.trim().length) {
+    return [];
+  }
+
+  var morphKeys = [];
+  // Will parsed out the morph string to its 12 places, the 1st being language,
+  // 2nd always empty, 3rd role, 4th type, and so on
+  var regex = /([A-Z0-9,.][a-z]*)/g; // Delimited by boundry of a comma, period, or uppercase letter
+  var codes = morph.match(regex).map(function (code) {
+    return [',', '.'].includes(code) ? null : code;
+  });
+  if (codes.length < 3) {
+    return morph;
+  }
+
+  var morpMapGrk = _morphCodeLocalizationMap.morphCodeLocalizationMapSrGrk;
+  if (morpMapGrk[2].hasOwnProperty(codes[2])) {
+    morphKeys.push(morpMapGrk[2][codes[2]].key);
+  } else {
+    morphKeys.push('*' + codes[2]); // no known localization key, so prefixing with '*'
+  }
+  if (codes[4]) {
+    var col2 = morpMapGrk[2];
+    var col2Form = col2[codes[2]];
+    if (col2.hasOwnProperty(codes[2]) && col2Form[4] && col2Form[4].hasOwnProperty(codes[4])) {
+      morphKeys.push(col2Form[4][codes[4]]);
+    } else {
+      morphKeys.push('*' + codes[4]);
+    } // unknown type, prefixing with '*'
+  }
+  codes.forEach(function (code, index) {
+    // 0 and 1  are ignored, already did 2 and 3 above
+    if (index < 5 || !code) {
+      return;
+    }
+    if (morpMapGrk[index].hasOwnProperty(code)) {
+      morphKeys.push(morpMapGrk[index][code]);
+    } else {
+      morphKeys.push('*' + code);
+    } // unknown code, prefixing with '*'
+  });
+  return morphKeys;
 };

package/lib/js/utils/morphCodeLocalizationMap.js

@@ -3,7 +3,7 @@
 Object.defineProperty(exports, "__esModule", {
   value: true
 });
-exports.morphCodeLocalizationMapGrk = exports.morphCodeLocalizationMapAr = exports.morphCodeLocalizationMapHeb = undefined;
+exports.morphCodeLocalizationMapGrk = exports.morphCodeLocalizationMapSrGrk = exports.morphCodeLocalizationMapAr = exports.morphCodeLocalizationMapHeb = undefined;
 
 var _lodash = require('lodash');
 
@@ -190,6 +190,97 @@ morphCodeLocalizationMapAr.verb_stems = {
   G: 'ittaphal'
 };
 
+// These reflect the columns on for the SR  on pages 9 and 10 of: https://greekcntr.org/resources/NTGRG.pdf
+var formVI = { // Form for VI
+  I: 'indicative',
+  M: 'imperative',
+  S: 'subjunctive',
+  O: 'optative',
+  N: 'infinitive',
+  P: 'participle'
+};
+
+var formNADX = { // Form for NADX
+  C: 'comparative',
+  S: 'superlatives',
+  D: 'diminutive',
+  I: 'indeclinable'
+};
+
+var morphCodeLocalizationMapSrGrk = exports.morphCodeLocalizationMapSrGrk = {
+  2: { // Function
+    N: {
+      key: 'noun',
+      4: formNADX
+    },
+    R: {
+      key: 'pronoun'
+    },
+    A: {
+      key: 'adjective',
+      4: formNADX
+    },
+    V: {
+      key: 'verb',
+      4: formVI
+    },
+    D: {
+      key: 'adverb',
+      4: formNADX
+    },
+    P: {
+      key: 'preposition'
+    },
+    C: {
+      key: 'conjunction'
+    },
+    I: {
+      key: 'interjection',
+      4: formVI
+    },
+    X: {
+      key: 'determiner', // new
+      4: formNADX
+    }
+  },
+  5: { // Tense
+    P: 'present',
+    I: 'imperfect',
+    F: 'future',
+    A: 'aorist',
+    E: 'perfect',
+    L: 'pluperfect'
+  },
+  6: { // Voice
+    A: 'active',
+    M: 'middle',
+    P: 'passive'
+  },
+  7: { // Person
+    1: 'first',
+    2: 'second',
+    3: 'third'
+  },
+  8: { // Case
+    N: 'nominative',
+    G: 'genitive',
+    D: 'dative',
+    A: 'accusative',
+    V: 'vocative'
+  },
+  9: { // Gender
+    M: 'masculine',
+    F: 'feminine',
+    N: 'neuter',
+    A: 'any' // new
+  },
+  10: { // Number
+    S: 'singular',
+    P: 'plural',
+    A: 'any' // new
+  }
+};
+
 // These reflect the columns on page 55 of https://greekcntr.org/downloads/project.pdf
 // This helps us translate codes starting and the 3rd place (the 2nd index) of a morph string
 // The numbered keys are the index of that code in the string, where the letter index is the code

package/lib/js/utils/verseObjects.js

@@ -528,17 +528,29 @@ var getWordListFromVerseObjectArray = exports.getWordListFromVerseObjectArray =
   return wordList;
 };
 
-var addContentAttributeToChildren = function addContentAttributeToChildren(childrens, parentObject, grandParentObject) {
+/**
+ * maps original language content to each child and flattens them into array, recursive processing
+ * @param {Array} childrens - list to process
+ * @param {Array} ancestors - ordered list of all original language ancestors
+ * @return {[]} returns flat array of all children
+ */
+var addContentAttributeToChildren = function addContentAttributeToChildren(childrens, ancestors) {
   var childrensWithAttribute = [];
 
-  for (var i = 0; i < childrens.length; i++) {
+  for (var i = 0, lc = childrens.length; i < lc; i++) {
     var child = childrens[i];
     if (child.children) {
-      child = addContentAttributeToChildren(child.children, child, parentObject);
-    } else if (!child.content && parentObject.content) {
-      var childrenContent = [parentObject];
-      if (grandParentObject) childrenContent.push(grandParentObject);
-      child.content = childrenContent;
+      child = addContentAttributeToChildren(child.children, [child].concat((0, _toConsumableArray3.default)(ancestors)));
+    } else if (ancestors[0].content) {
+      if (!child.content) {
+        child.content = [];
+      }
+      for (var j = 0, la = ancestors.length; j < la; j++) {
+        var ancestor = ancestors[j];
+        if (ancestor.content) {
+          child.content.push(ancestor);
+        }
+      }
     }
     childrensWithAttribute.push(child);
   }
@@ -552,7 +564,7 @@ var addContentAttributeToChildren = function addContentAttributeToChildren(child
  * @param {array} words - output array that will be filled with flattened verseObjects
  */
 var flattenVerseObjects = function flattenVerseObjects(verse, words) {
-  for (var i = 0; i < verse.length; i++) {
+  for (var i = 0, l = verse.length; i < l; i++) {
     var object = verse[i];
     if (object) {
       if (object.type === 'word') {
@@ -561,7 +573,7 @@ var flattenVerseObjects = function flattenVerseObjects(verse, words) {
       } else if (object.type === 'milestone') {
         // get children of milestone
         // add content attibute to children
-        var newObject = addContentAttributeToChildren(object.children, object);
+        var newObject = addContentAttributeToChildren(object.children, [object]);
         flattenVerseObjects(newObject, words);
       } else {
         words.push(object);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "word-aligner",
-  "version": "0.4.0",
+  "version": "1.0.1-alpha",
   "description": "A library for handling word alignment",
   "main": "lib/index.js",
   "scripts": {
@@ -51,6 +51,7 @@
     "eslint": "^5.12.1",
     "eslint-config-google": "^0.12.0",
     "eslint-plugin-jest": "^22.1.3",
+    "json-stringify-safe": "5.0.1",
     "jest": "^23.6.0",
     "ospath": "1.2.2",
     "usfm-js": "2.1.0"

package/src/js/aligner.js

@@ -29,14 +29,14 @@ export const hasAlignments = (alignments) => {
  * @return {Array} - sorted array of verseObjects to be used for verseText of targetLanguage
  */
 export const merge = (alignments, wordBank, verseString,
-    useVerseText = false) => {
+  useVerseText = false) => {
   // get the definitive list of verseObjects from the verse, unaligned but in order
   const {newVerseObjects: unalignedOrdered, wordMap} =
         VerseObjectUtils.getOrderedVerseObjectsFromString(verseString);
   // assign verseObjects with unaligned objects to be replaced with aligned ones
   // check each word in the verse string is also in the word bank or alignments
   const verseObjectsNotInAlignmentData = verseStringWordsContainedInAlignments(
-      alignments, wordBank, wordMap);
+    alignments, wordBank, wordMap);
   if (verseObjectsNotInAlignmentData.length > 0) {
     if (hasAlignments(alignments)) { // if verse has some alignments
       const verseWordsJoined = verseObjectsNotInAlignmentData.map(({text}) => text).join(', ');
@@ -53,9 +53,9 @@ export const merge = (alignments, wordBank, verseString,
   for (let i = 0; i < wbLen; i++) {
     const bottomWord = wordBank[i];
     const verseObject = VerseObjectUtils.wordVerseObjectFromBottomWord(
-        bottomWord);
+      bottomWord);
     const index = VerseObjectUtils.indexOfVerseObject(
-        wordMap, verseObject);
+      wordMap, verseObject);
     if (index > -1) {
       const location = wordMap[index];
       location.array[location.pos] = verseObject;
@@ -132,7 +132,7 @@ export const merge = (alignments, wordBank, verseString,
  *                      the given alignments
  */
 export function verseStringWordsContainedInAlignments(
-    alignments, wordBank, wordMap) {
+  alignments, wordBank, wordMap) {
   const unalignedMap = wordMap.filter((wordItem) => {
     const verseObject = wordItem.array[wordItem.pos];
     const checkIfWordMatches = function(verseObject) {
@@ -251,7 +251,7 @@ export const orderAlignments = function(alignmentVerse, alignmentUnOrdered) {
   let orderedObjects = null;
   if (typeof alignmentVerse === 'string') {
     orderedObjects = VerseObjectUtils.getOrderedVerseObjectsFromString(
-        alignmentVerse);
+      alignmentVerse);
   } else {
     orderedObjects = VerseObjectUtils.getOrderedVerseObjects(alignmentVerse);
   }
@@ -288,7 +288,7 @@ export const orderAlignments = function(alignmentVerse, alignmentUnOrdered) {
         }
         if (index < 0) { // if still not found in topWords, it's an unaligned topWord
           const wordObject = VerseObjectUtils.alignmentObjectFromVerseObject(
-              nextWord);
+            nextWord);
           alignment.push({topWords: [wordObject], bottomWords: []});
         }
       }
@@ -309,7 +309,7 @@ export const orderAlignments = function(alignmentVerse, alignmentUnOrdered) {
 export const addVerseObjectToAlignment = (verseObject, alignment) => {
   if (verseObject.type === 'milestone' && verseObject.children.length > 0) {
     const wordObject = VerseObjectUtils.alignmentObjectFromVerseObject(
-        verseObject);
+      verseObject);
     const duplicate = alignment.topWords.find(function(obj) {
       return (obj.word === wordObject.word) &&
         (obj.occurrence === wordObject.occurrence);
@@ -322,7 +322,7 @@ export const addVerseObjectToAlignment = (verseObject, alignment) => {
     });
   } else if (verseObject.type === 'word' && !verseObject.children) {
     const wordObject = VerseObjectUtils.alignmentObjectFromVerseObject(
-        verseObject);
+      verseObject);
     alignment.bottomWords.push(wordObject);
   }
 };
@@ -457,7 +457,7 @@ export const generateWordBank = (verseData) => {
  * @return {{alignments, wordBank}} - Reset alignments data
  */
 export const getBlankAlignmentDataForVerse = (
-    ugntVerse, targetLanguageVerse) => {
+  ugntVerse, targetLanguageVerse) => {
   const alignments = generateBlankAlignments(ugntVerse);
   const wordBank = generateWordBank(targetLanguageVerse);
   return {alignments, wordBank};

package/src/js/utils/MorphUtils.js

@@ -1,5 +1,10 @@
 /* eslint-disable no-use-before-define */
-import {morphCodeLocalizationMapGrk, morphCodeLocalizationMapAr, morphCodeLocalizationMapHeb} from './morphCodeLocalizationMap';
+import {
+  morphCodeLocalizationMapGrk,
+  morphCodeLocalizationMapAr,
+  morphCodeLocalizationMapHeb,
+  morphCodeLocalizationMapSrGrk
+} from './morphCodeLocalizationMap';
 
 /**
  * @description - Get a list of all the localization keys for a morph string in Greek
@@ -17,6 +22,9 @@ export const getMorphLocalizationKeys = (morph) => {
 
     case 'gr':
     default:
+      if (morph && morph.length === 12) {
+        return getMorphLocalizationKeysGreekSR(morph);
+      }
       return getMorphLocalizationKeysGreek(morph);
   }
 };
@@ -115,3 +123,51 @@ export const getMorphLocalizationKeysGreek = (morph) => {
   });
   return morphKeys;
 };
+/**
+ * @description - Get a list of all the localization keys for a morph string in Greek
+ * @param {String} morph - the morph string, e.g. Gr,N,,,,,GMS,
+ * @return {Array} - List of localization keys (unknown codes are prefixed with `*`)
+ */
+export const getMorphLocalizationKeysGreekSR = (morph) => {
+  if (!morph || typeof morph !== 'string' || !morph.trim().length) {
+    return [];
+  }
+
+  const morphKeys = [];
+  // Will parsed out the morph string to its 12 places, the 1st being language,
+  // 2nd always empty, 3rd role, 4th type, and so on
+  const regex = /([A-Z0-9,.][a-z]*)/g; // Delimited by boundry of a comma, period, or uppercase letter
+  const codes = morph.match(regex).map((code) => [',', '.'].includes(code) ? null : code);
+  if (codes.length < 3) {
+    return morph;
+  }
+
+  const morpMapGrk = morphCodeLocalizationMapSrGrk;
+  if (morpMapGrk[2].hasOwnProperty(codes[2])) {
+    morphKeys.push(morpMapGrk[2][codes[2]].key);
+  } else {
+    morphKeys.push('*' + codes[2]); // no known localization key, so prefixing with '*'
+  }
+  if (codes[4]) {
+    const col2 = morpMapGrk[2];
+    const col2Form = col2[codes[2]];
+    if (col2.hasOwnProperty(codes[2]) && col2Form[4] && col2Form[4].hasOwnProperty(codes[4])) {
+      morphKeys.push(col2Form[4][codes[4]]);
+    } else {
+      morphKeys.push('*' + codes[4]);
+    } // unknown type, prefixing with '*'
+  }
+  codes.forEach((code, index) => {
+    // 0 and 1  are ignored, already did 2 and 3 above
+    if (index < 5 || !code) {
+      return;
+    }
+    if (morpMapGrk[index].hasOwnProperty(code)) {
+      morphKeys.push(morpMapGrk[index][code]);
+    } else {
+      morphKeys.push('*' + code);
+    } // unknown code, prefixing with '*'
+  });
+  return morphKeys;
+};
+

package/src/js/utils/morphCodeLocalizationMap.js

@@ -183,6 +183,97 @@ morphCodeLocalizationMapAr.verb_stems = {
   G: 'ittaphal',
 };
 
+// These reflect the columns on for the SR  on pages 9 and 10 of: https://greekcntr.org/resources/NTGRG.pdf
+const formVI = { // Form for VI
+  I: 'indicative',
+  M: 'imperative',
+  S: 'subjunctive',
+  O: 'optative',
+  N: 'infinitive',
+  P: 'participle',
+};
+
+const formNADX = { // Form for NADX
+  C: 'comparative',
+  S: 'superlatives',
+  D: 'diminutive',
+  I: 'indeclinable',
+};
+
+export const morphCodeLocalizationMapSrGrk = {
+  2: { // Function
+    N: {
+      key: 'noun',
+      4: formNADX,
+    },
+    R: {
+      key: 'pronoun',
+    },
+    A: {
+      key: 'adjective',
+      4: formNADX,
+    },
+    V: {
+      key: 'verb',
+      4: formVI,
+    },
+    D: {
+      key: 'adverb',
+      4: formNADX,
+    },
+    P: {
+      key: 'preposition',
+    },
+    C: {
+      key: 'conjunction',
+    },
+    I: {
+      key: 'interjection',
+      4: formVI,
+    },
+    X: {
+      key: 'determiner', // new
+      4: formNADX,
+    },
+  },
+  5: { // Tense
+    P: 'present',
+    I: 'imperfect',
+    F: 'future',
+    A: 'aorist',
+    E: 'perfect',
+    L: 'pluperfect',
+  },
+  6: { // Voice
+    A: 'active',
+    M: 'middle',
+    P: 'passive',
+  },
+  7: { // Person
+    1: 'first',
+    2: 'second',
+    3: 'third',
+  },
+  8: { // Case
+    N: 'nominative',
+    G: 'genitive',
+    D: 'dative',
+    A: 'accusative',
+    V: 'vocative',
+  },
+  9: { // Gender
+    M: 'masculine',
+    F: 'feminine',
+    N: 'neuter',
+    A: 'any', // new
+  },
+  10: { // Number
+    S: 'singular',
+    P: 'plural',
+    A: 'any', // new
+  },
+};
+
 // These reflect the columns on page 55 of https://greekcntr.org/downloads/project.pdf
 // This helps us translate codes starting and the 3rd place (the 2nd index) of a morph string
 // The numbered keys are the index of that code in the string, where the letter index is the code

package/src/js/utils/verseObjects.js

@@ -428,17 +428,29 @@ export const getWordListFromVerseObjectArray = (verseObjects) => {
   return wordList;
 };
 
-const addContentAttributeToChildren = (childrens, parentObject, grandParentObject) => {
+/**
+ * maps original language content to each child and flattens them into array, recursive processing
+ * @param {Array} childrens - list to process
+ * @param {Array} ancestors - ordered list of all original language ancestors
+ * @return {[]} returns flat array of all children
+ */
+const addContentAttributeToChildren = (childrens, ancestors) => {
   const childrensWithAttribute = [];
 
-  for (let i = 0; i < childrens.length; i++) {
+  for (let i = 0, lc = childrens.length; i < lc; i++) {
     let child = childrens[i];
     if (child.children) {
-      child = addContentAttributeToChildren(child.children, child, parentObject);
-    } else if (!child.content && parentObject.content) {
-      const childrenContent = [parentObject];
-      if (grandParentObject) childrenContent.push(grandParentObject);
-      child.content = childrenContent;
+      child = addContentAttributeToChildren(child.children, [child, ...ancestors]);
+    } else if (ancestors[0].content) {
+      if (!child.content) {
+        child.content = [];
+      }
+      for (let j = 0, la = ancestors.length; j < la; j++) {
+        const ancestor = ancestors[j];
+        if (ancestor.content) {
+          child.content.push(ancestor);
+        }
+      }
     }
     childrensWithAttribute.push(child);
   }
@@ -452,7 +464,7 @@ const addContentAttributeToChildren = (childrens, parentObject, grandParentObjec
  * @param {array} words - output array that will be filled with flattened verseObjects
  */
 const flattenVerseObjects = (verse, words) => {
-  for (let i = 0; i < verse.length; i++) {
+  for (let i = 0, l = verse.length; i < l; i++) {
     const object = verse[i];
     if (object) {
       if (object.type === 'word') {
@@ -460,8 +472,7 @@ const flattenVerseObjects = (verse, words) => {
         words.push(object);
       } else if (object.type === 'milestone') { // get children of milestone
         // add content attibute to children
-        const newObject = addContentAttributeToChildren(object.children,
-          object);
+        const newObject = addContentAttributeToChildren(object.children, [object]);
         flattenVerseObjects(newObject, words);
       } else {
         words.push(object);

package/.idea/encodings.xml

@@ -1,4 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<project version="4">
-  <component name="Encoding" addBOMForNewFiles="with NO BOM" />
-</project>