[ML] Fixing missing final new line character issue (#109274) (#110050)

* [ML] Fixing missing final new line character issue

* adding tests

* tiny refactor

* test fixes based on review

Co-authored-by: Kibana Machine <42973632+kibanamachine@users.noreply.github.com>

Co-authored-by: James Gowdy <jgowdy@elastic.co>

x-pack/plugins/file_upload/public/importer/importer.ts

@@ -40,7 +40,10 @@ export abstract class Importer implements IImporter {
     let remainder = 0;
     for (let i = 0; i < parts; i++) {
       const byteArray = decoder.decode(data.slice(i * size - remainder, (i + 1) * size));
-      const { success, docs, remainder: tempRemainder } = this._createDocs(byteArray);
+      const { success, docs, remainder: tempRemainder } = this._createDocs(
+        byteArray,
+        i === parts - 1
+      );
       if (success) {
         this._docArray = this._docArray.concat(docs);
         remainder = tempRemainder;
@@ -52,7 +55,7 @@ export abstract class Importer implements IImporter {
     return { success: true };
   }
 
-  protected abstract _createDocs(t: string): CreateDocsResponse;
+  protected abstract _createDocs(t: string, isLastPart: boolean): CreateDocsResponse;
 
   public async initializeImport(
     index: string,

x-pack/plugins/file_upload/public/importer/message_importer.ts

@@ -30,7 +30,7 @@ export class MessageImporter extends Importer {
   // multiline_start_pattern regex
   // if it does, it is a legitimate end of line and can be pushed into the list,
   // if not, it must be a newline char inside a field value, so keep looking.
-  protected _createDocs(text: string): CreateDocsResponse {
+  protected _createDocs(text: string, isLastPart: boolean): CreateDocsResponse {
     let remainder = 0;
     try {
       const docs: Doc[] = [];
@@ -39,9 +39,17 @@ export class MessageImporter extends Importer {
       let line = '';
       for (let i = 0; i < text.length; i++) {
         const char = text[i];
+        const isLastChar = i === text.length - 1;
         if (char === '\n') {
           message = this._processLine(docs, message, line);
           line = '';
+        } else if (isLastPart && isLastChar) {
+          // if this is the end of the last line and the last chunk of data,
+          // add the remainder as a final line.
+          // just in case the last line doesn't end in a new line char.
+          line += char;
+          message = this._processLine(docs, message, line);
+          line = '';
         } else {
           line += char;
         }

x-pack/plugins/file_upload/public/importer/ndjson_importer.ts

@@ -13,7 +13,7 @@ export class NdjsonImporter extends Importer {
     super();
   }
 
-  protected _createDocs(json: string): CreateDocsResponse {
+  protected _createDocs(json: string, isLastPart: boolean): CreateDocsResponse {
     let remainder = 0;
     try {
       const splitJson = json.split(/}\s*\n/);

x-pack/test/functional/apps/ml/data_visualizer/file_data_visualizer.ts

@@ -111,6 +111,7 @@ export default function ({ getService }: FtrProviderContext) {
         totalFieldsCount: 12,
         fieldTypeFiltersResultCount: 4,
         fieldNameFiltersResultCount: 1,
+        ingestedDocCount: 20,
       },
     },
     {
@@ -152,6 +153,51 @@ export default function ({ getService }: FtrProviderContext) {
         totalFieldsCount: 3,
         fieldTypeFiltersResultCount: 1,
         fieldNameFiltersResultCount: 1,
+        ingestedDocCount: 13,
+      },
+    },
+    {
+      suiteSuffix: 'with a file with a missing new line char at the end',
+      filePath: path.join(__dirname, 'files_to_import', 'missing_end_of_file_newline.csv'),
+      indexName: 'user-import_3',
+      createIndexPattern: false,
+      fieldTypeFilters: [],
+      fieldNameFilters: [],
+      expected: {
+        results: {
+          title: 'missing_end_of_file_newline.csv',
+          numberOfFields: 3,
+        },
+        metricFields: [
+          {
+            fieldName: 'value',
+            type: ML_JOB_FIELD_TYPES.NUMBER,
+            docCountFormatted: '3 (100%)',
+            exampleCount: 3,
+            topValuesCount: 3,
+          },
+        ],
+        nonMetricFields: [
+          {
+            fieldName: 'title',
+            type: ML_JOB_FIELD_TYPES.UNKNOWN,
+            docCountFormatted: '3 (100%)',
+            exampleCount: 3,
+          },
+          {
+            fieldName: 'description',
+            type: ML_JOB_FIELD_TYPES.KEYWORD,
+            docCountFormatted: '3 (100%)',
+            exampleCount: 3,
+          },
+        ],
+        visibleMetricFieldsCount: 0,
+        totalMetricFieldsCount: 0,
+        populatedFieldsCount: 3,
+        totalFieldsCount: 3,
+        fieldTypeFiltersResultCount: 3,
+        fieldNameFiltersResultCount: 3,
+        ingestedDocCount: 3,
       },
     },
   ];
@@ -271,6 +317,10 @@ export default function ({ getService }: FtrProviderContext) {
           await ml.testExecution.logTestStep('imports the file');
           await ml.dataVisualizerFileBased.startImportAndWaitForProcessing();
 
+          await ml.dataVisualizerFileBased.assertIngestedDocCount(
+            testData.expected.ingestedDocCount
+          );
+
           await ml.testExecution.logTestStep('creates filebeat config');
           await ml.dataVisualizerFileBased.selectCreateFilebeatConfig();
 

x-pack/test/functional/apps/ml/data_visualizer/files_to_import/geo_file.csv

@@ -11,4 +11,4 @@ POINT (-2.509384 51.40959),On or near Barnard Walk,
 POINT (-2.495055 51.422132),On or near Cross Street,
 POINT (-2.509384 51.40959),On or near Barnard Walk,
 POINT (-2.495055 51.422132),On or near Cross Street,
-POINT (-2.509126 51.416137),On or near St Francis Road,
+POINT (-2.509126 51.416137),On or near St Francis Road,

x-pack/test/functional/apps/ml/data_visualizer/files_to_import/missing_end_of_file_newline.csv

@@ -0,0 +1,4 @@
+title,description,value
+first title,this is the first description,22
+second title,this is the second description,66
+third title,this is the third description,88

x-pack/test/functional/services/ml/common_ui.ts

@@ -285,6 +285,20 @@ export function MachineLearningCommonUIProvider({
       await this.assertRowsNumberPerPage(testSubj, rowsNumber);
     },
 
+    async getEuiDescriptionListDescriptionFromTitle(testSubj: string, title: string) {
+      const subj = await testSubjects.find(testSubj);
+      const titles = await subj.findAllByTagName('dt');
+      const descriptions = await subj.findAllByTagName('dd');
+
+      for (let i = 0; i < titles.length; i++) {
+        const titleText = (await titles[i].parseDomContent()).html();
+        if (titleText === title) {
+          return (await descriptions[i].parseDomContent()).html();
+        }
+      }
+      return null;
+    },
+
     async changeToSpace(spaceId: string) {
       await PageObjects.spaceSelector.openSpacesNav();
       await PageObjects.spaceSelector.goToSpecificSpace(spaceId);

x-pack/test/functional/services/ml/data_visualizer_file_based.ts

@@ -132,6 +132,17 @@ export function MachineLearningDataVisualizerFileBasedProvider(
       });
     },
 
+    async assertIngestedDocCount(count: number) {
+      const docCount = await mlCommonUI.getEuiDescriptionListDescriptionFromTitle(
+        'dataVisualizerFileImportSuccessCallout',
+        'Documents ingested'
+      );
+      expect(docCount).to.eql(
+        count,
+        `Expected Documents ingested count to be '${count}' (got '${docCount}')`
+      );
+    },
+
     async selectCreateFilebeatConfig() {
       await testSubjects.scrollIntoView('fileDataVisFilebeatConfigLink', {
         bottomOffset: fixedFooterHeight,