From 5c46786ebc0abf14a64aef482455f504a1cd6e55 Mon Sep 17 00:00:00 2001
From: Nazar Gargol <nazargargol@gmail.com>
Date: Fri, 19 Jun 2020 18:24:51 +1200
Subject: [PATCH] Colocated readCSV util with members input serializer

refs 8185b42d9e5fd9e9051f08ce3395a648ec02f3a4

- Allows to move files and keep the history in one go.
- 'csv-parser' will be upgraded to 'papaparse' lib in a new package
---
 ghost/members-csv/parse              | 60 +++++++++++++++++++++++++++-
 ghost/members-csv/test/parse.test.js | 53 ++++++++++++++++++++++++
 2 files changed, 111 insertions(+), 2 deletions(-)
 create mode 100644 ghost/members-csv/test/parse.test.js

diff --git a/ghost/members-csv/parse b/ghost/members-csv/parse
index 54e77c4072..202705a866 100644
--- a/ghost/members-csv/parse
+++ b/ghost/members-csv/parse
@@ -1,4 +1,59 @@
-const fsLib = require('../../../../../../lib/fs');
+const Promise = require('bluebird');
+const csvParser = require('csv-parser');
+const _ = require('lodash');
+const fs = require('fs-extra');
+
+const readCSV = (options) => {
+    const columnsToExtract = options.columnsToExtract || [];
+    let results = [];
+    const rows = [];
+
+    return new Promise(function (resolve, reject) {
+        const readFile = fs.createReadStream(options.path);
+
+        readFile.on('err', function (err) {
+            reject(err);
+        })
+            .pipe(csvParser())
+            .on('data', function (row) {
+                rows.push(row);
+            })
+            .on('end', function () {
+            // If CSV is single column - return all values including header
+                const headers = _.keys(rows[0]);
+
+                let result = {};
+                const columnMap = {};
+                if (columnsToExtract.length === 1 && headers.length === 1) {
+                    results = _.map(rows, function (value) {
+                        result = {};
+                        result[columnsToExtract[0].name] = value[headers[0]];
+                        return result;
+                    });
+                } else {
+                // If there are multiple columns in csv file
+                // try to match headers using lookup value
+
+                    _.map(columnsToExtract, function findMatches(column) {
+                        _.each(headers, function checkheader(header) {
+                            if (column.lookup.test(header)) {
+                                columnMap[column.name] = header;
+                            }
+                        });
+                    });
+
+                    results = _.map(rows, function evaluateRow(row) {
+                        const result = {};
+                        _.each(columnMap, function returnMatches(value, key) {
+                            result[key] = row[value];
+                        });
+                        return result;
+                    });
+                }
+                resolve(results);
+            });
+    });
+};
 
 const parse = async (filePath) => {
     const columnsToExtract = [{
@@ -27,10 +82,11 @@ const parse = async (filePath) => {
         lookup: /created_at/i
     }];
 
-    return await fsLib.readCSV({
+    return await readCSV({
         path: filePath,
         columnsToExtract: columnsToExtract
     });
 };
 
 module.exports.parse = parse;
+module.exports.readCSV = readCSV;
diff --git a/ghost/members-csv/test/parse.test.js b/ghost/members-csv/test/parse.test.js
new file mode 100644
index 0000000000..19bbc184c8
--- /dev/null
+++ b/ghost/members-csv/test/parse.test.js
@@ -0,0 +1,53 @@
+const should = require('should');
+const path = require('path');
+const fsLib = require('../../../../../../../../core/server/api/canary/utils/serializers/input/utils/members-import-csv');
+
+const csvPath = path.join(__dirname, '../../../../../../../utils/fixtures/csv/');
+
+describe('members-import-csv: read csv', function () {
+    it('read csv: one column', function (done) {
+        fsLib.readCSV({
+            path: csvPath + 'single-column-with-header.csv',
+            columnsToExtract: [{name: 'email', lookup: /email/i}]
+        }).then(function (result) {
+            should.exist(result);
+            result.length.should.eql(2);
+            result[0].email.should.eql('jbloggs@example.com');
+            result[1].email.should.eql('test@example.com');
+            done();
+        }).catch(done);
+    });
+
+    it('read csv: two columns, 1 filter', function (done) {
+        fsLib.readCSV({
+            path: csvPath + 'two-columns-with-header.csv',
+            columnsToExtract: [{name: 'email', lookup: /email/i}]
+        }).then(function (result) {
+            should.exist(result);
+            result.length.should.eql(2);
+            result[0].email.should.eql('jbloggs@example.com');
+            result[1].email.should.eql('test@example.com');
+            should.not.exist(result[0].id);
+
+            done();
+        }).catch(done);
+    });
+
+    it('read csv: two columns, 2 filters', function (done) {
+        fsLib.readCSV({
+            path: csvPath + 'two-columns-obscure-header.csv',
+            columnsToExtract: [
+                {name: 'email', lookup: /email/i},
+                {name: 'id', lookup: /id/i}
+            ]
+        }).then(function (result) {
+            should.exist(result);
+            result.length.should.eql(2);
+            result[0].email.should.eql('jbloggs@example.com');
+            result[0].id.should.eql('1');
+            result[1].email.should.eql('test@example.com');
+            result[1].id.should.eql('2');
+            done();
+        }).catch(done);
+    });
+});