From b32e3dae69dfe4bfc7847947ecfd8a8972885973 Mon Sep 17 00:00:00 2001
From: Harshit Hajela <hhajela@splunk.com>
Date: Wed, 28 Apr 2021 14:30:51 -0600
Subject: [PATCH] make string matching more lenient

---
 backend/companies.json | 26 +++++++++++++-------------
 backend/updater.py     |  7 ++++++-
 2 files changed, 19 insertions(+), 14 deletions(-)

diff --git a/backend/companies.json b/backend/companies.json
index eaf617ce7..12457d9b6 100644
--- a/backend/companies.json
+++ b/backend/companies.json
@@ -1,32 +1,32 @@
 {
-  "MMM": "3M Company",
+  "MMM": "3M",
   "AXP": "American Express",
   "AMGN": "Amgen",
-  "AAPL": "Apple Inc.",
+  "AAPL": "Apple",
   "BA": "Boeing",
-  "CAT": "Caterpillar Inc.",
-  "CVX": "Chevron Corporation",
-  "CSCO": "Cisco Systems",
-  "KO": "The Coca-Cola Company",
-  "DOW": "Dow Inc.",
+  "CAT": "Caterpillar",
+  "CVX": "Chevron",
+  "CSCO": "Cisco",
+  "KO": "Coca-Cola",
+  "DOW": "Dow",
   "GS": "Goldman Sachs",
-  "HD": "The Home Depot",
+  "HD": "Home Depot",
   "HON": "Honeywell",
   "IBM": "IBM",
   "INTC": "Intel",
   "JNJ": "Johnson & Johnson",
-  "JPM": "JPMorgan Chase",
+  "JPM": "JPMorgan",
   "MCD": "McDonald's",
-  "MRK": "Merck & Co.",
+  "MRK": "Merck",
   "MSFT": "Microsoft",
   "NKE": "Nike",
   "PG": "Procter & Gamble",
   "CRM": "Salesforce",
   "TRV": "The Travelers Companies",
-  "UNH": "UnitedHealth Group",
+  "UNH": "UnitedHealth",
   "VZ": "Verizon",
-  "V": "Visa Inc.",
+  "V": "Visa",
   "WBA": "Walgreens Boots Alliance",
   "WMT": "Walmart",
-  "DIS": "The Walt Disney Company"
+  "DIS": "Walt Disney"
 }
diff --git a/backend/updater.py b/backend/updater.py
index 9509ca3b6..0815e3110 100644
--- a/backend/updater.py
+++ b/backend/updater.py
@@ -40,8 +40,13 @@ def join(self):
 
     def fetchComments(self):
         for comment in self.sr_obj.stream.comments(skip_existing=True, pause_after=5):
+            comment_text = comment.body.casefold()
             for ticker in self.companies:
-                if ticker in comment.body or self.companies[ticker] in comment.body:
+                casefolded_company = self.companies[ticker].casefold()
+                if ('{0} '.format(ticker) in comment.body or
+                        ' {0}'.format(ticker) in comment.body or
+                        '{0} '.format(casefolded_company) in comment_text or
+                        ' {0}'.format(casefolded_company) in comment_text):
                     comment_obj = { "ticker": ticker, "text": comment.body, "timestamp": math.ceil(time.time_ns()/1000000) }
                     self.output(comment_obj)
                     break