From a30c1c1044ab6a22769b41c84cc842e0be860dce Mon Sep 17 00:00:00 2001
From: Bob Mottram <bob@freedombone.net>
Date: Sat, 21 Nov 2020 13:45:37 +0000
Subject: [PATCH] More efficient reading of tag index
---
webapp_hashtagswarm.py | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/webapp_hashtagswarm.py b/webapp_hashtagswarm.py
index 552f8045..e26a8bd3 100644
--- a/webapp_hashtagswarm.py
+++ b/webapp_hashtagswarm.py
@@ -76,6 +76,7 @@ def htmlHashTagSwarm(baseDir: str, actor: str, translate: {}) -> str:
currTime = datetime.utcnow()
daysSinceEpoch = (currTime - datetime(1970, 1, 1)).days
daysSinceEpochStr = str(daysSinceEpoch) + ' '
+ daysSinceEpochStr2 = str(daysSinceEpoch - 1) + ' '
recently = daysSinceEpoch - 1
tagSwarm = []
domainHistogram = {}
@@ -99,8 +100,12 @@ def htmlHashTagSwarm(baseDir: str, actor: str, translate: {}) -> str:
hashTagName = f.split('.')[0]
if isBlockedHashtag(baseDir, hashTagName):
continue
- if daysSinceEpochStr not in open(tagsFilename).read():
- continue
+ with open(tagsFilename, 'r') as fp:
+ # only read one line, which saves time and memory
+ lastTag = fp.readline()
+ if not lastTag.startswith(daysSinceEpochStr):
+ if not lastTag.startswith(daysSinceEpochStr2):
+ continue
with open(tagsFilename, 'r') as tagsFile:
while True:
line = tagsFile.readline()