A trivial bit of error handling

bookbyline.py

@@ -9,14 +9,13 @@
 import sys
 import os
 import sqlite3
-import re
 import tweepy
 sys.path.append("/Users/sth/scripts")
+
 # tweepy stuff
 auth=tweepy.BasicAuthHandler('robo_dante', 'beatrice')
 api=tweepy.API(auth)
-
-
+# create a SQLite connection, or create a new db and table
 connection=sqlite3.connect('dc.db')
 cursor=connection.cursor()
 try:
@@ -28,6 +27,11 @@
 	cursor.execute('INSERT INTO position VALUES (null, ?, ?, ?)',(lastline, "1", 0))
 	# set up a new blank table, and start off at line 0
 	# retry logic goes here
+	try:
+		cursor.execute('SELECT * FROM position ORDER BY POSITION DESC LIMIT 1')
+	except sqlite3.OperationalError:
+		print "Still couldn't execute the SQL query, even though I created a new table. Giving up."
+		sys.exit()
 
 # get the highest page number, and the line display offset
 row=cursor.fetchone()
@@ -41,12 +45,11 @@
 
 def format_tweet(input_string,next_string):
 	""" Properly format an input string based on whether it's a header line, or a poetry line
-	accepts 3 inputs: the current line from a book, the following line, and the current line display
-	off_set, which is incremented each time a header line is encountered, and subtracted from the
-	"real" line number, in order to display the correct line number.
-	returns a ready-to-tweet string, either a canto, or a poetry line. """
-
-	# If line is a new Canto, append the following line to it, instead of "l. "
+	accepts 2 inputs: the current line from a book, and the following line. If the current line
+	begins with "CANTO", it's a header, so instead of displaying a line number, we join the next line
+	and increment both the line number and the line offset by 1. This means the line numbers
+	don't jump, when a header is encountered, as the offset is subtracted from the display line.
+	Returns a ready-to-tweet string, either a canto, or a poetry line. """
 	#pattern='^CANTO'
 	#if re.search(pattern, input_string):
 	if input_string.startswith("CANTO"):