fix anomolies in CSV, more efficient processing

[phrases.git] / phrases.py

diff --git a/phrases.py b/phrases.py
index 39d6f0c7e59be1d3101d120fb89b57b669e68c5d..d1d934f2e324d9bf0cabd2b834e22d1103af8f1a 100755 (executable)
--- a/phrases.py
+++ b/phrases.py
@@ -5,56 +5,50 @@
 import argparse
 import random
 import sys
-import re
+import csv
 
-# Create list of phrases from phrase file
-def read_phrases(phrase_file):
-    f = open(phrase_file, 'r')
-    contents = f.read()
-
-    lines = [line.rstrip() for line in contents.split('\n')]
-    delim = re.compile(r'^%$')
-    phrases = []
-    cur = []
-
-    for line in lines:
-        if delim.match(line):
-            phrase = '\n'.join(cur)
-            if phrase.strip():
-                phrases.append(phrase)
-            cur = []
-            continue
-        cur.append(line)
-
-    return phrases
-
-# Main program
 def main(args=sys.argv[1:]):
+    # Argument parsing
     parser = argparse.ArgumentParser(description="Latin famous phrases in the terminal.")
-    parser.add_argument("-e", "--english", action='store_true', help="Output English.")
-    parser.add_argument("-l", "--latin", action='store_true', help="Output Latin (default)")
-    parser.add_argument("-n", "--notes", action='store_true', help="Output notes on phrase")
-    parser.add_argument("-m", "--min", type=int, help="Set the minimum length of latin")
-    parser.add_argument("-M", "--max", type=int, help="Set the maximum length of latin")
+    parser.add_argument("-e", "--english", action='store_true', help="Print the English translation.")
+    parser.add_argument("-i", "--id", action='store_true', help="Print the id of the phrase.")
+    parser.add_argument("-l", "--latin", action='store_true', help="Print the Latin phrase (default)")
+    parser.add_argument("-m", "--min", default=0, type=int, help="Set the minimum length of the Latin phrase")
+    parser.add_argument("-M", "--max", default=10000000, type=int, help="Set the maximum length of Latin phrase")
+    parser.add_argument("-n", "--notes", action='store_true', help="Print any notes on phrase")
     args = parser.parse_args()
 
-    phrases = list(read_phrases("/usr/share/phrases/phrases"))
-    randomRecord = random.randint(0, len(phrases) - 1)
-    randphrase = phrases[randomRecord]
-
-    if args.min:
-        while True:
-            randomRecord = random.randint(0, len(phrases) - 1)
-            randphrase = phrases[randomRecord]
-            if args.min <= len(randphrase.split('\n')[0]) <= args.max:
-                break
-    if args.latin:
-        print(randphrase.split('\n')[0])
-    if args.english:
-        print(randphrase.split('\n')[1])
-    if args.notes:
-        print(randphrase.split('\n')[2])
-    if not len(sys.argv) > 1:
-        print(randphrase.split('\n')[0])
-
-main()
+    right_length = []
+
+    # Find phrases of the right size
+    with open('/usr/share/phrases/phrases.csv') as f:
+        reader = csv.reader(f)
+        all_lines = list(reader)
+        next(reader, None) # skip header
+        for row in all_lines:
+            try:
+                if args.max >= int(row[4]) >= args.min: # generate a shortlist of phrases of the right length
+                    right_length.append(row[0])
+            except:
+                pass # skip malformed rows
+
+        try:
+            chosen = int(right_length[random.randint(0, len(right_length) - 1)]) # choose a random id from the shortlist
+        except:
+            sys.exit("No phrase within the given parameters!")
+
+        # Output as specified in flags
+        if not (args.english or args.latin or args.notes):
+            print(all_lines[chosen][1])
+        else:
+            if args.id:
+                print(all_lines[chosen][1])
+            if args.latin:
+                print(all_lines[chosen][1])
+            if args.english:
+                print(all_lines[chosen][2])
+            if args.notes:
+                print(all_lines[chosen][3])
+
+if __name__ == "__main__":
+    main()