fix anomolies in CSV, more efficient processing

[phrases.git] / phrases.py

diff --git a/phrases.py b/phrases.py
index bcf3f7367128bfb9de2ba3f9f1e66af48f071eee..d1d934f2e324d9bf0cabd2b834e22d1103af8f1a 100755 (executable)
--- a/phrases.py
+++ b/phrases.py
@@ -2,57 +2,53 @@
 # Display Latin famous phrases in the terminal
 # Armaan Bhojwani 2020
 
+import argparse
 import random
-import os
 import sys
-import re
-
-# Use system random if available, otherwise, use Python's
-def _random_int(start, end):
-    try:
-        r = random.SystemRandom()
-    except:
-        r = random
-
-    return r.randint(start, end)
-
-# 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 = []
-
-    def save_if_nonempty(buf):
-        phrase = '\n'.join(buf)
-        if phrase.strip():
-            phrases.append(phrase)
-
-    for line in lines:
-        if delim.match(line):
-            save_if_nonempty(cur)
-            cur = []
-            continue
-
-        cur.append(line)
-
-    if cur:
-        save_if_nonempty(cur)
-
-    return phrases
-
-# Return a random phrase from the phrases list
-def get_random_phrase(phrase_file):
-    phrases = list(_read_phrases(phrase_file))
-    randomRecord = _random_int(0, len(phrases) - 1)
-    randphrase = phrases[randomRecord]
-    return randphrase.partition('\n')[0]
-
-# Main program
-def main():
-    print(get_random_phrase("/usr/share/phrases/phrases"))
-
-main()
+import csv
+
+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="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()
+
+    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()