Improve the Queen award (#156)

Fixed bugs:
 - Sometimes two people would get the same username
 - Sometimes a whole post would have the same prefix
 - Links would be broken
 - Greentext would be broken
 - strings of periods do not become "and, and, and, and,"

New stuff:
 - New girl catchphrases
 - New girl pfps
 - Catchphrases have a chance to not be applied (50%, but configurable)
 - A lot more regexes
 	- everyone => literally everyone (and others)
    - into => totally into
    - hello => hiiiiiiii (and others)
    - just => like just
    - i mean => i mean like
    - ! => !!!

Co-authored-by: Chuck Sneed <sneed@formerlychucks.net>
Reviewed-on: #156
Co-authored-by: HeyMoon <heymoon@noreply.fsdfsd.net>
Co-committed-by: HeyMoon <heymoon@noreply.fsdfsd.net>

files/classes/user.py

@@ -818,7 +818,7 @@ class User(Base):
 		if self.rainbow:
 			return f"{SITE_FULL}/e/marseysalutepride.webp"
 		if self.queen:
-			number_of_girl_pfps = 8
+			number_of_girl_pfps = 25
 			pic_num = (self.id % number_of_girl_pfps) + 1
 			return f"{SITE_FULL}/i/pfps/girls/{pic_num}.webp"
 		if self.profileurl and self.can_see_my_shit:

files/helpers/config/const.py

@@ -101,8 +101,11 @@ AJ_REPLACEMENTS = {
 	'EVERYBODY': 'EVERYPONY',
 }
 
+PHRASE_CHANCE = 0.5
 GIRL_PHRASES = [
 	"ok so $",
+	"um $",
+	"also like $",
 	"literally, $",
 	"i feel like $",
 	"my heart is telling me $",
@@ -112,8 +115,10 @@ GIRL_PHRASES = [
     "$ and thats the tea, sis",
     "$ but go off i guess",
     "$ but go off",
-    "$, karen",
-	"$ but its whatever"
+	"$ but its whatever",
+	"$ and its EVERYTHING",
+	"$ *sips tea*",
+	"$ PERIODT"
 ]
 GIRL_NAME_PREFIX = [
 	'the',

files/helpers/regex.py

@@ -69,9 +69,20 @@ torture_regex = re.compile('(^|\s)(i|me)($|\s)', flags=re.I|re.A)
 torture_regex2 = re.compile("(^|\s)(i'm)($|\s)", flags=re.I|re.A)
 torture_regex3 = re.compile("(^|\s)(my|mine)($|\s)", flags=re.I|re.A)
 
-sentence_ending_regex = re.compile('(\.)', flags=re.I|re.A)
+#matches ". ", does not match "..." or a.b
+sentence_ending_regex = re.compile('(?<!\.)(\.)(?=$|\n|\s)', flags=re.I|re.A)
 normal_punctuation_regex = re.compile('(\"|\')', flags=re.I|re.A)
 more_than_one_comma_regex = re.compile('\,\,+', flags=re.I|re.A)
+#matches the various superlatives, but only if it as the start or end of a string or if it surrounded by spaces or is at the end of a word.
+superlative_regex = re.compile('(?<=^|(?<=\s))(everyone|everybody|nobody|all|none|every|any|no one|anything)(?=$|\n|\s|[.?!,])', flags=re.I|re.A)
+#like above, except only when totally doesn't already prefix
+totally_regex = re.compile('(?<=^|(?<=\s))(?<!totally )(into)(?=$|\n|\s|[.?!,])', flags=re.I|re.A)
+greeting_regex = re.compile('(?<=^|(?<=\s))(hello|hi|hey|hecko)(?=$|\n|\s|[.?!,])', flags=re.I|re.A)
+like_before_regex = re.compile('(?<=^|(?<=\s))(?<!like )(just|only)(?=$|\n|\s|[.?!,])', flags=re.I|re.A)
+like_after_regex = re.compile('(?<=^|(?<=\s))(i mean)(?! like)(?=$|\n|\s|[.?!,])', flags=re.I|re.A)
+#match ! or ? but only if it isn't touching another ! or ?, or is in front of a letter
+single_repeatable_punctuation = re.compile('(?<!!|\?)(!|\?)(?!!|\?)(?=\s|$)', flags=re.I|re.A)
+initial_part_regex = re.compile('(?<=^)(>+)', flags=re.I|re.A)
 
 image_check_regex = re.compile(f'!\[\]\(((?!(https:\/\/({hosts})\/|\/)).*?)\)', flags=re.A)
 

files/helpers/sanitize.py

@@ -688,14 +688,27 @@ def torture_ap(string, username):
 
 def torture_queen(string, key):
 	if not string: return string
+	result = initial_part_regex.search(string)
+	initial = result.group(1) if result else ""
 	string = string.lower()
+	string = initial_part_regex.sub("", string)
 	string = sentence_ending_regex.sub(", and", string)
+	string = superlative_regex.sub(r"literally \g<1>", string)
+	string = totally_regex.sub(r"totally \g<1>", string)
+	string = single_repeatable_punctuation.sub(r"\g<1>\g<1>\g<1>", string)
+	string = greeting_regex.sub(r"hiiiiiiiiii", string)
+	string = like_after_regex.sub(r"\g<1> like", string)
+	string = like_before_regex.sub(r"like \g<1>", string)
 	string = normal_punctuation_regex.sub("", string)
 	string = more_than_one_comma_regex.sub(",", string)
 	if string[-5:] == ', and':
 		string = string[:-5]
-	girl_phrase = GIRL_PHRASES[key%len(GIRL_PHRASES)]
-	string = girl_phrase.replace("$", string)
+    
+	random.seed(key)
+	if random.random() < PHRASE_CHANCE:
+		girl_phrase = random.choice(GIRL_PHRASES)
+		string = girl_phrase.replace("$", string)
+	string = initial + string
 	return string
 
 def torture_object(obj, torture_method):
@@ -706,7 +719,7 @@ def torture_object(obj, torture_method):
 		i = 0
 		for tag in tags:
 			i+=1
-			key = obj.id*i
+			key = obj.id+i
 			tag.string.replace_with(torture_method(tag.string, key))
 		obj.body_html = str(soup).replace('<html><body>','').replace('</body></html>','')
 

files/routes/awards.py

@@ -329,12 +329,15 @@ def award_thing(v, thing_type, id):
 			abort(409, f"{safe_username} is under the effect of a conflicting award: OwOify award!")
 
 		if not author.queen:
-
-			adjective = GIRL_NAME_ADJECTIVE[author.id%20].capitalize()
-			noun = GIRL_NAME_NOUN[(int(author.id/20))%20].capitalize()
-			prefix = GIRL_NAME_PREFIX[(int((id/20)/20))%8]
+			number_of_adjectives = len(GIRL_NAME_ADJECTIVE)
+			number_of_nouns = len(GIRL_NAME_NOUN)
+			number_of_prefix = len(GIRL_NAME_PREFIX)
+			numbers = get_number_tuple(author.id, [number_of_adjectives, number_of_nouns, number_of_prefix])
+			adjective = GIRL_NAME_ADJECTIVE[numbers[0]].capitalize()
+			noun = GIRL_NAME_NOUN[numbers[1]].capitalize()
+			prefix = GIRL_NAME_PREFIX[numbers[2]]
 			prefix = prefix[0].upper() + prefix[1:]
-			number = int(((author.id/20)/20)/8)
+			number = numbers[3]
 
 			new_name = f"{prefix}{adjective}{noun}"+ (str(number) if number > 0 else "")
 
@@ -551,3 +554,18 @@ def award_thing(v, thing_type, id):
 
 	return {"message": f"{AWARDS[kind]['title']} award given to {thing_type} successfully!"}
 
+def shift_number_down(input, mod):
+    if input <= 0:
+        return 0, 0
+    number = (input%mod)
+    input -= number
+    input /= mod
+    return int(number), int(input)
+
+def get_number_tuple(input, mods):
+    results = []
+    for mod in mods:
+        result, input = shift_number_down(input, mod)
+        results.append(result)
+    results.append(input)
+    return results