0.7.26 - improved wordlists.

Author: finnbear

Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "rustrict"
 authors = ["Finn Bear"]
-version = "0.7.25"
+version = "0.7.26"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/finnbear/rustrict/"

README.md

@@ -132,8 +132,7 @@ If you want to add custom profanities or safe words, enable the `customize` feat
 }
 ```
 
-But wait, there's more! If your use-case is chat moderation, and you can store data on a per-user basis, you
-might benefit from the `context` feature.
+If your use-case is chat moderation, and you store data on a per-user basis, you can use `rustrict::Context` as a reference implementation:
 
 ```rust
 #[cfg(feature = "context")]
@@ -178,7 +177,7 @@ is used as a dataset. Positive accuracy is the percentage of profanity detected
 
 | Crate | Accuracy | Positive Accuracy | Negative Accuracy | Time |
 |-------|----------|-------------------|-------------------|------|
-| [rustrict](https://crates.io/crates/rustrict) | 79.74%   | 94.00%            | 76.18%            | 9s   |
+| [rustrict](https://crates.io/crates/rustrict) | 79.74%   | 94.00%            | 76.19%            | 9s   |
 | [censor](https://crates.io/crates/censor) | 76.16%   | 72.76%            | 77.01%            | 23s  |
 
 ## Development

src/character_analyzer.rs

@@ -49,6 +49,7 @@ fn main() {
                 '🐿' => 20,
                 '𒐫' => 40,
                 '𒈙' => 35,
+                '༺' | '༻' => 25,
                 _ => {
                     let max_width = (max_width(c, &fonts) as f32 / 100f32).round() as u16;
                     if max_width > u8::MAX as u16 {

src/character_widths.bin

@@ -8,6 +8,10 @@ use std::time::{Duration, Instant};
 
 /// Context is useful for taking moderation actions on a per-user basis i.e. each user would get
 /// their own Context.
+/// 
+/// # Recommendation
+/// 
+/// Use this as a reference implementation e.g. by copying and adapting it.
 #[derive(Clone)]
 #[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
 #[cfg_attr(doc, doc(cfg(feature = "context")))]

src/context.rs

@@ -1,16 +1,21 @@
 #8
 # of
+(until
 2 secs
 3 secs
 4 secs
+45s
 5 secs
 6 secs
 7 secs
 8 secs
 88
 9 secs
+9 is still
 99
 0 secs
+300 bot
+600 bot
 twinkie
 two secs
 three secs
@@ -22,6 +27,7 @@ eight secs
 nine secs
 ten secs
 aboutit
+admit it's
 ain't it
 alt
 an ai
@@ -78,6 +84,7 @@ few secs
 ffa game
 fire cracker
 fire crackers
+forgot it's
 francoitalian
 franco italian
 freakin
@@ -101,6 +108,7 @@ hellen
 hellp
 h on keyboard
 h tier
+hi @Bla
 hi tirp
 ho ho ho
 honkeytonk
@@ -184,6 +192,7 @@ pp. 9
 pussinboots
 puss in boots
 ref'd
+refresh at
 rip
 saturated fat
 shoehorn your
@@ -197,6 +206,7 @@ suicide squad
 superbowlxxx
 tally ho
 tally-ho
+tea the
 test test test
 then i guess
 then talk
@@ -229,6 +239,7 @@ virgin islands
 wassup
 wasn't it
 wouldn't it
+xp or no
 yass
 yesturday
 zenga

src/dictionary_extra.txt

@@ -1,13 +1,18 @@
 # of
 #8
+(until
 0 secs
 2 secs
 3 secs
+300 bot
 4 secs
+45s
 5 secs
 6 secs
+600 bot
 7 secs
 8 secs
+9 is still
 9 secs
 a analog
 a analyse
@@ -147,6 +152,7 @@ adipex nissan
 adipex pee
 adipex rated
 adiposogenital
+admit it's
 ado lif
 adramelech
 adrammelech
@@ -2749,6 +2755,14 @@ bol lock
 bol locks
 bol look
 bol looks
+bomb china
+bomb india
+bomb iran
+bomb israel
+bomb palestine
+bomb russia
+bomb ukraine
+bomb usage
 bon ed
 bon eric
 bon erik
@@ -6863,6 +6877,7 @@ fore skin
 forebreast
 forget lost
 forget married
+forgot it's
 fork cocktail
 fork commission
 fork cook
@@ -7800,6 +7815,7 @@ heterosex
 heterotic
 hexadic
 hexanal
+hi @Bla
 hi little
 hi tier
 hi tile
@@ -7822,6 +7838,7 @@ highs perm
 highs seeks
 hilar
 hildebrandic
+hill hitting
 hill illus
 hill iv
 hill ju
@@ -9654,6 +9671,7 @@ junk until
 junk untitled
 junk unto
 jurisprude
+just cumulative
 justments cumulative
 justments ext
 justments hilt
@@ -9998,6 +10016,7 @@ kill twelve
 kill twenty
 kill twi
 kill ty
+killed yourself
 killian
 killing jewel
 killing palestinian
@@ -10351,6 +10370,7 @@ less blin
 less bo
 lets cumulative
 lets ext
+lets fake
 lets hilt
 lets hit
 lets lut
@@ -13515,6 +13535,7 @@ plumbaginaceous
 plumbum
 plumigerous
 plzz
+pmsg
 pn lips
 pn nigeria
 pnigerophobia
@@ -15134,6 +15155,7 @@ res perm
 res seeks
 resex
 resh aging
+resh at
 resh hilt
 resh hit
 resh it
@@ -17647,6 +17669,7 @@ tch linking
 tch links
 tch little
 tchincou
+tea the
 teanal
 teapottykin
 teataster
@@ -19880,6 +19903,7 @@ xnxx until
 xnxx untitled
 xnxx unto
 xnxx vie
+xp or no
 ya holes
 yacht its
 yacht texts