feat(core): many more Weir rules (#2742)

* feat(core): a ton of new Weir rules

* chore(core): move to the proper folder

Author: elijah-potter

harper-core/src/linting/closed_compounds.rs

@@ -25,6 +25,8 @@ pub fn lint_group() -> LintGroup {
         "Anyhow"          => ("any how", "anyhow"),
         "Anywhere"        => ("any where", "anywhere"),
         "Backplane"       => ("back plane", "backplane"),
+        "Bypass"          => ("by pass", "bypass"),
+        "Deadlift"        => ("dead lift", "deadlift"),
         "Desktop"         => ("desk top", "desktop"),
         "Devops"          => ("dev ops", "devops"),
         "Everybody"       => ("every body", "everybody"),
@@ -37,6 +39,8 @@ pub fn lint_group() -> LintGroup {
         "Instead"         => ("in stead", "instead"),
         "Intact"          => ("in tact", "intact"),
         "Itself"          => ("it self", "itself"),
+        "Keystroke"       => ("key stoke", "keystroke"),
+        "Keystrokes"      => ("key stokes", "keystrokes"),
         "Laptop"          => ("lap top", "laptop"),
         "Middleware"      => ("middle ware", "middleware"),
         "Misunderstand"   => ("miss understand", "misunderstand"),
@@ -59,17 +63,21 @@ pub fn lint_group() -> LintGroup {
         "Postpone"        => ("post pone", "postpone"),
         "Proofread"       => ("proof read", "proofread"),
         "Regardless"      => ("regard less", "regardless"),
+        "Shortcoming"     => ("short coming", "shortcoming"),
+        "Shortcomings"    => ("short comings", "shortcomings"),
         "Somebody"        => ("some body", "somebody"),
         "Somehow"         => ("some how", "somehow"),
         "Someone"         => ("some one", "someone"),
         "Somewhere"       => ("some where", "somewhere"),
+        "There"           => ("the re", "there"),
         "Therefore"       => ("there fore", "therefore"),
         "Thereupon"       => ("there upon", "thereupon"),
         "Underclock"      => ("under clock", "underclock"),
         "Upset"           => ("up set", "upset"),
         "Upward"          => ("up ward", "upward"),
         "Whereupon"       => ("where upon", "whereupon"),
         "Widespread"      => ("wide spread", "widespread"),
+        "Without"         => ("with out", "without"),
         "Worldwide"       => ("world wide", "worldwide"),
     });
 
@@ -223,4 +231,60 @@ mod tests {
         let expected = "They set off on their journey overnight.";
         assert_suggestion_result(test_sentence, lint_group(), expected);
     }
+
+    #[test]
+    fn by_pass() {
+        let test_sentence = "Please by pass this check for now.";
+        let expected = "Please bypass this check for now.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn dead_lift() {
+        let test_sentence = "I can dead lift 200 kg.";
+        let expected = "I can deadlift 200 kg.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn key_stoke() {
+        let test_sentence = "Use this key stoke to open search.";
+        let expected = "Use this keystroke to open search.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn key_stokes() {
+        let test_sentence = "These key stokes are hard to memorize.";
+        let expected = "These keystrokes are hard to memorize.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn with_out() {
+        let test_sentence = "We left with out a map.";
+        let expected = "We left without a map.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn the_re() {
+        let test_sentence = "The re are too many popups on this page.";
+        let expected = "There are too many popups on this page.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn short_coming() {
+        let test_sentence = "That bug is a short coming in the current release.";
+        let expected = "That bug is a shortcoming in the current release.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
+
+    #[test]
+    fn short_comings() {
+        let test_sentence = "We listed three short comings in the postmortem.";
+        let expected = "We listed three shortcomings in the postmortem.";
+        assert_suggestion_result(test_sentence, lint_group(), expected);
+    }
 }

weir_rules/AtLeasToLeast.weir

@@ -0,0 +1,34 @@
+expr main <([(at leas)]), leas>
+
+let message "Use the standard phrase with `least` here."
+let description "Fixes the frequent typo `at leas` when the intended expression is `at least`."
+let kind "Typo"
+let becomes "least"
+let strategy "MatchCase"
+
+# True positives
+
+test "At leas we shipped the patch." "At least we shipped the patch."
+test "at leas I remembered to push." "at least I remembered to push."
+test "We can, at leas, retry tomorrow." "We can, at least, retry tomorrow."
+test "AT LEAS TRY AGAIN." "AT LEAST TRY AGAIN."
+test "I guess at leas one test should pass." "I guess at least one test should pass."
+test "At leas, this error is reproducible." "At least, this error is reproducible."
+
+# True negatives / false positives
+
+test "At least we shipped the patch." "At least we shipped the patch."
+test "I have a leas on that apartment." "I have a leas on that apartment."
+test "Please at leisure review the draft." "Please at leisure review the draft."
+test "The least risky option won." "The least risky option won."
+test "They said: \"at leash\" as a joke." "They said: \"at leash\" as a joke."
+test "The release is at, leas, in notes." "The release is at, leas, in notes."
+
+# Boundary and false-negative coverage
+
+test "At leas? Maybe." "At least? Maybe."
+test "If not now, at leas tomorrow." "If not now, at least tomorrow."
+test "We are at leas halfway done." "We are at least halfway done."
+test "At leas!" "At least!"
+test "I wrote \"at leas\" twice: at leas." "I wrote \"at least\" twice: at least."
+test "Looking at least-cost models." "Looking at least-cost models."

weir_rules/BelieveInPreposition.weir

@@ -0,0 +1,37 @@
+expr main [(believe to)]
+
+let message "This verb usually takes `in` here."
+let description "Replaces `believe to` with `believe in` in common object-taking usage."
+let kind "Grammar"
+let becomes "believe in"
+let strategy "MatchCase"
+
+# True positives
+
+test "I believe to this plan." "I believe in this plan."
+test "They believe to ghosts." "They believe in ghosts."
+test "Many teens believe to luck." "Many teens believe in luck."
+test "Do you believe to fate?" "Do you believe in fate?"
+test "We still believe to science." "We still believe in science."
+test "I believe to your potential." "I believe in your potential."
+test "BELIEVE TO YOURSELF FOR ONCE." "BELIEVE IN YOURSELF FOR ONCE."
+test "I wonder if you believe to miracles." "I wonder if you believe in miracles."
+
+# True negatives / false positives
+
+allows "I believe in this plan."
+allows "They believe that this plan works."
+allows "I choose to believe this story."
+allows "We believe in each other."
+allows "Do you believe in fate?"
+
+# Known false positives (current boundary)
+
+test "I believe to this day that we were right." "I believe in this day that we were right."
+test "I believe to this moment it is true." "I believe in this moment it is true."
+
+# Potential false negatives (accepted gaps)
+
+allows "She believes to her team."
+allows "We believed to that strategy for years."
+allows "He is believing to the mission again."

weir_rules/BetterOffPhrase.weir

@@ -0,0 +1,34 @@
+expr main (better of)
+
+let message "This expression usually takes `off` instead of `of`."
+let description "Rewrites `better of` to `better off` in common comparative phrasing."
+let kind "Grammar"
+let becomes "better off"
+let strategy "MatchCase"
+
+# True positives
+
+test "People who are better of can donate more." "People who are better off can donate more."
+test "You will be better of waiting until tomorrow." "You will be better off waiting until tomorrow."
+test "I feel better of without that app." "I feel better off without that app."
+test "We are better of focusing on quality." "We are better off focusing on quality."
+test "They were better of before the update." "They were better off before the update."
+test "She is far better of now." "She is far better off now."
+test "He'd be better of at home." "He'd be better off at home."
+test "The team is BETTER OF with clear goals." "The team is BETTER OFF with clear goals."
+test "Are we better of if we delay launch?" "Are we better off if we delay launch?"
+
+# True negatives / false-positive checks
+
+test "People who are better off can donate more." "People who are better off can donate more."
+test "This is a matter of better options." "This is a matter of better options."
+test "Of course this is better overall." "Of course this is better overall."
+test "He offered better outcomes, not quick fixes." "He offered better outcomes, not quick fixes."
+test "They got off the train early." "They got off the train early."
+test "A better offer arrived yesterday." "A better offer arrived yesterday."
+
+# Boundary / accepted limits
+
+allows "She chose the better option from the list."
+allows "Pick the best design from these three mockups."
+allows "They discussed old and new methods in detail."

weir_rules/BluRayHyphen.weir

@@ -0,0 +1,32 @@
+expr main (blu ray)
+
+let message "Use the hyphenated spelling for this media format."
+let description "Joins the two-word spelling of the optical disc format into the standard compound form."
+let kind "Punctuation"
+let becomes "blu-ray"
+let strategy "MatchCase"
+
+# True positives
+
+test "I ordered a Blu ray player." "I ordered a Blu-ray player."
+test "This movie ships on blu ray." "This movie ships on blu-ray."
+test "Do you still buy BLU RAY discs?" "Do you still buy BLU-RAY discs?"
+test "A Blu ray drive is installed." "A Blu-ray drive is installed."
+test "They found cheap blu ray movies." "They found cheap blu-ray movies."
+test "BLU RAY" "BLU-RAY"
+
+# False positives / true negatives
+
+test "I ordered a Blu-ray player." "I ordered a Blu-ray player."
+test "The ray looked blue in that photo." "The ray looked blue in that photo."
+test "We watched a DVD last night." "We watched a DVD last night."
+test "The Blu rays are on the shelf." "The Blu rays are on the shelf."
+test "A blur ay effect appeared in the lens." "A blur ay effect appeared in the lens."
+test "The brand name is Blue Ray Labs." "The brand name is Blue Ray Labs."
+
+# False negatives and edge behavior checks
+
+test "blu ray" "blu-ray"
+test "Blu ray, DVD, and VHS." "Blu-ray, DVD, and VHS."
+test "I keep one blu ray." "I keep one blu-ray."
+test "Many people said BLU ray still works." "Many people said BLU-ray still works."

weir_rules/CauseItIsBecause.weir

@@ -0,0 +1,33 @@
+expr main (cause it is)
+
+let message "Prefer `because it is` in this clause."
+let description "Normalizes informal `cause it is` to the standard subordinating form in explanatory clauses."
+let kind "Style"
+let becomes "because it is"
+let strategy "MatchCase"
+
+# True positives
+
+test "Cause it is late, we should leave." "Because it is late, we should leave."
+test "cause it is noisy, I use headphones." "because it is noisy, I use headphones."
+test "I stayed home cause it is raining." "I stayed home because it is raining."
+test "We paused deployment cause it is unstable." "We paused deployment because it is unstable."
+test "CAUSE IT IS CLEAR NOW." "BECAUSE IT IS CLEAR NOW."
+test "Cause it is simple, ship it." "Because it is simple, ship it."
+test "They agreed cause it is practical." "They agreed because it is practical."
+
+# False negatives (intentionally out of scope)
+
+test "Cause it was late, we should leave." "Cause it was late, we should leave."
+test "Cause it's late, we should leave." "Cause it's late, we should leave."
+test "Cause it has changed, reread docs." "Cause it has changed, reread docs."
+test "Cause it had failed, we rolled back." "Cause it had failed, we rolled back."
+
+# True negatives / false positives
+
+test "Because it is late, we should leave." "Because it is late, we should leave."
+test "Stress can cause it is not obvious." "Stress can because it is not obvious."
+test "The cause it is unclear." "The because it is unclear."
+test "They discuss cause and effect." "They discuss cause and effect."
+test "Because it was late, we should leave." "Because it was late, we should leave."
+test "No one said cause it is wrong." "No one said because it is wrong."

weir_rules/ColdModalTypo.weir

@@ -0,0 +1,38 @@
+expr subjects [i, you, we, they, he, she, it]
+expr baseVerbs [be, have, do, go, get, make, take, see, say, know, think, find, use, work, help, start, stop, leave, feel, need, want]
+expr main <([(@subjects cold @baseVerbs), (@subjects cold ADV @baseVerbs), (@subjects cold not @baseVerbs), (@subjects cold never @baseVerbs), (@subjects cold ADV not @baseVerbs)]), cold>
+
+let message "This looks like a modal typo."
+let description "Rewrites `cold` to `could` when it appears in common subject-plus-verb modal contexts."
+let kind "Typo"
+let becomes "could"
+let strategy "MatchCase"
+
+# True positives
+
+test "I cold go now." "I could go now."
+test "You cold make dinner tonight." "You could make dinner tonight."
+test "We cold take the train." "We could take the train."
+test "They cold see the issue." "They could see the issue."
+test "He cold feel the difference." "He could feel the difference."
+test "She cold help us later." "She could help us later."
+test "It cold work this time." "It could work this time."
+test "I cold not find the setting." "I could not find the setting."
+test "We cold never know the reason." "We could never know the reason."
+test "They cold really use a break." "They could really use a break."
+test "I COLD GO IF NEEDED." "I COULD GO IF NEEDED."
+
+# False positives / true negatives
+
+allows "It was cold outside all day."
+allows "A cold drink helped a lot."
+allows "He caught a cold last week."
+allows "The soup went cold quickly."
+allows "I enjoy cold brew coffee."
+allows "She gave me a cold stare."
+
+# False negatives (accepted limits)
+
+allows "I cold call prospects each morning."
+allows "You cold text me later."
+allows "They cold build a prototype fast."

weir_rules/DoubleCheckHyphen.weir

@@ -0,0 +1,30 @@
+expr main (double check)
+
+let message "Use the hyphenated form in this compound."
+let description "Normalizes the common two-word form `double check` to `double-check`."
+let kind "Punctuation"
+let becomes "double-check"
+let strategy "MatchCase"
+
+# True positives
+
+test "Please double check the report before sending." "Please double-check the report before sending."
+test "I always double check numbers in production." "I always double-check numbers in production."
+test "Can you double check this again?" "Can you double-check this again?"
+test "We should double check with legal." "We should double-check with legal."
+test "You need to double check every file." "You need to double-check every file."
+test "DOUBLE CHECK THE FINAL TOTAL." "DOUBLE-CHECK THE FINAL TOTAL."
+test "double check this quickly." "double-check this quickly."
+test "If unsure, double check." "If unsure, double-check."
+
+# True negatives / false positives
+
+test "Please double-click the button." "Please double-click the button."
+test "Please double click the button." "Please double click the button."
+test "They double checked the config after deploy." "They double checked the config after deploy."
+test "We are double checking every migration." "We are double checking every migration."
+test "She double checks each invoice manually." "She double checks each invoice manually."
+test "A careful check is still needed." "A careful check is still needed."
+test "The team did one more review." "The team did one more review."
+test "double-check the report before sending." "double-check the report before sending."
+test "Double-checking takes discipline." "Double-checking takes discipline."

weir_rules/EachOthersPossessive.weir

@@ -0,0 +1,34 @@
+expr main <((each others) [NOUN, PROPN, ADJ, DET]), (each others)>
+
+let message "Use the reciprocal possessive form here."
+let description "Rewrites `each others` to `each other's` when it modifies a following noun phrase."
+let kind "Grammar"
+let becomes "each other's"
+let strategy "MatchCase"
+
+# True positives
+
+test "They reviewed each others code before release." "They reviewed each other's code before release."
+test "We checked each others notes after class." "We checked each other's notes after class."
+test "The two teams respected each others traditions." "The two teams respected each other's traditions."
+test "The siblings borrowed each others bikes." "The siblings borrowed each other's bikes."
+test "Please verify each others entries." "Please verify each other's entries."
+test "They corrected each others grammar in chat." "They corrected each other's grammar in chat."
+test "Each others schedules kept drifting." "Each other's schedules kept drifting."
+test "EACH OTHERS PROJECTS WERE MERGED." "EACH OTHER'S PROJECTS WERE MERGED."
+test "Writers should respect each others voices." "Writers should respect each other's voices."
+test "Designers critique each others drafts weekly." "Designers critique each other's drafts weekly."
+
+# False positives / true negatives
+
+test "They reviewed each other's code before release." "They reviewed each other's code before release."
+test "They reviewed each other during rehearsals." "They reviewed each other during rehearsals."
+test "Others' opinions were shared in the thread." "Others' opinions were shared in the thread."
+test "Each other was mentioned in the report." "Each other was mentioned in the report."
+test "They checked others code before release." "They checked others code before release."
+test "They checked others' code before release." "They checked others' code before release."
+
+# False negatives (known limitation: does not target misspelled reciprocal phrases with extra apostrophes)
+
+allows "They reviewed each others' code before release."
+allows "They corrected each others' grammar in chat."