Fix Linter

diff --git a/extra/lint/vocabs/vocabs-docs.factor b/extra/lint/vocabs/vocabs-docs.factor
index af18fe37757dd8ab34e699f2f20c063a20cb6ff2..66dd354359c8c87bfbe0c997eda9a412b0a0c713 100644 (file)
--- a/extra/lint/vocabs/vocabs-docs.factor
+++ b/extra/lint/vocabs/vocabs-docs.factor
@@ -57,49 +57,6 @@ HELP: find-unused.
     }
 } ;
 
-HELP: get-imported-words
-{ $values
-    { "string" string }
-    { "hashtable" hashtable }
-}
-{ $description 
-    "Gets all words that have been imported with " { $link \ USE: } " and " { $link \ USING: } " in the given string."
-} ;
-
-HELP: get-vocabs
-{ $values
-    { "string" string }
-    { "seq" sequence }
-}
-{ $description 
-    "Gets all the vocabularies imported in the given string." 
-} ;
-
-HELP: get-words
-{ $values
-    { "name" "a vocab name string" }
-    { "assoc" assoc }
-}
-{ $description 
-    "Gets all the words used in a given vocabulary." 
-} 
-{ $examples
-    { $example "USING: lint.vocabs prettyprint ;"
-        "\"lint.vocabs\" get-words ."
-"{
-    \"lint.vocabs\"
-    {
-        \"get-vocabs\"
-        \"get-words\"
-        \"find-unused-in-file\"
-        \"get-imported-words\"
-        \"find-unused-in-string\"
-        \"find-unused.\"
-        \"find-unused\"
-    }
-}"
-    }
-} ;
 
 ARTICLE: "lint.vocabs" "The Unused Vocabulary Linter"
 "The " { $vocab-link "lint.vocabs" } " vocabulary implements a set of words designed to find unused imports."

diff --git a/extra/lint/vocabs/vocabs-tests.factor b/extra/lint/vocabs/vocabs-tests.factor
index c34cea6ea3b439c34ce8c2f94521174857754e42..9b552e50218bc379ad0a0d98e4b12ca23ae12989 100644 (file)
--- a/extra/lint/vocabs/vocabs-tests.factor
+++ b/extra/lint/vocabs/vocabs-tests.factor
@@ -2,8 +2,9 @@
 ! See http://factorcode.org/license.txt for BSD license.
 USING: accessors arrays assocs compiler.units continuations
 formatting hash-sets hashtables io io.encodings.utf8 io.files
-kernel namespaces regexp sequences sequences.deep sets sorting
-splitting tools.test unicode vocabs vocabs.loader ;
+kernel namespaces regexp sequences sequences.deep
+sequences.parser sets sorting splitting tools.test unicode
+vocabs vocabs.loader ;
 IN: lint.vocabs
 
 <PRIVATE
@@ -46,6 +47,18 @@ CONSTANT: empty-using-statement     "USING: ; nop ( -- ) ;"
 : ---- ( -- ) "-------------------------------------------------------------------------" print ;
 PRIVATE>
 
+"next-token should get the next non-blank string in the stream:" print
+{ "hello" } [ "hello world!"              <sequence-parser> next-token ] unit-test
+{ "hello" } [ "\n    hello \n world!    " <sequence-parser> next-token ] unit-test
+
+----
+
+"next-token should ignore comments:" print
+{ "world!" } [ "! hello\nworld!"                 <sequence-parser> next-token ] unit-test
+{ "world!" } [ "! h\n! e\n! l\n! l\n! o\nworld!" <sequence-parser> next-token ] unit-test
+
+----
+
 "It should work on multiple lines, with multiple imports across the file: " print
 
 { { "hashtables" "math.primes" "sequences" "sets" "vocabs" } } [ mock-file find-unused-in-string ] unit-test

diff --git a/extra/lint/vocabs/vocabs.factor b/extra/lint/vocabs/vocabs.factor
index 2525bdfb94dbdf31195d0b2a8e74e9b44fab47f1..193883e8838c91c542dbc1c9d9e84569b0eb2dba 100644 (file)
--- a/extra/lint/vocabs/vocabs.factor
+++ b/extra/lint/vocabs/vocabs.factor
@@ -1,9 +1,8 @@
 ! Copyright (C) 2022 CapitalEx
 ! See http://factorcode.org/license.txt for BSD license.
-USING: accessors arrays assocs combinators combinators.extras
-combinators.short-circuit compiler.units formatting
-grouping.extras hash-sets hashtables io io.encodings.utf8
-io.files kernel literals multiline namespaces peg.ebnf regexp
+USING: accessors arrays assocs combinators 
+combinators.short-circuit compiler.units formatting hash-sets
+hashtables io io.encodings.utf8 io.files kernel namespaces
 sequences sequences.deep sequences.parser sets sorting splitting
 strings unicode vocabs vocabs.loader  ;
 FROM: namespaces => set ;
@@ -25,9 +24,6 @@ SYMBOL: old-dictionary
 : vocab-loaded? ( name -- ? )
     dictionary get key? ;
 
-USE: literals
-CONSTANT: USING-PATTERN $[ "USING: [^;]+ ;|USE: \\S+" <regexp> ]
-
 ! Helper words
 : tokenize ( string -- sequence-parser )
     <sequence-parser> ;
@@ -91,10 +87,11 @@ DEFER: next-token
 
         ! prune syntax stuff
         { "FROM:"     [ ";" skip-after f ] }
-        { "IN:"       [     skip-token f ] }
-        { "SYMBOL:"   [     skip-token f ] }
         { "SYMBOLS:"  [ ";" skip-after f ] }
+        { "R/"        [ "/" skip-after f ] }
         { "("         [ ")" skip-after f ] }
+        { "IN:"       [     skip-token f ] }
+        { "SYMBOL:"   [     skip-token f ] }
         { ":"         [     skip-token f ] }
         { "POSTPONE:" [     skip-token f ] }
         { "\\"        [     skip-token f ] }
@@ -161,12 +158,6 @@ DEFER: next-token
 : strip-code ( string -- string )
     tokenize V{ } clone swap (strip-code) ;
 
-: extract-imports ( string -- seq )
-    USING-PATTERN all-matching-subseqs ;
-
-: remove-imports ( string -- seq )
-    USING-PATTERN "" re-replace ;
-
 ! Words for finding the words used ina program, stripping out import statements
 : skip-imports ( sequence-parser -- sequence-parser string/? )
     dup next { 
@@ -191,8 +182,8 @@ DEFER: next-token
 : (find-imports) ( vector sequence-parser -- vector )
     dup take-imports rot prepend swap [ (find-imports) ] ?keep-parsing-with ;
 
-: find-imports ( vector -- set )
-    <sequence-parser> V{ } clone swap (find-imports) fast-set ;
+: find-imports ( vector -- seq )
+    <sequence-parser> V{ } clone swap (find-imports) ;
 
 : (get-words) ( name -- vocab )
     dup load-vocab words>> keys 2array ;
@@ -225,4 +216,27 @@ DEFER: next-token
 : print-no-unused-vocabs ( name _ -- )
     drop "No unused vocabs found in %s.\n" printf ;
 
-PRIVATE>
\ No newline at end of file
+: get-words ( name -- assoc )
+    dup vocab-exists? [ (get-words) ] [ no-vocab-found ] if ;
+
+: get-imported-words ( string -- hashtable )
+    save-dictionary
+        find-imports [ get-words ] map >hashtable
+    restore-dictionary ;
+
+PRIVATE>
+
+: find-unused-in-string ( string -- seq )
+    strip-code [ get-imported-words ] [ find-used-words ] bi
+        reject-unused-vocabs natural-sort ;
+
+: find-unused-in-file ( path -- seq )
+    utf8 file-contents find-unused-in-string ;
+
+: find-unused ( name -- seq )
+    vocab-source-path dup [ find-unused-in-file ] when ;
+
+: find-unused. ( name -- )
+    dup find-unused dup empty?
+        [ print-no-unused-vocabs ]
+           [ print-unused-vocabs ] if ;
\ No newline at end of file