The search ignore diacritics

Makefile

@@ -6,7 +6,7 @@ $(eval isGuix := $(shell command -v guix > /dev/null 2>&1 && echo t || echo f))
 $(eval destDir := $(shell [ "${isGuix}" = t ] && echo $${dotf}/bin || echo ~/bin))
 orgRoamLink := ${HOME}/org-roam
 
-all: show-environment clean install-deps
+all: show-environment clean install-deps test
 	[ ! -L "${orgRoamLink}" ] && ln -s ${dev}/notes/notes "${orgRoamLink}" || :
 	[ ! -d ${destDir} ] && mkdir ${destDir} || :
 	raco exe -o ${destDir}/search-notes main.rkt
@@ -32,3 +32,5 @@ install-deps:
 clean:
 	rm -rf ./compiled/ ./scribblings/compiled/
 
+test:
+	raco test ./

README.md

@@ -5,6 +5,7 @@ Installation: run `make`. See also Makefile.
 # Fix the 'loading code: version mismatch' error
 rm -rf ./compiled/ ./scribblings/compiled/
 raco pkg install --auto ansi-color
+# raco test ./  # optionally
 isGuix=$(command -v guix > /dev/null 2>&1 && echo t || echo f)
 [ ${isGuix} = t ] && destDir=$dotf/bin || destDir=~/bin
 [ ! -d ${destDir} ] && mkdir $destDir || :

main.rkt

@@ -3,10 +3,6 @@
 ;; TODO try #lang hacket - haskell + racket
 ;; https://lexi-lambda.github.io/hackett/index.html
 
-(module+ test
-  (require rackunit
-           racket/match))
-
 ;; Notice
 ;; To install (from within the package directory):
 ;;   $ raco pkg install
@@ -27,28 +23,88 @@
 ;; See the current version of the racket style guide here:
 ;; http://docs.racket-lang.org/style/index.html
 
-;; Code here
-
-
-
-(module+ test
-  ;; Any code in this `test` submodule runs when this file is run using DrRacket
-  ;; or with `raco test`. The code here does not run when this file is
-  ;; required by another module.
-
-  (check-equal? (+ 2 2) 4))
-
 (module+ main
   ;; (Optional) main submodule. Put code here if you need it to be executed when
   ;; this file is run using DrRacket or the `racket` executable.  The code here
   ;; does not run when this file is required by another module. Documentation:
   ;; http://docs.racket-lang.org/guide/Module_Syntax.html#%28part._main-and-test%29
 
+  (provide regexp-normalize-match* regexp-normalize-split)
+
   (require
    ;; (prefix-in com: "common.rkt")
    "notes.rkt" ;; is used indeed
    "notes-reader.rkt"
-   ansi-color)
+   ansi-color
+
+  ;; for string-replace
+  racket/string
+  )
+
+  (define diacritic-map
+    (hash "a" "[aáäàâæ]"
+          "c" "[cčç]"
+          "d" "[dď]"
+          "e" "[eéèêë]"
+          "i" "[iíîï]"
+          "l" "[lĺľ]"
+          "n" "[nň]"
+          "o" "[oóôöœ]"
+          "r" "[rŕř]"
+          "s" "[sš]"
+          "t" "[tť]"
+          "u" "[uúûüù]"
+          "y" "[yý]"
+          "z" "[zž]"
+          "A" "[AÁÄÀÂÆ]"
+          "C" "[CČÇ]"
+          "D" "[DĎ]"
+          "E" "[EÉÈÊË]"
+          "I" "[IÍÎÏ]"
+          "L" "[LĹĽ]"
+          "N" "[NŇ]"
+          "O" "[OÓÔÖŒ]"
+          "R" "[RŔŘ]"
+          "S" "[SŠ]"
+          "T" "[TŤ]"
+          "U" "[UÚÛÜÙ]"
+          "Y" "[YÝ]"
+          "Z" "[ZŽ]"
+          "ß" "ß")) ; German sharp S
+
+  (define (string-normalize s)
+    ;; Normalization Form C, Canonical Decomposition followed by Canonical
+    ;; Composition:
+    ;; Decompose characters and then recomposes them using canonical
+    ;; equivalence. E.g., 'é' would first be split into 'e' and the combining
+    ;; accent, and then recomposed back into 'é'.
+    ;; Use this when you want to normalize characters to their composed forms
+    ;; while still respecting canonical equivalence.
+    (string-normalize-nfc s))
+
+  (define (regexp-normalize-match* rxs target-str)
+    (let* ((normalized-target (string-normalize target-str))
+           (modified-regex
+            (string-append*
+             (map (lambda (char)
+                    (hash-ref diacritic-map (string char) (string char)))
+                  (string->list rxs)))))
+      ;; (printf "modified-regex: ~a\n" modified-regex)
+      ;; (printf "target-str: ~a\n" target-str)
+      ;; (printf "normalized-target: ~a\n" normalized-target)
+      (regexp-match* (regexp modified-regex) normalized-target)))
+
+  (define (regexp-normalize-split rxs target-str)
+    (let* ((normalized-target (string-normalize target-str))
+           (modified-regex
+            (string-append*
+             (map (lambda (char)
+                    (hash-ref diacritic-map (string char) (string char)))
+                  (string->list rxs)))))
+      ;; (printf "modified-regex: ~a\n" modified-regex)
+      ;; (printf "target-str: ~a\n" target-str)
+      ;; (printf "normalized-target: ~a\n" normalized-target)
+      (regexp-split (regexp modified-regex) normalized-target)))
 
   (define pattern-param (make-parameter ""))
   (define filepaths-param (make-parameter ""))
@@ -164,10 +220,10 @@ racket main.rkt -e \"/home/bost/der/search-notes/main.rkt /home/bost/der/search-
                        (displayln first-file-string))
                    (colorize colorize-matches?
                              display-fn
-                             (regexp-split  regexp-split-match
-                                            relevant-file-strings-joined)
-                             (regexp-match* regexp-split-match
-                                            relevant-file-strings-joined))
+                             (regexp-normalize-split
+                              regexp-split-match relevant-file-strings-joined)
+                             (regexp-normalize-match*
+                              regexp-split-match relevant-file-strings-joined))
                    (printf "\n\n")))
                relevant-file-strings)))
     (curry map

scribblings/search-notes.scrbl

@@ -1,5 +1,6 @@
 #lang scribble/manual
-@require[@for-label[search-notes
+@require[@for-label[
+                    ;; search-notes
                     racket/base]]
 
 @title{search-notes}

test.rkt

@@ -0,0 +1,31 @@
+#lang racket
+
+(module+ test
+  (require
+   ;; main
+   rackunit
+   racket/match
+   "main.rkt"
+   (submod "main.rkt" main)
+   )
+
+  ;; Any code in this `test` submodule runs when this file is run using DrRacket
+  ;; or with `raco test`. The code here does not run when this file is
+  ;; required by another module.
+
+  ;; (test-case "Test for add function"
+  ;;   (check-equal? (+ 2 2) 4))
+
+  (test-case "Test diacritics"
+    (define sdiacr "jkl \n abčd \n xyz \n 123 \n ábc \n 567")
+    (define sd sdiacr)
+    (define splain "jkl \n abcd \n xyz \n 123 \n abc \n 567")
+    (define sp splain)
+    (define rxs "abc")
+
+    (check-equal? (length (regexp-normalize-match* rxs sdiacr))
+                  (length (regexp-match* rxs splain)))
+    (check-equal? (regexp-normalize-split rxs sdiacr)
+                  (regexp-split rxs splain))
+    (check-equal? (regexp-normalize-split rxs splain)
+                  (regexp-split rxs splain))))