From d7168e8575794b7af4e7e2bde51ca0663bbc2eac Mon Sep 17 00:00:00 2001 From: =?utf8?q?Mattias=20Engdeg=C3=A5rd?= Date: Wed, 5 Jul 2023 15:25:30 +0200 Subject: [PATCH] Improve some big regexps * lisp/language/lao-util.el (lao-transcription-pattern): Remove duplicate definition. * lisp/language/tibetan.el (tibetan-regexp, tibetan-precomposed-regexp) (tibetan-precomposition-rule-regexp): Use regexp-opt. --- lisp/language/lao-util.el | 19 ------------------ lisp/language/tibetan.el | 41 +++++++++++---------------------------- 2 files changed, 11 insertions(+), 49 deletions(-) diff --git a/lisp/language/lao-util.el b/lisp/language/lao-util.el index 6b9a8984414..33e98a60d63 100644 --- a/lisp/language/lao-util.el +++ b/lisp/language/lao-util.el @@ -298,25 +298,6 @@ ("\\\\" . "ໆ") ("\\$" . "ຯ"))) -(defconst lao-transcription-pattern - (concat - "\\(" - (mapconcat 'car lao-transcription-consonant-alist "\\|") - "\\)\\(" - (mapconcat 'car lao-transcription-semi-vowel-alist "\\|") - "\\)?\\(\\(" - (mapconcat 'car lao-transcription-vowel-alist "\\|") - "\\)\\(" - (mapconcat 'car lao-transcription-maa-sakod-alist "\\|") - "\\)?\\(" - (mapconcat (lambda (x) (regexp-quote (car x))) - lao-transcription-tone-alist "\\|") - "\\)?\\)?\\|" - (mapconcat (lambda (x) (regexp-quote (car x))) - lao-transcription-punctuation-alist "\\|") - ) - "Regexp of Roman transcription pattern for one Lao syllable.") - (defconst lao-transcription-pattern (concat "\\(" diff --git a/lisp/language/tibetan.el b/lisp/language/tibetan.el index 1292092c582..2249fa467b1 100644 --- a/lisp/language/tibetan.el +++ b/lisp/language/tibetan.el @@ -558,48 +558,29 @@ ("སྨ" . "ö†°"))) (defconst tibetan-regexp - (let (pattern) - (dolist (alist (list tibetan-precomposed-transcription-alist - tibetan-consonant-transcription-alist - tibetan-vowel-transcription-alist - tibetan-modifier-transcription-alist - tibetan-subjoined-transcription-alist) - (apply #'concat (nreverse (cdr pattern)))) - (dolist (key-val alist) - (setq pattern (cons "\\|" (cons (regexp-quote (car key-val)) - pattern)))))) + (regexp-opt + (mapcar (lambda (x) (regexp-quote (car x))) + (append tibetan-precomposed-transcription-alist + tibetan-consonant-transcription-alist + tibetan-vowel-transcription-alist + tibetan-modifier-transcription-alist + tibetan-subjoined-transcription-alist))) "Regexp matching a Tibetan transcription of a composable Tibetan sequence. The result of matching is to be used for indexing alists at conversion from a roman transcription to the corresponding Tibetan character.") (defvar tibetan-precomposed-regexp (purecopy - (let ((l tibetan-precomposed-transcription-alist) - temp) - (setq temp "^\\(") - (setq temp - (concat temp (car (car l)))) - (setq l (cdr l)) - (while l - (setq temp - (concat temp "\\|" (car (car l)))) - (setq l (cdr l))) - (concat temp "\\)"))) + (concat "^" (regexp-opt + (mapcar #'car tibetan-precomposed-transcription-alist) + t))) "Regexp string to match a romanized Tibetan complex consonant. The result of matching is to be used for indexing alists when the input key from an input method is converted to the corresponding precomposed glyph.") (defvar tibetan-precomposition-rule-regexp (purecopy - (let ((l tibetan-precomposition-rule-alist) - temp) - (setq temp "\\(") - (setq temp (concat temp (car (car l)))) - (setq l (cdr l)) - (while l - (setq temp (concat temp "\\|" (car (car l)))) - (setq l (cdr l))) - (concat temp "\\)"))) + (regexp-opt (mapcar #'car tibetan-precomposition-rule-alist) t)) "Regexp string to match a sequence of Tibetan consonantic components. That is, one base consonant and one or more subjoined consonants. The result of matching is to be used for indexing alist when the component -- 2.39.2