summaryrefslogtreecommitdiff
path: root/module
diff options
context:
space:
mode:
authorEkaitz Zarraga <ekaitz@elenq.tech>2024-12-22 21:01:07 +0100
committerLudovic Courtès <ludo@gnu.org>2024-12-22 22:14:12 +0100
commit38ad264979b06535200ed59dcc80b30cb02777fc (patch)
treea5ef576f74640dd835ce943e09688877fe9ce3d6 /module
parentc86a48a92f65ea499cc5becfb24c0de5ee23e035 (diff)
downloadguile-38ad264979b06535200ed59dcc80b30cb02777fc.tar.gz
PEG: string-peg: Better support for escaping.
* module/ice-9/peg/string-peg.scm (peg-as-peg): Augment with rules for hexadecimal digits, “\uXXX” for characters, “\t” for tabs, etc. Signed-off-by: Ludovic Courtès <ludo@gnu.org>
Diffstat (limited to 'module')
-rw-r--r--module/ice-9/peg/string-peg.scm26
1 files changed, 23 insertions, 3 deletions
diff --git a/module/ice-9/peg/string-peg.scm b/module/ice-9/peg/string-peg.scm
index 745d8e8e7..9891f2ae5 100644
--- a/module/ice-9/peg/string-peg.scm
+++ b/module/ice-9/peg/string-peg.scm
@@ -67,9 +67,10 @@ Literal <-- SQUOTE (!SQUOTE Char)* SQUOTE Spacing
NotInClass <-- OPENBRACKET NOTIN (!CLOSEBRACKET Range)* CLOSEBRACKET Spacing
Class <-- OPENBRACKET !NOTIN (!CLOSEBRACKET Range)* CLOSEBRACKET Spacing
Range <-- Char DASH Char / Char
-Char <-- '\\\\' [nrt'\"\\[\\]\\\\]
+Char <-- '\\\\' [nrtf'\"\\[\\]\\\\]
/ '\\\\' [0-7][0-7][0-7]
/ '\\\\' [0-7][0-7]?
+ / '\\\\' 'u' HEX HEX HEX HEX
/ !'\\\\' .
# NOTE: `<--` and `<` are extensions
@@ -79,6 +80,7 @@ DQUOTE < [\"]
DASH < '-'
OPENBRACKET < '['
CLOSEBRACKET < ']'
+HEX <- [0-9a-fA-F]
NOTIN < '^'
SLASH < '/' Spacing
AND <-- '&' Spacing
@@ -92,7 +94,7 @@ DOT <-- '.' Spacing
Spacing < (Space / Comment)*
Comment < '#' (!EndOfLine .)* EndOfLine
-Space < ' ' / '\t' / EndOfLine
+Space < ' ' / '\\t' / EndOfLine
EndOfLine < '\\r\\n' / '\\n' / '\\r'
EndOfFile < !.
")
@@ -144,12 +146,15 @@ EndOfFile < !.
(define-sexp-parser Range all
(or (and Char DASH Char) Char))
(define-sexp-parser Char all
- (or (and "\\" (or "n" "r" "t" "'" "\"" "[" "]" "\\"))
+ (or (and "\\" (or "n" "r" "t" "f" "'" "\"" "[" "]" "\\"))
(and "\\" (range #\0 #\7) (range #\0 #\7) (range #\0 #\7))
(and "\\" (range #\0 #\7) (? (range #\0 #\7)))
+ (and "\\" "u" HEX HEX HEX HEX)
(and (not-followed-by "\\") peg-any)))
(define-sexp-parser LEFTARROW body
(and (or "<--" "<-" "<") Spacing)) ; NOTE: <-- and < are extensions
+(define-sexp-parser HEX body
+ (or (range #\0 #\9) (range #\a #\f) (range #\A #\F)))
(define-sexp-parser NOTIN none
(and "^"))
(define-sexp-parser SLASH none
@@ -372,12 +377,27 @@ EndOfFile < !.
(* (- (char->integer x) (char->integer #\0)) y))
(reverse (string->list charstr 1))
'(1 8 64)))))
+ ((char=? #\u (string-ref charstr 1))
+ (integer->char
+ (reduce + 0
+ (map
+ (lambda (x y)
+ (* (cond
+ ((char-numeric? x)
+ (- (char->integer x) (char->integer #\0)))
+ ((char-alphabetic? x)
+ (+ 10 (- (char->integer x) (char->integer #\a)))))
+ y))
+ (reverse (string->list (string-downcase charstr) 2))
+ '(1 16 256 4096)))))
(else
(case (string-ref charstr 1)
((#\n) #\newline)
((#\r) #\return)
((#\t) #\tab)
+ ((#\f) #\page)
((#\') #\')
+ ((#\") #\")
((#\]) #\])
((#\\) #\\)
((#\[) #\[))))))