further improve html/mathml import:

accept no-namespace math tag (as per html5) accept mathml-only (files and "paste as html") better handle various imports to wide add many math html entities
texmacs · Dec 28, 2022 · a4f9bc9 · a4f9bc9
1 parent a5cd683
commit a4f9bc9
Show file tree

Hide file tree

Showing 4 changed files with 630 additions and 13 deletions.
diff --git a/TeXmacs/progs/convert/html/htmltm.scm b/TeXmacs/progs/convert/html/htmltm.scm
@@ -352,6 +352,17 @@
   (if (sxhtml-list? (xpath-parent env))
       '()
       '((next-line))))
+
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+;; Unlike previously, Mathml nodes in HTML5 have no namespace prefix 
+;; => xmltm.scm prefixes them with h:
+;; Yet, the import code in mathtm.scm expects m: prefix : replace prefix (hacky)
+;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+
+(define (htmltm-math env a c)
+(with cm `(m:math (@ ,@a) ,(replace-nsprefix-in-stree c "h:" "m:"))
+  `(,(mathtm-as-serial env cm))
+))
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;; MathJax extension
@@ -433,7 +444,7 @@
 (tm-define (htmltm-as-serial root)
   ;; As htmltm, but returns a serial node.
   ;; Actually also initializes the dynamic enviroment.
-  ;; FIXME: move the htmlinitialization elsewhere for symmetry with htmltm.
+  ;; FIXME: move the html initialization elsewhere for symmetry with htmltm.
   (define (sub env)
     (htmltm-serial (htmltm-preserve-space? env)
 		   (htmltm env (cleanup-root env root))))
@@ -565,6 +576,9 @@
   ;;; Extensions
   (mathjax (handler :collapse :inline htmltm-mathjax))
 
+  ;;; Math tag in HTML5 (no namespace prefix)
+  (math (handler :mixed :block htmltm-math))
+
   ;; Tags present in the previous converter
   ;; Unknown: FIG FN NOTE AU LANG PERSON
   ;; ABBREV instead of ABBR
@@ -609,6 +623,13 @@
     (else
       st))) ; unexpected entry
 
+(define (replace-nsprefix-in-stree st from to)
+  (cond ((and (nnull? st) (symbol? st) (string-starts? (symbol->string st) from))
+           (string->symbol (string-append to 
+            (string-drop (symbol->string st) (string-length from))))) 
+        ((list? st) (map (lambda (x) (replace-nsprefix-in-stree x from to)) st))
+        (else st)))
+
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;; Post processing
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

diff --git a/TeXmacs/progs/convert/html/init-html.scm b/TeXmacs/progs/convert/html/init-html.scm
@@ -24,6 +24,7 @@
 	((format-test? s pos "<body") #t)
 	((format-test? s pos "<title") #t)
 	((format-test? s pos "<!doctype html") #t)
+	((format-test? s pos "<math") #t)
 	((format-test? s pos "<?xml ")
 	 (html-recognizes-at? s (format-skip-line s pos)))
 	((format-test? s pos "<!doctype ")