[uptex] implement ^^^^xxyy (#170)

texjporg · Jun 23, 2024 · 8d3be7e · 8d3be7e
1 parent 5a74e9a
commit 8d3be7e
Show file tree

Hide file tree

Showing 3 changed files with 48 additions and 6 deletions.
diff --git a/source/texk/web2c/euptexdir/tests/eup-ofm.tex b/source/texk/web2c/euptexdir/tests/eup-ofm.tex
@@ -35,7 +35,7 @@
 \catcode"C6=11  \catcode"D8=11  \catcode"DF=11
 \catcode"E6=11  \catcode"F8=11
 
-\char"C6\char"D8\char"DF\char"E6\char"F8
+\char"C6\char"D8\char"DF^^e6^^^^00f8
 
 ÆØßæø
 
@@ -54,14 +54,14 @@
 \catcode"393=11  \catcode"394=11  \catcode"398=11
 \catcode"39B=11  \catcode"39E=11
 
-\char"393\char"394\char"398\char"39B\char"39E
+\char"393\char"394\char"398\char"39B^^^^039e
 
 ΓΔΘΛΞ
 
 \kcatcode"2010=14
 \catcode"2010=11  \catcode"2018=11  \catcode"2019=11
 
-\char"2010\char"2018\char"2019
+\char"2010\char"2018^^^^2019
 
 ‐‘’
 

diff --git a/source/texk/web2c/uptexdir/tests/up-ofm.tex b/source/texk/web2c/uptexdir/tests/up-ofm.tex
@@ -35,7 +35,7 @@
 \catcode"C6=11  \catcode"D8=11  \catcode"DF=11
 \catcode"E6=11  \catcode"F8=11
 
-\char"C6\char"D8\char"DF\char"E6\char"F8
+\char"C6\char"D8\char"DF^^e6^^^^00f8
 
 ÆØßæø
 
@@ -54,14 +54,14 @@
 \catcode"393=11  \catcode"394=11  \catcode"398=11
 \catcode"39B=11  \catcode"39E=11
 
-\char"393\char"394\char"398\char"39B\char"39E
+\char"393\char"394\char"398\char"39B^^^^039e
 
 ΓΔΘΛΞ
 
 \kcatcode"2010=14
 \catcode"2010=11  \catcode"2018=11  \catcode"2019=11
 
-\char"2010\char"2018\char"2019
+\char"2010\char"2018^^^^2019
 
 ‐‘’
 

diff --git a/source/texk/web2c/uptexdir/uptex-m.ch b/source/texk/web2c/uptexdir/uptex-m.ch
@@ -431,6 +431,14 @@ if ((kcp mod @'10)>0)and(nrestmultichr(kcp)>0) then p:=p-(kcp mod @'10);
       s:=get_avail; info(s):=ktokentochr(info(loc));
 @z
 
+@x
+@!c,@!cc:ASCII_code; {constituents of a possible expanded code}
+@!d:2..3; {number of excess characters in an expanded code}
+@y
+@!c,@!cc,@!cd,@!ce:ASCII_code; {constituents of a possible expanded code}
+@!d:2..7; {number of excess characters in an expanded code}
+@z
+
 @x
   begin cur_chr:=buffer[loc]; incr(loc);
     if multistrlen(ustringcast(buffer), limit+1, loc-1)=2 then
@@ -481,6 +489,40 @@ hangul_code(mid_kanji):
   state:=mid_line;
 @z
 
+@x
+@d hex_to_cur_chr==
+  if c<="9" then cur_chr:=c-"0" @+else cur_chr:=c-"a"+10;
+  if cc<="9" then cur_chr:=16*cur_chr+cc-"0"
+  else cur_chr:=16*cur_chr+cc-"a"+10
+@y
+@d hex_to_cur_chr==
+  if c<="9" then cur_chr:=c-"0" @+else cur_chr:=c-"a"+10;
+  if cc<="9" then cur_chr:=16*cur_chr+cc-"0"
+  else cur_chr:=16*cur_chr+cc-"a"+10
+@d long_hex_to_cur_chr==
+  if c<="9" then cur_chr:=c-"0" @+else cur_chr:=c-"a"+10;
+  if cc<="9" then cur_chr:=16*cur_chr+cc-"0"
+  else cur_chr:=16*cur_chr+cc-"a"+10;
+  if cd<="9" then cur_chr:=16*cur_chr+cd-"0"
+  else cur_chr:=16*cur_chr+cd-"a"+10;
+  if ce<="9" then cur_chr:=16*cur_chr+ce-"0"
+  else cur_chr:=16*cur_chr+ce-"a"+10
+@z
+
+@x
+  begin c:=buffer[loc+1]; @+if c<@'200 then {yes we have an expanded char}
+@y
+  begin if (cur_chr=buffer[loc+1]) and (cur_chr=buffer[loc+2]) and
+           ((loc+6)<=limit) then
+     begin c:=buffer[loc+3]; cc:=buffer[loc+4];
+       cd:=buffer[loc+5]; ce:=buffer[loc+6];
+       if is_hex(c) and is_hex(cc) and is_hex(cd) and is_hex(ce) then
+       begin loc:=loc+7; long_hex_to_cur_chr; goto reswitch;
+       end
+     end;
+  c:=buffer[loc+1]; @+if c<@'200 then {yes we have an expanded char}
+@z
+
 @x
 else  begin k:=loc; cur_chr:=buffer[k]; incr(k);
   if multistrlen(ustringcast(buffer), limit+1, k-1)=2 then