diff options
author | Akira Kakuto <kakuto@fuk.kindai.ac.jp> | 2018-02-03 14:10:32 +0000 |
---|---|---|
committer | Akira Kakuto <kakuto@fuk.kindai.ac.jp> | 2018-02-03 14:10:32 +0000 |
commit | f704f19b7964a4247ef69e6d4a66797e0e90bbb1 (patch) | |
tree | 411d075f5b5815fe02effc41f71011226ac17057 /Build/source | |
parent | 7fa6eb304b185976cfce57891da0e4d1615def95 (diff) |
xetex.web: sync with the upstream
git-svn-id: svn://tug.org/texlive/trunk@46530 c570f23f-e606-0410-a88d-b1316a301751
Diffstat (limited to 'Build/source')
-rw-r--r-- | Build/source/texk/web2c/xetexdir/ChangeLog | 6 | ||||
-rw-r--r-- | Build/source/texk/web2c/xetexdir/xetex.web | 54 |
2 files changed, 33 insertions, 27 deletions
diff --git a/Build/source/texk/web2c/xetexdir/ChangeLog b/Build/source/texk/web2c/xetexdir/ChangeLog index 0b7d1fdfad4..f1cc12e7ae9 100644 --- a/Build/source/texk/web2c/xetexdir/ChangeLog +++ b/Build/source/texk/web2c/xetexdir/ChangeLog @@ -12,6 +12,12 @@ native fonts. Fixes #145. * XeTeX_ext.[ch]: Patch from Hironobu Yamashita to properly close input pipes. Fixes 147. + * xetex.web: Leave text as UTF-16 in trick_buf for better + error context printing. (Mostly) fixes #146. + It's still possible that error context could include half of + a supplementary-plane character (i.e. an unpaired surrogate). + Should be a lot less likely, and less troublesome than broken + UTF-8, however. * XeTeXFontMgr.cpp, XeTeXFontMgr.h: Convert optical size info in OpenType 'size' feature from PostScript to TeX points. Note that this will change behavior for existing documents diff --git a/Build/source/texk/web2c/xetexdir/xetex.web b/Build/source/texk/web2c/xetexdir/xetex.web index 08f825c47b2..83965c769a1 100644 --- a/Build/source/texk/web2c/xetexdir/xetex.web +++ b/Build/source/texk/web2c/xetexdir/xetex.web @@ -1697,6 +1697,9 @@ if @<Character |s| is the current new-line character@> then begin print_visible_char("^"); print_visible_char("^"); print_lc_hex((s mod @"100) div @"10); print_lc_hex(s mod @"10); end + else if selector=pseudo then + print_visible_char(s) { Don't UTF8-encode text in |trick_buf|, + we'll handle that when printing error context. } else begin { |char >= 128|: encode as UTF8 } if s<@"800 then begin @@ -6077,7 +6080,7 @@ A global boolean variable called |no_new_control_sequence| is set to @d prim_prime=431 {about 85\pct! of |primitive_size|} @d prim_base=1 @d prim_next(#) == prim[#].lh {link for coalesced lists} -@d prim_text(#) == prim[#].rh {string number for control sequence name, plus one} +@d prim_text(#) == prim[#].rh {string number for control sequence name} @d prim_is_full == (prim_used=prim_base) {test if all positions are occupied} @d prim_eq_level_field(#)==#.hh.b1 @d prim_eq_type_field(#)==#.hh.b0 @@ -6188,30 +6191,27 @@ var h:integer; {hash code} @!k:pointer; {index in string pool} @!j,@!l:integer; begin -if s<=biggest_char then begin - if s<0 then begin p:=undefined_primitive; goto found; end - else p:=(s mod prim_prime)+prim_base; {we start searching here} - end +if s<256 then begin + p:=s; + if (p<0) or (prim_eq_level(p)<>level_one) then + p:=undefined_primitive; +end else begin j:=str_start_macro(s); if s = str_ptr then l:=cur_length else l:=length(s); @<Compute the primitive code |h|@>; - p:=h+prim_base; {we start searching here; note that |0<=h<prim_prime|} - end; -loop@+begin - if prim_text(p)>1+biggest_char then { |p| points a multi-letter primitive } - begin if length(prim_text(p)-1)=l then - if str_eq_str(prim_text(p)-1,s) then goto found; - end - else if prim_text(p)=1+s then goto found; { |p| points a single-letter primitive } - if prim_next(p)=0 then - begin if no_new_control_sequence then - p:=undefined_primitive - else @<Insert a new primitive after |p|, then make - |p| point to it@>; - goto found; + p:=h+prim_base; {we start searching here; note that |0<=h<hash_prime|} + loop@+begin if prim_text(p)>0 then if length(prim_text(p))=l then + if str_eq_str(prim_text(p),s) then goto found; + if prim_next(p)=0 then + begin if no_new_control_sequence then + p:=undefined_primitive + else @<Insert a new primitive after |p|, then make + |p| point to it@>; + goto found; + end; + p:=prim_next(p); end; - p:=prim_next(p); end; found: prim_lookup:=p; end; @@ -6224,7 +6224,7 @@ begin if prim_text(p)>0 then until prim_text(prim_used)=0; {search for an empty location in |prim|} prim_next(p):=prim_used; p:=prim_used; end; -prim_text(p):=s+1; +prim_text(p):=s; end @ The value of |prim_prime| should be roughly 85\pct! of @@ -6297,7 +6297,7 @@ var k:pool_pointer; {index into |str_pool|} @!prim_val:integer; {needed to fill |prim_eqtb|} begin if s<256 then begin cur_val:=s+single_base; - prim_val:=prim_lookup(s); + prim_val:=s; end else begin k:=str_start_macro(s); l:=str_start_macro(s+1)-k; {we will move |s| into the (possibly non-empty) |buffer|} @@ -7700,11 +7700,11 @@ if l+first_count<=half_error_line then else begin print("..."); p:=l+first_count-half_error_line+3; n:=half_error_line; end; -for q:=p to first_count-1 do print_visible_char(trick_buf[q mod error_line]); +for q:=p to first_count-1 do print_char(trick_buf[q mod error_line]); print_ln; for q:=1 to n do print_visible_char(" "); {print |n| spaces to begin line~2} if m+n<=error_line then p:=first_count+m else p:=first_count+(error_line-n-3); -for q:=first_count to p-1 do print_visible_char(trick_buf[q mod error_line]); +for q:=first_count to p-1 do print_char(trick_buf[q mod error_line]); if m+n>error_line then print("...") @ But the trick is distracting us from our current goal, which is to @@ -8638,7 +8638,7 @@ temporary file. begin save_scanner_status:=scanner_status; scanner_status:=normal; get_token; scanner_status:=save_scanner_status; if cur_cs < hash_base then - cur_cs:=prim_lookup(cur_cs-single_base) + cur_cs:=prim_lookup(cur_cs-257) else cur_cs:=prim_lookup(text(cur_cs)); if cur_cs<>undefined_primitive then begin @@ -11218,7 +11218,7 @@ if_primitive_code: begin get_next; scanner_status:=save_scanner_status; if cur_cs < hash_base then - m:=prim_lookup(cur_cs-single_base) + m:=prim_lookup(cur_cs-257) else m:=prim_lookup(text(cur_cs)); b :=((cur_cmd<>undefined_cs) and @@ -24113,7 +24113,7 @@ any_mode(ignore_spaces): begin get_next; scanner_status:=t; if cur_cs < hash_base then - cur_cs:=prim_lookup(cur_cs-single_base) + cur_cs:=prim_lookup(cur_cs-257) else cur_cs :=prim_lookup(text(cur_cs)); if cur_cs<>undefined_primitive then begin |