diff --git a/manual/luatex-languages.tex b/manual/luatex-languages.tex
index 990a8dcc038d568ccf74fe8cc6686ab5bd8cfc38..54a7b390d5ffac254c3eba712a693703ba6be912 100644
--- a/manual/luatex-languages.tex
+++ b/manual/luatex-languages.tex
@@ -263,7 +263,82 @@ The word end is determined as follows:
\NC adjust \NC when hyphenationbounds 2 or 3 \NC \NR
\stoptabulate
-(Future versions of \LUATEX\ might provide more granularity.)
+% (Future versions of \LUATEX\ might provide more granularity.)
+
+In traditional \TEX\ ligature building and hyphenation are interwoven with the
+line break mechanism. In \LUATEX\ these phases are isolated. As a consequence we
+deal differently with (a sequence of) explicit hyphens. We already have added
+some control over aspects of the hyphenation and yet another one concerns
+automatic hyphens (e.g.\ \type {-} characters in the input).
+
+When \type {\automatichyphenmode} has a value of 0, a hyphen will be turned into
+an automatic discretionary. The snippets before and after it will not be
+hyphenated. A side effect is that a leading hyphen can lead to a split but one
+will seldom run into that situation. Setting a pre and post character makes this
+more prominent. A value of 1 will prevent this side effect and a value of 2 will
+not turn the hyphen into a discretionary. Experiments with other options, like
+permitting hyphenation, of the words on both sides were discarded.
+
+\startbuffer[a]
+before-after \par
+before--after \par
+before---after \par
+\stopbuffer
+
+\startbuffer[b]
+-before \par
+after- \par
+--before \par
+after-- \par
+---before \par
+after--- \par
+\stopbuffer
+
+\startbuffer[c]
+before-after \par
+before--after \par
+before---after \par
+\stopbuffer
+
+We show three samples:
+
+Input A: \typebuffer[a]
+Input B: \typebuffer[b]
+Input C: \typebuffer[c]
+
+\startbuffer[demo]
+\startcombination[nx=4,ny=3,location=top]
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\zerocount \hsize6em \getbuffer[a]}} {A~0~6em}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\zerocount \hsize2pt \getbuffer[a]}} {A~0~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\plusone \hsize2pt \getbuffer[a]}} {A~1~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\plustwo \hsize2pt \getbuffer[a]}} {A~2~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\zerocount \hsize6em \getbuffer[b]}} {B~0~6em}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\zerocount \hsize2pt \getbuffer[b]}} {B~0~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\plusone \hsize2pt \getbuffer[b]}} {B~1~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\plustwo \hsize2pt \getbuffer[b]}} {B~2~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\zerocount \hsize6em \getbuffer[c]}} {C~0~6em}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\zerocount \hsize2pt \getbuffer[c]}} {C~0~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\plusone \hsize2pt \getbuffer[c]}} {C~1~2pt}
+ {\framed[align=normal,strut=no,top=\vskip.5ex,bottom=\vskip.5ex]{\automatichyphenmode\plustwo \hsize2pt \getbuffer[c]}} {C~2~2pt}
+\stopcombination
+\stopbuffer
+
+\startplacefigure[reference=automatic:1,title={The automatic modes \type {0} (default), \type {1} and \type {2}, with a \type {\hsize}
+of 6em and 2pt (which triggers a linebreak).}]
+ \dontcomplain \tt \getbuffer[demo]
+\stopplacefigure
+
+\startplacefigure[reference=automatic:2,title={The automatic modes \type {0} (default), \type {1} and \type {2}, with \type
+{\preexhyphenchar} and \type {\postexhyphenchar} set to characters \type {A} and \type {B}.}]
+ \postexhyphenchar`A\relax
+ \preexhyphenchar `B\relax
+ \dontcomplain \tt \getbuffer[demo]
+\stopplacefigure
+
+As with primitive companions of other single character commands, the \type {\-}
+command has a more verbose primitive version in \type {\explicitdiscretionary}
+and the normally intercepted in the hyphenator character \type {-} (or whatever
+is configured) is available as \type {\automaticdiscretionary}.
\section{The main control loop}
diff --git a/manual/luatex-nodes.tex b/manual/luatex-nodes.tex
index 21f1789e5a99b08462d7b78c7b516a2876fec246..d3c2c06dae1a3bb51a4354cb144ebeb10517f3ed 100644
--- a/manual/luatex-nodes.tex
+++ b/manual/luatex-nodes.tex
@@ -301,11 +301,15 @@ a \type {userskip} with subtype zero).
\starttabulate[|lT|l|p|]
\NC \rmbf field \NC \bf type \NC \bf explanation \NC \NR
-\NC subtype \NC number \NC not used \NC \NR
+\NC subtype \NC number \NC \showsubtypes{penalty} \NC \NR
\NC attr \NC node \NC list of attributes \NC \NR
\NC penalty \NC number \NC the penalty value \NC \NR
\stoptabulate
+The subtypes are just informative and \TEX\ itself doesn't use them. When you
+run into an \type {linebreakpenalty} you need to keep in mind that it's a
+accumulation of \type {club}, \type{widow} and other relevant penalties.
+
\subsubsection[glyphnodes]{glyph nodes}
\starttabulate[|lT|l|p|]
@@ -1705,6 +1709,13 @@ tail node.
node.slide(<node> n)
\stopfunctioncall
+After some callbacks automatic sliding takes place. This feature can be turned
+off with \type {node.fix_node_lists(false)} but you better make sure then that
+you don't mess up lists. In most cases \TEX\ itself only uses \type {next}
+pointers but your other callbacks might expect proper \type {prev} pointers too.
+Future versions of \LUATEX\ can add more checking but this will not influence
+usage.
+
\subsubsection{\type {node.check_discretionary} and \type {node.check_discretionaries}}
When you fool around with disc nodes you need to be aware of the fact that they
diff --git a/manual/luatex-tex.tex b/manual/luatex-tex.tex
index 650f3ca056db95185732b5c2b9c25e2c7c6f9a47..fac1af39353a0809729542d727ce3cb525605f63 100644
--- a/manual/luatex-tex.tex
+++ b/manual/luatex-tex.tex
@@ -161,7 +161,7 @@ The current list is:
\NC pool_ptr \NC string pool index \NC \NR
\NC pool_size \NC current size allocated for string characters \NC \NR
\NC save_size \NC save stack size \NC \NR
-\NC shell_escape \NC \type {0} means disabled, \type {1} means anything is permitted and \type {2} is restricted \NC \NR
+\NC shell_escape \NC \type {0} means disabled, \type {1} means anything is permitted and \type {2} is restricted \NC \NR
\NC safer_option \NC \type {1} means safer is enforced \NC \NR
\NC kpse_used \NC \type {1} means that kpse is used \NC \NR
\NC stack_size \NC input stack size \NC \NR
diff --git a/manual/luatex.pdf b/manual/luatex.pdf
index 006bfc7856ebcbc182574062de7e8c5b984448ac..277bf24122390d3909133f74766222663d38f1de 100644
Binary files a/manual/luatex.pdf and b/manual/luatex.pdf differ
diff --git a/manual/luatex.tex b/manual/luatex.tex
index 1ccc9b7a59a39552d9eaac9a2a7bafcd5174e6a6..1613b920b206b2b2c5a64f9b3369792a370c5db0 100644
--- a/manual/luatex.tex
+++ b/manual/luatex.tex
@@ -29,8 +29,8 @@
\dontcomplain
\startdocument
- [status=intermediate release,
- version=1.0.4]
+ [status=release,
+ version=1.0.5]
\component luatex-titlepage
diff --git a/source/texk/web2c/luatexdir/font/luafont.w b/source/texk/web2c/luatexdir/font/luafont.w
index 4e874d9becb896585da678362505845b75366417..d7a5c5f904d3c4f6ff5970724e36c767fc417df4 100644
--- a/source/texk/web2c/luatexdir/font/luafont.w
+++ b/source/texk/web2c/luatexdir/font/luafont.w
@@ -1995,8 +1995,9 @@ halfword handle_ligaturing(halfword head, halfword tail)
vlink(tail) = null;
}
- /* |if (fix_node_lists)| */
- fix_node_list(head);
+ if (fix_node_lists) {
+ fix_node_list(head);
+ }
prev = head;
cur = vlink(prev);
@@ -2184,7 +2185,9 @@ static halfword run_lua_ligkern_callback(halfword head, halfword tail, int callb
luatex_error(Luas, (i == LUA_ERRRUN ? 0 : 1));
return tail;
}
- fix_node_list(head);
+ if (fix_node_lists) {
+ fix_node_list(head);
+ }
lua_settop(Luas, top);
return tail;
}
diff --git a/source/texk/web2c/luatexdir/lang/hyphen.w b/source/texk/web2c/luatexdir/lang/hyphen.w
index 8e13896090d769a68b025dba26403c1a7bf58d84..35befb743debbf7e34b4429727f7a3edc25b6d8d 100644
--- a/source/texk/web2c/luatexdir/lang/hyphen.w
+++ b/source/texk/web2c/luatexdir/lang/hyphen.w
@@ -776,6 +776,8 @@ void hnj_hyphen_load(HyphenDict * dict, const unsigned char *f)
}
@ @c
+extern halfword insert_syllable_discretionary(halfword t, lang_variables * lan);
+
void hnj_hyphen_hyphenate(HyphenDict * dict,
halfword first1,
halfword last1,
@@ -791,9 +793,9 @@ void hnj_hyphen_hyphenate(HyphenDict * dict,
char *hyphens = hnj_malloc(hyphen_len + 1);
/* Add a '.' to beginning and end to facilitate matching */
- set_vlink(begin_point, first1);
- set_vlink(end_point, get_vlink(last1));
- set_vlink(last1, end_point);
+ vlink(begin_point) = first1;
+ vlink(end_point) = vlink(last1);
+ vlink(last1) = end_point;
for (char_num = 0; char_num < hyphen_len; char_num++) {
hyphens[char_num] = '0';
@@ -801,16 +803,16 @@ void hnj_hyphen_hyphenate(HyphenDict * dict,
hyphens[hyphen_len] = 0;
/* now, run the finite state machine */
- for (char_num = 0, here = begin_point; here != get_vlink(end_point);
- here = get_vlink(here)) {
+ for (char_num = 0, here = begin_point; here != vlink(end_point);
+ here = vlink(here)) {
int ch;
if (here == begin_point || here == end_point) {
ch = '.';
} else {
- ch = get_hj_code(char_lang(here),get_character(here));
+ ch = get_hj_code(char_lang(here),character(here));
if (ch <= 32) {
- ch = get_character(here);
+ ch = character(here);
}
}
while (state != -1) {
@@ -857,14 +859,14 @@ void hnj_hyphen_hyphenate(HyphenDict * dict,
}
/* restore the correct pointers */
- set_vlink(last1, get_vlink(end_point));
+ vlink(last1) = vlink(end_point);
/* pattern is \.{\^.\^w\^o\^r\^d\^.\^} |word_len|=4, |ext_word_len|=6, |hyphens|=7
* check \.{ \^ \^ \^ } so drop first two and stop after |word_len-1|
*/
- for (here = first1, char_num = 2; here != left; here = get_vlink(here))
+ for (here = first1, char_num = 2; here != left; here = vlink(here))
char_num++;
- for (; here != right; here = get_vlink(here)) {
+ for (; here != right; here = vlink(here)) {
if (hyphens[char_num] & 1)
here = insert_syllable_discretionary(here, lan);
char_num++;
diff --git a/source/texk/web2c/luatexdir/lang/texlang.h b/source/texk/web2c/luatexdir/lang/texlang.h
index 5e7a046195f2c092253b36bcd0d43a6a1de1c31e..0fa11d4c82f79f7b2905298c92965d904c12452a 100644
--- a/source/texk/web2c/luatexdir/lang/texlang.h
+++ b/source/texk/web2c/luatexdir/lang/texlang.h
@@ -67,7 +67,7 @@ extern int get_post_exhyphen_char(int lan);
extern void set_hyphenation_min(int lan, int val);
extern int get_hyphenation_min(int lan);
-extern halfword compound_word_break(halfword t, int clang);
+/* extern halfword compound_word_break(halfword t, int clang); */
extern void dump_language_data(void);
extern void undump_language_data(void);
@@ -82,5 +82,7 @@ extern void new_post_exhyphen_char(void);
extern void new_hyphenation_min(void);
extern void new_hj_code(void);
+extern void set_disc_field(halfword f, halfword t);
+extern halfword insert_syllable_discretionary(halfword t, lang_variables * lan);
#endif
diff --git a/source/texk/web2c/luatexdir/lang/texlang.w b/source/texk/web2c/luatexdir/lang/texlang.w
index bd519af632ab8192a136c0ce982805ebcc717be8..337d2eca51dfebcfc08bc7806e177c8380d309ad 100644
--- a/source/texk/web2c/luatexdir/lang/texlang.w
+++ b/source/texk/web2c/luatexdir/lang/texlang.w
@@ -60,7 +60,8 @@ struct tex_language *new_language(int n)
lang->post_exhyphen_char = 0;
lang->hyphenation_min = -1;
if (saving_hyph_codes_par) {
- hj_codes_from_lc_codes(l); /* for now, we might just use specific value for whatever task */
+ /* for now, we might just use specific value for whatever task */
+ hj_codes_from_lc_codes(l);
}
return lang;
} else {
@@ -198,10 +199,10 @@ void load_tex_patterns(int curlang, halfword head)
const char *clean_hyphenation(int id, const char *buff, char **cleaned)
{
int items = 0;
- unsigned char word[MAX_WORD_LEN + 1]; /* work buffer for bytes */
- unsigned uword[MAX_WORD_LEN + 1] = { 0 }; /* work buffer for unicode */
- int u = 0; /* unicode buffer value */
- int i = 0; /* index into buffer */
+ unsigned char word[MAX_WORD_LEN + 1]; /* work buffer for bytes */
+ unsigned uword[MAX_WORD_LEN + 1] = { 0 }; /* work buffer for unicode */
+ int u = 0; /* unicode buffer value */
+ int i = 0; /* index into buffer */
char *uindex = (char *)word;
const char *s = buff;
@@ -218,12 +219,12 @@ const char *clean_hyphenation(int id, const char *buff, char **cleaned)
/* now convert the input to unicode */
word[i] = '\0';
utf2uni_strcpy(uword, (const char *)word);
-
/* build the new word string */
i = 0;
while (uword[i]>0) {
u = uword[i++];
- if (u == '-') { /* skip */
+ if (u == '-') {
+ /* skip */
} else if (u == '=') {
STORE_CHAR(id,'-');
} else if (u == '{') {
@@ -253,7 +254,7 @@ const char *clean_hyphenation(int id, const char *buff, char **cleaned)
if (u == '}') {
items++;
}
- if (items != 3) { /* syntax error */
+ if (items != 3) {
*cleaned = NULL;
tex_error("exception syntax error", NULL);
return s;
@@ -322,51 +323,61 @@ void load_tex_hyphenation(int curlang, halfword head)
}
@ @c
-halfword insert_discretionary(halfword t, halfword pre, halfword post, halfword replace, int penalty)
+static halfword insert_discretionary(halfword t, halfword pre, halfword post, halfword replace, int penalty)
{
- halfword g, n;
+ halfword g;
int f;
- n = new_node(disc_node, syllable_disc);
- disc_penalty(n) = penalty;
- try_couple_nodes(n, vlink(t));
- couple_nodes(t, n);
- if (replace != null)
+ halfword d = new_node(disc_node, syllable_disc);
+ int attr = node_attr(t) ;
+ disc_penalty(d) = penalty;
+ if (t == replace) {
+ /* prev disc next-next */
+ try_couple_nodes(d, vlink(t));
+ try_couple_nodes(alink(t), d);
+ alink(t) = null;
+ vlink(t) = null;
+ replace = t ;
+ } else {
+ /* prev disc next */
+ try_couple_nodes(d, vlink(t));
+ couple_nodes(t, d);
+ }
+ if (replace != null) {
f = font(replace);
- else
- f = get_cur_font(); /* for compound words following explicit hyphens */
+ } else {
+ /* For compound words following explicit hyphens. */
+ f = get_cur_font();
+ }
for (g = pre; g != null; g = vlink(g)) {
font(g) = f;
- if (node_attr(t) != null) {
+ if (attr != null) {
delete_attribute_ref(node_attr(g));
- node_attr(g) = node_attr(t);
- attr_list_ref(node_attr(t)) += 1;
+ node_attr(g) = attr;
+ attr_list_ref(attr) += 1;
}
}
for (g = post; g != null; g = vlink(g)) {
font(g) = f;
- if (node_attr(t) != null) {
+ if (attr != null) {
delete_attribute_ref(node_attr(g));
- node_attr(g) = node_attr(t);
- attr_list_ref(node_attr(t)) += 1;
+ node_attr(g) = attr;
+ attr_list_ref(attr) += 1;
}
}
- for (g = replace; g != null; g = vlink(g)) {
- if (node_attr(t) != null) {
+ if (attr != null) {
+ for (g = replace; g != null; g = vlink(g)) {
delete_attribute_ref(node_attr(g));
- node_attr(g) = node_attr(t);
- attr_list_ref(node_attr(t)) += 1;
+ node_attr(g) = attr;
+ attr_list_ref(attr) += 1;
}
+ delete_attribute_ref(node_attr(d));
+ node_attr(d) = attr;
+ attr_list_ref(attr) += 1;
}
- if (node_attr(t) != null) {
- delete_attribute_ref(node_attr(vlink(t)));
- node_attr(vlink(t)) = node_attr(t);
- attr_list_ref(node_attr(t)) += 1;
- }
- t = vlink(t);
- set_disc_field(pre_break(t), pre);
- set_disc_field(post_break(t), post);
- set_disc_field(no_break(t), replace);
- return t;
+ set_disc_field(pre_break(d), pre);
+ set_disc_field(post_break(d), post);
+ set_disc_field(no_break(d), replace);
+ return d;
}
halfword insert_syllable_discretionary(halfword t, lang_variables * lan)
@@ -413,43 +424,38 @@ halfword insert_syllable_discretionary(halfword t, lang_variables * lan)
}
@ @c
-halfword compound_word_break(halfword t, int clang)
+static halfword insert_character(halfword t, int c)
+{
+ halfword p;
+ p = new_node(glyph_node, 0);
+ set_to_character(p);
+ character(p) = c;
+ if (t != null) {
+ couple_nodes(t, p);
+ }
+ return p;
+}
+
+static halfword compound_word_break(halfword t, int clang)
{
halfword disc = null;
halfword pre = null;
halfword pos = null;
halfword pre_exhyphen_char = get_pre_exhyphen_char(clang);
halfword post_exhyphen_char = get_post_exhyphen_char(clang);
- if (pre_exhyphen_char > 0)
+ if (pre_exhyphen_char > 0) {
pre = insert_character(null,pre_exhyphen_char);
+ } else {
+ pre = insert_character(null,ex_hyphen_char_par);
+ }
if (post_exhyphen_char > 0)
pos = insert_character(null,post_exhyphen_char);
- disc = insert_discretionary(t,pre,pos,null,ex_hyphen_penalty_par);
+ disc = insert_discretionary(t,pre,pos,t,ex_hyphen_penalty_par);
subtype(disc) = automatic_disc;
set_automatic_disc_penalty(disc);
return disc;
}
-halfword insert_complex_discretionary(halfword t, lang_variables * lan,
- halfword pre, halfword pos,
- halfword replace)
-{
- (void) lan;
- return insert_discretionary(t,pre,pos,replace,hyphen_penalty_par);
-}
-
-halfword insert_character(halfword t, int c)
-{
- halfword p;
- p = new_node(glyph_node, 0);
- set_to_character(p);
- character(p) = c;
- if (t != null) {
- couple_nodes(t, p);
- }
- return p;
-}
-
@ @c
void set_disc_field(halfword f, halfword t)
{
@@ -713,6 +719,20 @@ there was not the best idea ever.
*/
+/*
+ We only accept an explicit hyphen when there is a preceding glyph and we skip a sequence of
+ explicit hyphens as that normally indicates a -- or --- ligature in which case we can in a
+ worse case usage get bad node lists later on due to messed up ligature building as these
+ dashes are ligatures in base fonts. This is a side effect of the separating the hyphenation,
+ ligaturing and kerning steps. A test is cmr with ------.
+
+ A font handler can collapse successive hyphens but it's not nice to put the burden there. A
+ somewhat messy border case is ---- but in LuaTeX we don't treat -- and --- special. Also,
+ traditional TeX will break a line at -foo but this can be disabled by setting the automatic
+ mode to 1.
+
+*/
+
static halfword find_next_wordstart(halfword r, halfword first_language, halfword strict_bound)
{
register int l;
@@ -757,27 +777,31 @@ static halfword find_next_wordstart(halfword r, halfword first_language, halfwor
if (is_simple_character(r)) {
chr = character(r) ;
if (chr == ex_hyphen_char_par) {
- /*
- We only accept an explicit hyphen when there is a preceding glyph and
- we skip a sequence of explicit hyphens as that normally indicates a
- -- or --- ligature in which case we can in a worse case usage get bad
- node lists later on due to messed up ligature building as these dashes
- are ligatures in base fonts. This is a side effect of the separating the
- hyphenation, ligaturing and kerning steps. A test is cmr with ------.
- */
t = vlink(r) ;
- if ((start_ok == 0) && (t!=null) && (type(t) == glyph_node) && (character(t) != ex_hyphen_char_par)) {
- compound_word_break(r, char_lang(r));
- start_ok = 1 ;
+ if ((automatic_hyphen_mode_par == 0) && (t != null) && (type(t) == glyph_node) && (character(t) != ex_hyphen_char_par)) {
+ /* we have no word yet and the next character is a non hyphen */
+ r = compound_word_break(r, char_lang(r));
} else {
- start_ok = 0;
+ /* we jump over the sequence of hyphens */
+ while ((t != null) && (type(t) == glyph_node) && (character(t) == ex_hyphen_char_par)) {
+ r = t ;
+ t = vlink(r) ;
+ }
+ if (t == null) {
+ /* we reached the end of the list so we have no word start */
+ return null;
+ }
}
+ /* we need a restart */
+ start_ok = 0;
} else if (start_ok && (char_lang(r)>=first_language) && ((l = get_hj_code(char_lang(r),chr)) > 0)) {
if (char_uchyph(r) || l == chr || l <= 32) {
return r;
} else {
start_ok = 0;
}
+ } else {
+ /* go on */
}
}
break;
@@ -839,18 +863,23 @@ void hnj_hyphenation(halfword head, halfword tail)
halfword strict_bound = hyphenation_bounds_par;
halfword s, r = head, wordstart = null, save_tail1 = null, left = null, right = null;
- /* this first movement assures two things:
- \item{a)} that we won't waste lots of time on something that has been
- handled already (in that case, none of the glyphs match |simple_character|).
- \item{b)} that the first word can be hyphenated. if the movement was
- not explicit, then the indentation at the start of a paragraph
- list would make |find_next_wordstart()| look too far ahead.
- */
+ /*
+ This first movement assures two things:
+
+ \item{a)} that we won't waste lots of time on something that has been handled already
+ (in that case, none of the glyphs match |simple_character|).
+
+ \item{b)} that the first word can be hyphenated. if the movement was not explicit,
+ then the indentation at the start of a paragraph list would make |find_next_wordstart()|
+ look too far ahead.
+ */
while (r != null && (type(r) != glyph_node || !is_simple_character(r))) {
r = vlink(r);
}
- /* this will make |r| a glyph node with subtype character */
+ /*
+ This will make |r| a glyph node with subtype character.
+ */
r = find_next_wordstart(r,first_language,strict_bound);
if (r == null)
return;
@@ -860,15 +889,17 @@ void hnj_hyphenation(halfword head, halfword tail)
s = new_penalty(0,word_penalty);
couple_nodes(tail, s);
- while (r != null) { /* could be while(1), but let's be paranoid */
+ while (r != null) { /* could be while(1), but let's be paranoid */
int clang, lhmin, rhmin, hmin;
halfword hyf_font;
halfword end_word = r;
wordstart = r;
assert(is_simple_character(wordstart));
hyf_font = font(wordstart);
- if (hyphen_char(hyf_font) < 0) /* for backward compat */
+ if (hyphen_char(hyf_font) < 0) {
+ /* For backward compatibility: */
hyf_font = 0;
+ }
clang = char_lang(wordstart);
lhmin = char_lhmin(wordstart);
rhmin = char_rhmin(wordstart);
@@ -889,6 +920,7 @@ void hnj_hyphenation(halfword head, halfword tail)
) {
if (character(r) == ex_hyphen_char_par) {
explicit_hyphen = true;
+ break;
}
wordlen++;
if (lchar <= 32) {
@@ -911,12 +943,27 @@ void hnj_hyphenation(halfword head, halfword tail)
lchar = character(r) ;
}
hy = uni2string(hy, (unsigned) lchar);
- /* this should not be needed any more */
- /*if (vlink(r)!=null) alink(vlink(r))=r; */
end_word = r;
r = vlink(r);
}
- if ( valid_wordend(r,strict_bound)
+ if (explicit_hyphen == true) {
+ /* we are not at the start, so we only need to look ahead */
+ halfword t = vlink(r) ;
+ if ((automatic_hyphen_mode_par == 0 || automatic_hyphen_mode_par == 1) && (t != null) && ((type(t) == glyph_node) && (character(t) != ex_hyphen_char_par))) {
+ /* we have a word already but the next character may not be a hyphen too */
+ r = compound_word_break(r, char_lang(r));
+ } else {
+ /* we jump over the sequence of hyphens */
+ while ((t != null) && (type(t) == glyph_node) && (character(t) == ex_hyphen_char_par)) {
+ r = t ;
+ t = vlink(r) ;
+ }
+ if (t == null) {
+ /* we reached the end of the list and will quit the loop later */
+ r = null;
+ }
+ }
+ } else if ( valid_wordend(r,strict_bound)
&& clang >= first_language
&& wordlen >= lhmin + rhmin
&& (hmin <= 0 || wordlen >= hmin)
@@ -924,35 +971,10 @@ void hnj_hyphenation(halfword head, halfword tail)
&& (lang = tex_languages[clang]) != NULL
) {
*hy = 0;
- if ( lang->exceptions != 0
- && (replacement = hyphenation_exception(lang->exceptions, utf8word)) != NULL
- ) {
-#ifdef VERBOSE
- formatted_warning("hyphenation","replacing %s (c=%d) by %s", utf8word, clang, replacement);
-#endif
+ if (lang->exceptions != 0 && (replacement = hyphenation_exception(lang->exceptions, utf8word)) != NULL) {
+ /* handle the exception and go on to the next word */
do_exception(wordstart, r, replacement);
free(replacement);
- } else if (explicit_hyphen == true) {
- /*
- insert an explicit discretionary after each of the last in a
- set of explicit hyphens
- */
- halfword rr = r;
-#ifdef VERBOSE
- formatted_warning("hyphenation","explicit hyphen(s) found in %s (c=%d)", utf8word, clang);
-#endif
- while (rr != wordstart) {
- if (is_simple_character(rr)) {
- if (character(rr) == ex_hyphen_char_par) {
- compound_word_break(rr, clang);
- while (character(alink(rr)) == ex_hyphen_char_par)
- rr = alink(rr);
- if (rr == wordstart)
- break;
- }
- }
- rr = alink(rr);
- }
} else if (lang->patterns != NULL) {
left = wordstart;
for (i = lhmin; i > 1; i--) {
@@ -960,10 +982,7 @@ void hnj_hyphenation(halfword head, halfword tail)
while (!is_simple_character(left)) {
left = vlink(left);
}
- /*
- if (!left)
- break ;
- */
+ /* if (!left) break; */
/* what is left overruns right .. a bit messy */
}
right = r;
@@ -972,20 +991,10 @@ void hnj_hyphenation(halfword head, halfword tail)
while (!is_simple_character(right)) {
right = alink(right);
}
- /*
- if (!right)
- break ;
- */
+ /* if (!right) break; */
/* what is right overruns left .. a bit messy */
}
- /* maybe an extra check ... */
- /* |if (left && right) {| */
-#ifdef VERBOSE
- formatted_warning("hyphenation","hyphenate %s (c=%d,l=%d,r=%d) from %c to %c",
- utf8word, clang, lhmin, rhmin, character(left), character(right));
-#endif
- (void) hnj_hyphen_hyphenate(lang->patterns, wordstart, end_word, wordlen, left, right, &langdata);
- /* |}| */
+ (void) hnj_hyphen_hyphenate(lang->patterns, wordstart, end_word, wordlen, left, right, &langdata);
}
}
explicit_hyphen = false;
@@ -1026,7 +1035,7 @@ void new_hyphenation(halfword head, halfword tail)
}
}
-@ dumping and undumping languages
+@ Dumping and undumping languages.
@c
#define dump_string(a) \
@@ -1139,20 +1148,19 @@ void new_hyph_exceptions(void)
flush_list(def_ref);
}
-@ Similarly, when \TeX\ has scanned `\.{\\patterns}', it calls on a
-procedure named |new_patterns|.
+@ Similarly, when \TeX\ has scanned `\.{\\patterns}', it calls on a procedure named
+|new_patterns|.
@c
void new_patterns(void)
-{ /* initializes the hyphenation pattern data */
+{
(void) scan_toks(false, true);
load_tex_patterns(language_par, def_ref);
flush_list(def_ref);
}
-@ `\.{\\prehyphenchar}', sets the |pre_break| character, and
-`\.{\\posthyphenchar}' the |post_break| character. Their respective defaults are
-ascii hyphen ("-") and zero (nul).
+@ `\.{\\prehyphenchar}', sets the |pre_break| character, and `\.{\\posthyphenchar}' the
+|post_break| character. Their respective defaults are ascii hyphen ("-") and zero (nul).
@c
void new_pre_hyphen_char(void)
@@ -1169,9 +1177,8 @@ void new_post_hyphen_char(void)
set_post_hyphen_char(language_par, cur_val);
}
-@ `\.{\\preexhyphenchar}', sets the |pre_break| character, and
-`\.{\\postexhyphenchar}' the |post_break| character. Their defaults are both zero
-(nul).
+@ `\.{\\preexhyphenchar}', sets the |pre_break| character, and `\.{\\postexhyphenchar}' the
+|post_break| character. Their defaults are both zero (nul).
@c
void new_pre_exhyphen_char(void)
diff --git a/source/texk/web2c/luatexdir/lua/liolibext.c b/source/texk/web2c/luatexdir/lua/liolibext.c
index e910139c2ce5c0cf74b10fbde3df963eb3e07c62..2e09c4541d40ad5c2313f4f2e039706d9e6cfa9c 100644
--- a/source/texk/web2c/luatexdir/lua/liolibext.c
+++ b/source/texk/web2c/luatexdir/lua/liolibext.c
@@ -29,11 +29,9 @@
#endif
#include "lualib.h"
-
-
#ifdef LuajitTeX
/* luajit has its own way for io, which is a mix of */
-/* lua 5.1 and lua 5.2 . We use the stock luajit. */
+/* lua 5.1 and lua 5.2 . We use the stock luajit. */
#else
/*
** {======================================================
@@ -61,7 +59,6 @@
/* }====================================================== */
-
#if defined(LUA_USE_POSIX)
#define l_fseek(f,o,w) fseeko(f,o,w)
@@ -92,9 +89,6 @@
#endif/* #ifdef LuajitTeX */
-
-
-
static FILE *tofile (lua_State *L) {
#ifdef LuajitTeX
FILE **f = luaL_checkudata(L,1,LUA_FILEHANDLE);
@@ -110,6 +104,8 @@ static FILE *tofile (lua_State *L) {
#endif
}
+#define uchar(c) ((unsigned char)(c))
+
/*
HH: A few helpers to avoid reading numbers as strings. For now we put them in their
own namespace. We also have a few helpers that can make io functions tex friendly.
@@ -125,6 +121,19 @@ static int readcardinal1(lua_State *L) {
return 1;
}
+static int readcardinal1_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p]);
+ lua_pushinteger(L, a);
+ }
+ return 1;
+}
+
static int readcardinal2(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -137,6 +146,20 @@ static int readcardinal2(lua_State *L) {
return 1;
}
+static int readcardinal2_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+1 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p]);
+ lua_pushinteger(L, 0x100 * a + b);
+ }
+ return 1;
+}
+
static int readcardinal3(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -150,6 +173,21 @@ static int readcardinal3(lua_State *L) {
return 1;
}
+static int readcardinal3_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+2 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p++]);
+ int c = uchar(s[p]);
+ lua_pushinteger(L, 0x10000 * a + 0x100 * b + c);
+ }
+ return 1;
+}
+
static int readcardinal4(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -164,6 +202,22 @@ static int readcardinal4(lua_State *L) {
return 1;
}
+static int readcardinal4_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+3 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p++]);
+ int c = uchar(s[p++]);
+ int d = uchar(s[p]);
+ lua_pushinteger(L,0x1000000 * a + 0x10000 * b + 0x100 * c + d);
+ }
+ return 1;
+}
+
static int readinteger1(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -176,6 +230,22 @@ static int readinteger1(lua_State *L) {
return 1;
}
+static int readinteger1_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p]);
+ if (a >= 0x80)
+ lua_pushinteger(L, a - 0x100);
+ else
+ lua_pushinteger(L, a);
+ }
+ return 1;
+}
+
static int readinteger2(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -189,6 +259,23 @@ static int readinteger2(lua_State *L) {
return 1;
}
+static int readinteger2_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+1 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p]);
+ if (a >= 0x80)
+ lua_pushinteger(L, 0x100 * a + b - 0x10000);
+ else
+ lua_pushinteger(L, 0x100 * a + b);
+ }
+ return 1;
+}
+
static int readinteger3(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -203,6 +290,24 @@ static int readinteger3(lua_State *L) {
return 1;
}
+static int readinteger3_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+2 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p++]);
+ int c = uchar(s[p]);
+ if (a >= 0x80)
+ lua_pushinteger(L, 0x10000 * a + 0x100 * b + c - 0x1000000);
+ else
+ lua_pushinteger(L, 0x10000 * a + 0x100 * b + c);
+ }
+ return 1;
+}
+
static int readinteger4(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -218,6 +323,25 @@ static int readinteger4(lua_State *L) {
return 1;
}
+static int readinteger4_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+3 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p++]);
+ int c = uchar(s[p++]);
+ int d = uchar(s[p]);
+ if (a >= 0x80)
+ lua_pushinteger(L, 0x1000000 * a + 0x10000 * b + 0x100 * c + d - 0x100000000);
+ else
+ lua_pushinteger(L, 0x1000000 * a + 0x10000 * b + 0x100 * c + d);
+ }
+ return 1;
+}
+
static int readfixed2(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -225,9 +349,26 @@ static int readfixed2(lua_State *L) {
if (b == EOF)
lua_pushnil(L);
else if (a >= 0x80)
- lua_pushinteger(L, a + b/0xFFFF - 0x100);
+ lua_pushinteger(L, (a - 0x100) + b/0x100);
else
- lua_pushinteger(L, a + b/0xFFFF);
+ lua_pushinteger(L, (a ) + b/0x100);
+ return 1;
+}
+
+static int readfixed2_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+3 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p]);
+ if (a >= 0x80)
+ lua_pushinteger(L, (a - 0x100) + b/0x100);
+ else
+ lua_pushinteger(L, (a ) + b/0x100);
+ }
return 1;
}
@@ -240,15 +381,34 @@ static int readfixed4(lua_State *L) {
if (d == EOF)
lua_pushnil(L);
else if (a >= 0x80)
- lua_pushnumber(L, (0x1000000 * a + 0x10000 * b + 0x100 * c + d - 0x100000000)/65536.0);
+ lua_pushnumber(L, (0x100 * a + b - 0x10000) + (0x100 * c + d)/0x10000);
else
- lua_pushnumber(L, (0x1000000 * a + 0x10000 * b + 0x100 * c + d)/65536.0);
+ lua_pushnumber(L, (0x100 * a + b ) + (0x100 * c + d)/0x10000);
/* from ff */
/* int n = 0x1000000 * a + 0x10000 * b + 0x100 * c + d; */
/* lua_pushnumber(L,(real) (n>>16) + ((n&0xffff)/65536.0)); */
return 1;
}
+static int readfixed4_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+3 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p++]);
+ int c = uchar(s[p++]);
+ int d = uchar(s[p]);
+ if (a >= 0x80)
+ lua_pushnumber(L, (0x100 * a + b - 0x10000) + (0x100 * c + d)/0x10000);
+ else
+ lua_pushnumber(L, (0x100 * a + b ) + (0x100 * c + d)/0x10000);
+ }
+ return 1;
+}
+
static int read2dot14(lua_State *L) {
FILE *f = tofile(L);
int a = getc(f);
@@ -263,6 +423,21 @@ static int read2dot14(lua_State *L) {
return 1;
}
+static int read2dot14_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ if (p < 0 || p+1 >= l) {
+ lua_pushnil(L);
+ } else {
+ int a = uchar(s[p++]);
+ int b = uchar(s[p]);
+ int n = 0x100 * a + b;
+ lua_pushnumber(L,(real) ((n<<16)>>(16+14)) + ((n&0x3fff)/16384.0));
+ }
+ return 1;
+}
+
static int getposition(lua_State *L) {
FILE *f = tofile(L);
long p = ftell(f);
@@ -317,6 +492,28 @@ static int readbytetable(lua_State *L) {
return 1;
}
+static int readbytetable_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ int n = lua_tointeger(L,3);
+ if (p < 0 || p >= l) {
+ lua_pushnil(L);
+ } else {
+ int i ;
+ if (p + n >= l) {
+ n = l - p ;
+ }
+ lua_createtable(L, n, 0);
+ for (i=1;i<=n;i++) {
+ int a = uchar(s[p++]);
+ lua_pushinteger(L, a);
+ lua_rawseti(L,-2,i);
+ }
+ }
+ return 1;
+}
+
static int readbytes(lua_State *L) {
FILE *f = tofile(L);
int n = lua_tointeger(L,2);
@@ -332,6 +529,27 @@ static int readbytes(lua_State *L) {
return n;
}
+static int readbytes_s(lua_State *L) {
+ size_t l;
+ const char *s = luaL_checklstring(L, 1, &l);
+ size_t p = luaL_checkinteger(L, 2) - 1;
+ int n = lua_tointeger(L,3);
+ if (p < 0 || p >= l) {
+ return 0;
+ } else {
+ int i ;
+ if (p + n >= l) {
+ n = l - p ;
+ }
+ lua_createtable(L, n, 0);
+ for (i=1;i<=n;i++) {
+ int a = uchar(s[p++]);
+ lua_pushinteger(L, a);
+ }
+ return n;
+ }
+}
+
static int recordfilename(lua_State *L)
{
const char *fname = luaL_checkstring(L, 1);
@@ -448,7 +666,35 @@ static const luaL_Reg fiolib[] = {
{NULL, NULL}
};
+static const luaL_Reg siolib[] = {
+ { "readcardinal1", readcardinal1_s },
+ { "readcardinal2", readcardinal2_s },
+ { "readcardinal3", readcardinal3_s },
+ { "readcardinal4", readcardinal4_s },
+ { "readinteger1", readinteger1_s },
+ { "readinteger2", readinteger2_s },
+ { "readinteger3", readinteger3_s },
+ { "readinteger4", readinteger4_s },
+ { "readfixed2", readfixed2_s },
+ { "readfixed4", readfixed4_s },
+ { "read2dot14", read2dot14_s },
+ { "readbytes", readbytes_s },
+ { "readbytetable", readbytetable_s },
+ /* done */
+ {NULL, NULL}
+};
+
+/*
+ The sio helpers are experimental and might be handy at some point. Speed-wise
+ there is no gain over file access because with ssd and caching we basically
+ operate in memory too. We keep them as complement to the file ones. I did
+ consider using an userdata object for the position etc but some simple tests
+ demonstrated that there is no real gain and the current ones permits to wrap
+ up whatever interface one likes.
+*/
+
int luaopen_fio(lua_State *L) {
luaL_register(L, "fio", fiolib);
+ luaL_register(L, "sio", siolib);
return 1;
}
diff --git a/source/texk/web2c/luatexdir/lua/lnodelib.c b/source/texk/web2c/luatexdir/lua/lnodelib.c
index 49929fa27926e5c1db04dfd2f75f6ddc7c328f20..eeb93b3a09b7b38b67300c8e5211ac6eb3e1f4d4 100644
--- a/source/texk/web2c/luatexdir/lua/lnodelib.c
+++ b/source/texk/web2c/luatexdir/lua/lnodelib.c
@@ -268,6 +268,19 @@ static int get_valid_node_subtype_id(lua_State * L, int n)
return i;
}
+/* enable fixer ... more fixers can be added in the future */
+
+static int lua_nodelib_fix_node_lists(lua_State * L)
+{
+ if (lua_type(L, -1) == LUA_TBOOLEAN) {
+ fix_node_lists = lua_toboolean(L,-1);
+ } else if (lua_type(L, -1) == LUA_TNUMBER) {
+ fix_node_lists = lua_tointeger(L,-1);
+ }
+ return 0;
+}
+
+
/* two simple helpers to speed up and simplify lua code (replaced by getnext and getprev) */
static int lua_nodelib_next(lua_State * L)
@@ -3198,7 +3211,7 @@ static int lua_nodelib_set_glue(lua_State * L)
stretch_order(n) = ((top > 4 && lua_type(L, 5) == LUA_TNUMBER)) ? lua_tointeger(L,5) : 0;
shrink_order(n) = ((top > 5 && lua_type(L, 6) == LUA_TNUMBER)) ? lua_tointeger(L,6) : 0;
}
- return 0;
+ return 0;
}
return luaL_error(L, "glue (spec) expected");
}
@@ -7588,6 +7601,99 @@ static int lua_nodelib_check_discretionary(lua_State * L) {
return 0;
}
+/* synctex but not */
+
+static int lua_nodelib_set_synctex_mode(lua_State * L)
+{
+ halfword mode = lua_tointeger(L, 1);
+ synctex_set_mode(mode);
+ return 0;
+}
+
+static int lua_nodelib_set_synctex_tag(lua_State * L)
+{
+ halfword tag = lua_tointeger(L, 1);
+ synctex_set_tag(tag);
+ return 0;
+}
+
+static int lua_nodelib_set_synctex_line(lua_State * L)
+{
+ halfword line = lua_tointeger(L, 1);
+ synctex_set_line(line);
+ return 0;
+}
+
+static int lua_nodelib_direct_set_synctex_fields(lua_State * L)
+{
+ halfword n = lua_tointeger(L, 1);
+ halfword tag = lua_tointeger(L, 2);
+ halfword line = lua_tointeger(L, 3);
+ if (n != null) {
+ switch (type(n)) {
+ case math_node:
+ if (tag) synctex_tag_math(n) = tag;
+ if (line) synctex_line_math(n) = line;
+ break;
+ case glue_node:
+ if (tag) synctex_tag_glue(n) = tag;
+ if (line) synctex_line_glue(n) = line;
+ break;
+ case kern_node:
+ if (tag) synctex_tag_kern(n) = tag;
+ if (line) synctex_line_kern(n) = line;
+ break;
+ case hlist_node:
+ case vlist_node:
+ case unset_node:
+ if (tag) synctex_tag_box(n) = tag;
+ if (line) synctex_line_box(n) = line;
+ break;
+ case rule_node:
+ if (tag) synctex_tag_rule(n) = tag;
+ if (line) synctex_line_rule(n) = line;
+ break;
+ }
+ }
+ return 0;
+}
+
+static int lua_nodelib_direct_get_synctex_fields(lua_State * L)
+{
+ halfword n = lua_tointeger(L, 1);
+ if (n != null) {
+ switch (type(n)) {
+ case math_node:
+ lua_pushinteger(L,synctex_tag_math(n));
+ lua_pushinteger(L,synctex_line_math(n));
+ break;
+ case glue_node:
+ lua_pushinteger(L,synctex_tag_glue(n));
+ lua_pushinteger(L,synctex_line_glue(n));
+ break;
+ case kern_node:
+ lua_pushinteger(L,synctex_tag_kern(n));
+ lua_pushinteger(L,synctex_line_kern(n));
+ break;
+ case hlist_node:
+ case vlist_node:
+ case unset_node:
+ lua_pushinteger(L,synctex_tag_box(n));
+ lua_pushinteger(L,synctex_line_box(n));
+ break;
+ case rule_node:
+ lua_pushinteger(L,synctex_tag_rule(n));
+ lua_pushinteger(L,synctex_line_rule(n));
+ break;
+ }
+ return 2;
+ } else {
+ return 0;
+ }
+}
+
+/* done */
+
static const struct luaL_Reg nodelib_p[] = {
{"__index", lua_nodelib_get_property_t},
{"__newindex", lua_nodelib_set_property_t},
@@ -7733,6 +7839,9 @@ static const struct luaL_Reg direct_nodelib_f[] = {
{"check_discretionary", lua_nodelib_direct_check_discretionary},
{"check_discretionaries", lua_nodelib_direct_check_discretionaries},
/* done */
+ {"set_synctex_fields", lua_nodelib_direct_set_synctex_fields},
+ {"get_synctex_fields", lua_nodelib_direct_get_synctex_fields},
+ /* done */
{NULL, NULL} /* sentinel */
};
@@ -7824,6 +7933,13 @@ static const struct luaL_Reg nodelib_f[] = {
{"check_discretionary", lua_nodelib_check_discretionary},
{"check_discretionaries", lua_nodelib_check_discretionaries},
/* done */
+ {"set_synctex_mode", lua_nodelib_set_synctex_mode},
+ {"set_synctex_tag", lua_nodelib_set_synctex_tag},
+ {"set_synctex_line", lua_nodelib_set_synctex_line},
+ /* {"set_synctex_fields", lua_nodelib_set_synctex_fields}, */
+ /* {"get_synctex_fields", lua_nodelib_get_synctex_fields}, */
+ /* done */
+ {"fix_node_lists", lua_nodelib_fix_node_lists},
{NULL, NULL} /* sentinel */
};
diff --git a/source/texk/web2c/luatexdir/lua/luanode.w b/source/texk/web2c/luatexdir/lua/luanode.w
index af15748bacd6392a217ea520da44347505c32870..c09cac1ff01b4ecef0c8bd89805110ea02070384 100644
--- a/source/texk/web2c/luatexdir/lua/luanode.w
+++ b/source/texk/web2c/luatexdir/lua/luanode.w
@@ -96,21 +96,18 @@ void lua_node_filter(int filterid, int extrainfo, halfword head_node, halfword *
start_node = vlink(head_node);
if (start_node != null) {
/* maybe just always slide (harmless and fast) */
- last_node = vlink(start_node);
if (fix_node_lists) {
- while (last_node != null) {
- alink(last_node) = start_node;
- start_node = last_node;
- last_node = vlink(start_node);
- }
+ /* slides and returns last node */
+ *tail_node = fix_node_list(start_node);
} else {
+ last_node = vlink(start_node);
while (last_node != null) {
start_node = last_node;
last_node = vlink(start_node);
}
+ /* we're at the end now */
+ *tail_node = start_node;
}
- /* we're at the end now */
- *tail_node = start_node;
} else {
/* we're already at the end */
*tail_node = head_node;
diff --git a/source/texk/web2c/luatexdir/luatex.c b/source/texk/web2c/luatexdir/luatex.c
index 3eba56679b8d1e9e2d088b321f031c55cd0f9509..fac1851d9eafcbc15efd5c71500200b97511a1c7 100644
--- a/source/texk/web2c/luatexdir/luatex.c
+++ b/source/texk/web2c/luatexdir/luatex.c
@@ -29,9 +29,9 @@
#define TeX
int luatex_version = 100; /* \.{\\luatexversion} */
-int luatex_revision = '4'; /* \.{\\luatexrevision} */
-int luatex_date_info = 2017033109; /* the compile date is now hardwired :YEAR MONTH DAY HOUR*/
-const char *luatex_version_string = "1.0.4";
+int luatex_revision = '5'; /* \.{\\luatexrevision} */
+int luatex_date_info = 2017033110; /* the compile date is now hardwired :YEAR MONTH DAY HOUR*/
+const char *luatex_version_string = "1.0.5";
const char *engine_name = my_name; /* the name of this engine */
#include <kpathsea/c-ctype.h>
diff --git a/source/texk/web2c/luatexdir/tex/commands.w b/source/texk/web2c/luatexdir/tex/commands.w
index 84235914c64a269e989dcd12f6e2611feeea99c1..27c5ac9cea644792abdfb32958209e5bbfc65a25 100644
--- a/source/texk/web2c/luatexdir/tex/commands.w
+++ b/source/texk/web2c/luatexdir/tex/commands.w
@@ -160,6 +160,7 @@ void initialize_commands(void)
primitive_luatex("hyphenpenaltymode", assign_int_cmd, int_base + hyphen_penalty_mode_code, int_base);
primitive_luatex("automatichyphenpenalty", assign_int_cmd, int_base + automatic_hyphen_penalty_code, int_base);
primitive_luatex("explicithyphenpenalty", assign_int_cmd, int_base + explicit_hyphen_penalty_code, int_base);
+ primitive_luatex("automatichyphenmode", assign_int_cmd, int_base + automatic_hyphen_mode_code, int_base);
/* Many of \TeX's primitives need no |equiv|, since they are identifiable
by their |eq_type| alone. These primitives are loaded into the hash table
@@ -445,8 +446,10 @@ void initialize_commands(void)
primitive_tex("unhcopy", un_hbox_cmd, copy_code, 0);
primitive_tex("unvbox", un_vbox_cmd, box_code, 0);
primitive_tex("unvcopy", un_vbox_cmd, copy_code, 0);
- primitive_tex("-", discretionary_cmd, explicit_disc, 0);
+ primitive_tex("-", discretionary_cmd, explicit_disc, 0); /* good old tex */
primitive_tex("discretionary", discretionary_cmd, discretionary_disc, 0);
+ primitive_luatex("explicitdiscretionary", discretionary_cmd, explicit_disc, 0);
+ primitive_luatex("automaticdiscretionary", discretionary_cmd, automatic_disc, 0);
primitive_luatex("localleftbox", assign_local_box_cmd, 0, 0);
primitive_luatex("localrightbox", assign_local_box_cmd, 1, 0);
diff --git a/source/texk/web2c/luatexdir/tex/dumpdata.w b/source/texk/web2c/luatexdir/tex/dumpdata.w
index aa0d1655d6dce960eee38bf71180f03efd350167..4f8b1fc896a08e9af5bd3ee09a686d474c983861 100644
--- a/source/texk/web2c/luatexdir/tex/dumpdata.w
+++ b/source/texk/web2c/luatexdir/tex/dumpdata.w
@@ -23,7 +23,7 @@
/* we start with 907: the sum of the values of the bytes of "don knuth" */
-#define FORMAT_ID (907+28)
+#define FORMAT_ID (907+29)
#if ((FORMAT_ID>=0) && (FORMAT_ID<=256))
#error Wrong value for FORMAT_ID.
#endif
diff --git a/source/texk/web2c/luatexdir/tex/equivalents.h b/source/texk/web2c/luatexdir/tex/equivalents.h
index d7ca036e25afa8caa832a79e3d0526b7c7396441..a853f55872c37c5a28ea786947e747758430a6c0 100644
--- a/source/texk/web2c/luatexdir/tex/equivalents.h
+++ b/source/texk/web2c/luatexdir/tex/equivalents.h
@@ -291,6 +291,7 @@ the |number_regs| \.{\\dimen} registers.
# define hyphen_penalty_mode_code 100
# define automatic_hyphen_penalty_code 101
# define explicit_hyphen_penalty_code 102
+# define automatic_hyphen_mode_code 103
# define math_option_code (explicit_hyphen_penalty_code+1)
@@ -773,6 +774,7 @@ extern halfword last_cs_name;
#define hyphen_penalty_mode_par int_par(hyphen_penalty_mode_code)
#define automatic_hyphen_penalty_par int_par(automatic_hyphen_penalty_code)
#define explicit_hyphen_penalty_par int_par(explicit_hyphen_penalty_code)
+#define automatic_hyphen_mode_par int_par(automatic_hyphen_mode_code)
#define cur_lang_par int_par(cur_lang_code)
#define cur_font_par equiv(cur_font_loc)
diff --git a/source/texk/web2c/luatexdir/tex/maincontrol.w b/source/texk/web2c/luatexdir/tex/maincontrol.w
index 0baba872e8e90622fa2d8a8f6bb165d832259936..1d8b020cf0a2f1d19c4de3d6112f872a42d03bab 100644
--- a/source/texk/web2c/luatexdir/tex/maincontrol.w
+++ b/source/texk/web2c/luatexdir/tex/maincontrol.w
@@ -1759,18 +1759,42 @@ void append_discretionary(void)
if (cur_chr == explicit_disc) {
/* \- */
c = get_pre_hyphen_char(cur_lang_par);
- if (c != 0) {
+ if (c > 0) {
vlink(pre_break(tail)) = new_char(equiv(cur_font_loc), c);
alink(vlink(pre_break(tail))) = pre_break(tail);
tlink(pre_break(tail)) = vlink(pre_break(tail));
}
c = get_post_hyphen_char(cur_lang_par);
- if (c != 0) {
+ if (c > 0) {
vlink(post_break(tail)) = new_char(equiv(cur_font_loc), c);
alink(vlink(post_break(tail))) = post_break(tail);
tlink(post_break(tail)) = vlink(post_break(tail));
}
set_explicit_disc_penalty(tail);
+ } else if (cur_chr == automatic_disc) {
+ /* - as done in hyphenator */
+ c = get_pre_exhyphen_char(cur_lang_par);
+ if (c <= 0) {
+ c = ex_hyphen_char_par;
+ }
+ if (c > 0) {
+ vlink(pre_break(tail)) = new_char(equiv(cur_font_loc), c);
+ alink(vlink(pre_break(tail))) = pre_break(tail);
+ tlink(pre_break(tail)) = vlink(pre_break(tail));
+ }
+ c = get_post_exhyphen_char(cur_lang_par);
+ if (c > 0) {
+ vlink(post_break(tail)) = new_char(equiv(cur_font_loc), c);
+ alink(vlink(post_break(tail))) = post_break(tail);
+ tlink(post_break(tail)) = vlink(post_break(tail));
+ }
+ c = ex_hyphen_char_par;
+ if (c > 0) {
+ vlink(no_break(tail)) = new_char(equiv(cur_font_loc), c);
+ alink(vlink(no_break(tail))) = no_break(tail);
+ tlink(no_break(tail)) = vlink(no_break(tail));
+ }
+ set_automatic_disc_penalty(tail);
} else {
/* \discretionary */
if (scan_keyword("penalty")) {
diff --git a/source/texk/web2c/luatexdir/tex/texfileio.w b/source/texk/web2c/luatexdir/tex/texfileio.w
index 2281de12c4c858652b47a963712898c5e9ba46ec..c1610d10d904353743b98053ca593235b7e21e37 100644
--- a/source/texk/web2c/luatexdir/tex/texfileio.w
+++ b/source/texk/web2c/luatexdir/tex/texfileio.w
@@ -967,8 +967,9 @@ void start_input(void)
update_terminal();
istate = new_line;
/* Prepare new file {\sl Sync\TeX} information */
- synctexstartinput(); /* Give control to the {\sl Sync\TeX} controller */
-
+ if (synctex_par) {
+ synctexstartinput(); /* Give control to the {\sl Sync\TeX} controller */
+ }
/* Read the first line of the new file */
/* Here we have to remember to tell the |lua_input_ln| routine not to
start with a |get|. If the file is empty, it is considered to
diff --git a/source/texk/web2c/luatexdir/tex/texnodes.h b/source/texk/web2c/luatexdir/tex/texnodes.h
index b838dd0a2f95570e01d5fd5575de8d8d3ae2b8be..bc21d52603313c653b272251fa13940c998e6606 100644
--- a/source/texk/web2c/luatexdir/tex/texnodes.h
+++ b/source/texk/web2c/luatexdir/tex/texnodes.h
@@ -23,19 +23,6 @@
#ifndef __NODES_H__
# define __NODES_H__
-/* these are in texlang.c */
-
-# define set_vlink(a,b) vlink(a)=b
-# define get_vlink(a) vlink(a)
-# define get_character(a) character(a)
-
-extern halfword insert_discretionary(halfword t, halfword pre, halfword post, halfword replace, int penalty);
-extern halfword insert_syllable_discretionary(halfword t, lang_variables * lan);
-extern halfword insert_word_discretionary(halfword t, lang_variables * lan);
-extern halfword insert_complex_discretionary(halfword t, lang_variables * lan, halfword pre, halfword post, halfword replace);
-extern halfword insert_character(halfword t, int n);
-extern void set_disc_field(halfword f, halfword t);
-
# define varmemcast(a) (memory_word *)(a)
extern memory_word *volatile varmem;
@@ -928,7 +915,7 @@ extern halfword do_copy_node_list(halfword, halfword);
extern halfword copy_node_list(halfword);
extern halfword copy_node(const halfword);
extern void check_node(halfword);
-extern void fix_node_list(halfword);
+extern halfword fix_node_list(halfword);
extern int fix_node_lists;
extern char *sprint_node_mem_usage(void);
extern halfword raw_glyph_node(void);
@@ -1013,5 +1000,9 @@ extern int lua_properties_use_metatable ;
extern halfword make_local_par_node(int mode);
+extern void synctex_set_mode(int mode);
+extern void synctex_set_tag(int tag);
+extern void synctex_set_line(int line);
+
#endif
diff --git a/source/texk/web2c/luatexdir/tex/texnodes.w b/source/texk/web2c/luatexdir/tex/texnodes.w
index 7967006ec35986ec79b4a9754d2318f6c5675385..6a467e11545c42384ea61fe446e506fbc26727a1 100644
--- a/source/texk/web2c/luatexdir/tex/texnodes.w
+++ b/source/texk/web2c/luatexdir/tex/texnodes.w
@@ -827,6 +827,25 @@ static int copy_error(halfword p)
return 0;
}
+@ @c
+static halfword synctex_anyway_mode = 0;
+static halfword synctex_line_field = 0;
+
+void synctex_set_mode(int m)
+{
+ synctex_anyway_mode = m;
+};
+
+void synctex_set_tag(int t)
+{
+ cur_input.synctex_tag_field = t;
+};
+
+void synctex_set_line(int l)
+{
+ synctex_line_field = l;
+};
+
@ @c
halfword new_node(int i, int j)
{
@@ -892,7 +911,32 @@ halfword new_node(int i, int j)
default:
break;
}
- if (synctex_par) {
+ if (synctex_anyway_mode) {
+ switch (i) {
+ case math_node:
+ synctex_tag_math(n) = cur_input.synctex_tag_field;
+ synctex_line_math(n) = (synctex_line_field) ? synctex_line_field : line;
+ break;
+ case glue_node:
+ synctex_tag_glue(n) = cur_input.synctex_tag_field;
+ synctex_line_glue(n) = (synctex_line_field) ? synctex_line_field : line;
+ break;
+ case kern_node:
+ synctex_tag_kern(n) = cur_input.synctex_tag_field;
+ synctex_line_kern(n) = (synctex_line_field) ? synctex_line_field : line;
+ break;
+ case hlist_node:
+ case vlist_node:
+ case unset_node:
+ synctex_tag_box(n) = cur_input.synctex_tag_field;
+ synctex_line_box(n) = (synctex_line_field) ? synctex_line_field : line;
+ break;
+ case rule_node:
+ synctex_tag_rule(n) = cur_input.synctex_tag_field;
+ synctex_line_rule(n) = (synctex_line_field) ? synctex_line_field : line;
+ break;
+ }
+ } else if (synctex_par) {
/* handle synctex extension */
switch (i) {
case math_node:
@@ -1718,18 +1762,19 @@ void check_node(halfword p)
}
@ @c
-void fix_node_list(halfword head)
+halfword fix_node_list(halfword head)
{
- halfword p, q;
+ halfword next, tail;
if (head == null)
- return;
- p = head;
- q = vlink(p);
- while (q != null) {
- alink(q) = p;
- p = q;
- q = vlink(p);
+ return null;
+ tail = head;
+ next = vlink(head);
+ while (next != null) {
+ alink(next) = tail;
+ tail = next;
+ next = vlink(tail);
}
+ return tail;
}
@ @c