Fix syllabification.

* configure.ac: Add AC_PROG_YACC
* src/ellinika/phoneme.y: New file.
* src/ellinika/yyrename: New file.
* src/ellinika/syllabificator.c: New file.
* src/ellinika/.gitignore: Update.
* src/ellinika/elchr.c (char_info_st): Move to header.
(el_basic_ctype):
(elchr_info): Remove static qualifier.
Return a pointer to const.
(elchr_letter,elchr_phoneme): New functions.
(elchr_diphthong): Remove.
* src/ellinika/elmorph.c (elstr)<phoneme,phoneme_count>: New members.
(_elstr_syllabize): Rewrite.
(invalidate_maps)" New static function.
(_elstr_alloc): Initialize new fields, take function name
as argument, for diagnostic purposes.
(_elstr_print): Rewrite
(deftab): Update.
(elstr-syllable-prop,elstr-syllable)
(_elstr_set_accent,_elstr_set_accent_on_char): Rewrite.
(elstr-char-phoneme,elstr->phonetic-map): New functions.
* src/ellinika/elmorph.h (CHF_DIPH1,CHF_DIPH2): Remove.
(CHF_DIPHTHONG): New flag.
(PHON_.*): New constants.
(phoneme,syllable): New structures.
(char_info_st)<letter,phoneme>: New members.
(elchr_info,elchr_letter)
(elchr_phoneme,phoneme_map)
(syllable_map): New protos.
(elchr_diphthong): Remove protos.
* src/ellinika/elmorph.scm4: Move public definitions
to elmorph-public.scm; include it here.
* src/ellinika/xlat.scm (ellinika:sounds-like): Rewrite as a
wrapper over elstr->soundslike.

Describe Milesian numbers.

* style.css (img.ellinika-img): New class.
* xml/lingua.conf.in (IMAGE): New tag.
* xml/pl/alfabhta.xml: Describe Milesian numbers.

Various fixes.

* data/dbverb.struct: fix a typo in flection.
Use 'sub' theme for pas/sub/aor.
* data/irregular-verbs.xml: Add more verbs.
* scm/conjugator.scm: Various fixes.
* scm/verbop.scm: Accept empty mood and voice declarations.

12 files changed, 1206 insertions, 366 deletions

diff --git a/src/ellinika/.gitignore b/src/ellinika/.gitignore
index 9422f9a..11bf478 100644
--- a/src/ellinika/.gitignore
+++ b/src/ellinika/.gitignore
@@ -3,3 +3,5 @@ cgi.scm
 config.scm
 elmorph.scm
 elmorph.x
+phoneme.c
+phoneme.h
diff --git a/src/ellinika/Makefile.am b/src/ellinika/Makefile.am
index 274eea8..b8988d4 100644
--- a/src/ellinika/Makefile.am
+++ b/src/ellinika/Makefile.am
@@ -1,5 +1,5 @@
 # This file is part of Ellinika project.
-# Copyright (C) 2004,2006,2007,2008 Sergey Poznyakoff
+# Copyright (C) 2004,2006,2007,2008,2011 Sergey Poznyakoff
 #
 # Ellinika is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -15,7 +15,14 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 guiledir=$(GUILE_SITE)/$(PACKAGE)
-guile_DATA=xlat.scm cgi.scm i18n.scm config.scm dico.scm elmorph.scm
+guile_DATA=\
+ xlat.scm\
+ cgi.scm\
+ i18n.scm\
+ config.scm\
+ dico.scm\
+ elmorph.scm\
+ tenses.scm
 
 cgi.m4: Makefile
         echo 'divert(-1)' > $@
@@ -39,11 +46,11 @@ cgi.m4: Makefile
 SUFFIXES = .scm4 .scm .x
 
 .scm4.scm:
-        m4 cgi.m4 $< > $@ 
+        m4 -I$(srcdir) cgi.m4 $< > $@ 
 
 cgi.scm: cgi.scm4 cgi.m4
 config.scm: config.scm4 cgi.m4
-elmorph.scm: elmorph.scm4 cgi.m4
+elmorph.scm: elmorph.scm4 elmorph-public.scm cgi.m4
 
 pkglib_LTLIBRARIES=libelmorph.la
 
@@ -52,7 +59,9 @@ libelmorph_la_SOURCES = \
  utf8.c\
  elchr.c\
  elmorph.c\
- elmorph.h
+ elmorph.h\
+ phoneme.y\
+ syllabificator.c
 
 DOT_X_FILES = elmorph.x
 
@@ -80,4 +89,6 @@ install-data-hook:
          done; \
          cd $$here
 
-
+AM_YFLAGS = -d
+YACCCOMPILE = $(srcdir)/yyrename '$(YACC) $(YFLAGS) $(AM_YFLAGS)'
+EXTRA_DIST = yyrename elmorph-public.scm
\ No newline at end of file
diff --git a/src/ellinika/elchr.c b/src/ellinika/elchr.c
index 3142b6f..621ac03 100644
--- a/src/ellinika/elchr.c
+++ b/src/ellinika/elchr.c
@@ -23,20 +23,7 @@
 #include <libguile.h>
 #include "utf8.h"
 #include "elmorph.h"
-
-struct char_info_st {
-        unsigned ch;           /* Characters */
-        int flags;             /* Flags (see above) */
-        unsigned base;         /* for vowels - a corresponding vowel with all diacritics
-                                  removed */
-        unsigned trans;        /* a counter-case equivalent, i.e. a corresponding uppercase
-                                  letter if flags & CHF_LOWER and a corresponding lowerrcase
-                                  letter if flags & CHF_UPPER */
-        unsigned numval;       /* Numeric value */
-        unsigned accented[3];  /* For vowels - corresponding accented variant */
-        unsigned deaccent;     /* For accented vowels with diaeresis - corresponding
-                                  non-accented character */
-};
+#include "phoneme.h"
 
 /* See http://www.unicode.org/charts/PDF/Unicode-5.1/U51-0370.pdf */
 struct char_info_st el_basic_ctype[] = {
@@ -174,80 +161,149 @@ struct char_info_st el_basic_ctype[] = {
         { 0x0383, },
         { 0x0384, CHF_MODIFIER }, /* Oxeia */
         { 0x0385, CHF_MODIFIER }, /* dialytika */
-        { 0x0386, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x0391, 0x03AC }, /* Ά */
+        { 0x0386, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x0391, 0x03AC,
+          0, { 0, 0, 0}, 0, LETTER_A_ACC, PHON_A }, /* Ά */
         { 0x0387, CHF_PUNCT }, /* ano teleia */
-        { 0x0388, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x0395, 0x03AD }, /* Έ */
-        { 0x0389, CHF_VOWEL|CHF_UPPER|CHF_OXEIA|CHF_DIPH2, 0x0397, 0x03AE }, /* Ή */
-        { 0x038A, CHF_VOWEL|CHF_UPPER|CHF_OXEIA|CHF_DIPH2, 0x0399, 0x03AF }, /* Ί */
+        { 0x0388, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x0395, 0x03AD,
+          0, { 0, 0, 0}, 0, LETTER_E_ACC, PHON_E }, /* Έ */
+        { 0x0389, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x0397, 0x03AE,
+          0, { 0, 0, 0}, 0, LETTER_H_ACC, PHON_I }, /* Ή */
+        { 0x038A, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x0399, 0x03AF,
+          0, { 0, 0, 0}, 0, LETTER_I_ACC, PHON_I }, /* Ί */
         { 0x038B, },
-        { 0x038C, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x039F, 0x03CC }, /* Ό */
+        { 0x038C, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x039F, 0x03CC,
+          0, { 0, 0, 0}, 0, LETTER_OMICRON_ACC, PHON_O }, /* Ό */
         { 0x038D, },
-        { 0x038E, CHF_VOWEL|CHF_UPPER|CHF_OXEIA|CHF_DIPH2, 0x03A5, 0x03CD }, /* Ύ */
-        { 0x038F, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x03A9, 0x03CE }, /* Ώ */
-        { 0x0390, CHF_VOWEL|CHF_LOWER|CHF_TREMA|CHF_OXEIA, 0x03B9, 0, 0, 0, 0, 0x03CA }, /* ΐ */
-        { 0x0391, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC|CHF_DIPH1, 0, 0x03B1, 1, 0x0386 }, /* Α */
-        { 0x0392, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B2, 2 }, /* Β */
-        { 0x0393, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B3, 3 }, /* Γ */
-        { 0x0394, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B4, 4 }, /* Δ */
-        { 0x0395, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC|CHF_DIPH1, 0, 0x03B5, 5, 0x0388 }, /* Ε */
-        { 0x0396, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B6, 7 }, /* Ζ */
-        { 0x0397, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03B7, 8, 0x0389 }, /* Η */
-        { 0x0398, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B8, 9 }, /* Θ */
-        { 0x0399, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC|CHF_DIPH2, 0, 0x03B9, 10, 0x038A }, /* Ι */
-        { 0x039A, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BA, 20 }, /* Κ */
-        { 0x039B, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BB, 30 }, /* Λ */
-        { 0x039C, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BC, 40 }, /* Μ */
-        { 0x039D, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BD, 50 }, /* Ν */
-        { 0x039E, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BE, 60 }, /* Ξ */
-        { 0x039F, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03BF, 70, 0x038C }, /* Ο */
-        { 0x03A0, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C0, 80 }, /* Π */
-        { 0x03A1, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C1, 100 }, /* Ρ */
+        { 0x038E, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x03A5, 0x03CD,
+          0, { 0, 0, 0}, 0, LETTER_Y_ACC, PHON_I }, /* Ύ */
+        { 0x038F, CHF_VOWEL|CHF_UPPER|CHF_OXEIA, 0x03A9, 0x03CE,
+          0, { 0, 0, 0}, 0, LETTER_OMEGA_ACC, PHON_O }, /* Ώ */
+        { 0x0390, CHF_VOWEL|CHF_LOWER|CHF_TREMA|CHF_OXEIA, 0x03B9, 0,
+          0, { 0, 0, 0}, 0x03CA, LETTER_I_TREMA_ACC, PHON_I }, /* ΐ */
+        { 0x0391, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03B1,
+          1, { 0x0386, 0, 0}, 0, LETTER_A, PHON_A }, /* Α */
+        { 0x0392, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B2,
+          2, {0, 0, 0}, 0, LETTER_B, PHON_BH },/* Β */
+        { 0x0393, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B3,
+          3, {0, 0, 0}, 0, LETTER_G, PHON_GH }, /* Γ */
+        { 0x0394, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B4,
+          4, {0, 0, 0}, 0, LETTER_D, PHON_DH }, /* Δ */
+        { 0x0395, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03B5,
+          5, { 0x0388, 0, 0}, 0, LETTER_E, PHON_E }, /* Ε */
+        { 0x0396, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B6,
+          7, {0, 0, 0}, 0, LETTER_Z, PHON_Z }, /* Ζ */
+        { 0x0397, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03B7,
+          8, {0x0389, 0, 0}, 0, LETTER_H, PHON_I }, /* Η */
+        { 0x0398, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03B8,
+          9, {0, 0, 0}, 0, LETTER_TH, PHON_TH }, /* Θ */
+        { 0x0399, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03B9,
+          10, { 0x038A, 0, 0}, 0, LETTER_I, PHON_I }, /* Ι */
+        { 0x039A, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BA,
+          20, {0, 0, 0}, 0, LETTER_K, PHON_K }, /* Κ */
+        { 0x039B, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BB,
+          30, {0, 0, 0}, 0, LETTER_L, PHON_L }, /* Λ */
+        { 0x039C, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BC,
+          40, {0, 0, 0}, 0, LETTER_M, PHON_M }, /* Μ */
+        { 0x039D, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BD,
+          50, {0, 0, 0}, 0, LETTER_N, PHON_N }, /* Ν */
+        { 0x039E, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03BE,
+          60, {0, 0, 0}, 0, LETTER_KS, PHON_X }, /* Ξ */
+        { 0x039F, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03BF,
+          70, { 0x038C, 0, 0}, 0, LETTER_OMICRON, PHON_O }, /* Ο */
+        { 0x03A0, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C0,
+          80, {0, 0, 0}, 0, LETTER_P, PHON_P }, /* Π */
+        { 0x03A1, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C1,
+          100, {0, 0, 0}, 0, LETTER_R, PHON_R }, /* Ρ */
         { 0x03A2, },
-        { 0x03A3, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C3, 200 }, /* Σ */
-        { 0x03A4, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C4, 300 }, /* Τ */
-        { 0x03A5, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC|CHF_DIPH1|CHF_DIPH2, 0, 0x03C5, 400, 0x038E }, /* Υ */
-        { 0x03A6, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C6, 500 }, /* Φ */
-        { 0x03A7, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C7, 600 }, /* Χ */
-        { 0x03A8, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C8, 700 }, /* Ψ */
-        { 0x03A9, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03C9, 800, 0x038F }, /* Ω */
-        { 0x03AA, CHF_VOWEL|CHF_UPPER|CHF_TREMA|CHF_DIPH2, 0x0399, 0x03CA }, /* Ϊ */
+        { 0x03A3, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C3,
+          200, {0, 0, 0}, 0, LETTER_S, PHON_S }, /* Σ */
+        { 0x03A4, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C4,
+          300, {0, 0, 0}, 0, LETTER_T, PHON_T }, /* Τ */
+        { 0x03A5, CHF_VOWEL|CHF_UPPER|CHF_NUMERIC, 0, 0x03C5,
+          400, { 0x038E, 0, 0}, 0, LETTER_Y, PHON_I }, /* Υ */
+        { 0x03A6, CHF_CONSONANT|CHF_UPPER|CHF_NUMERIC, 0, 0x03C6,
+          500, {0, 0, 0}, 0, LETTER_F, PHON_F }, /* Φ */