glaemscribe 1.1.14 → 1.2.0

data/glaemresources/modes/{sindarin.glaem → sindarin-tengwar-general_use.glaem}

@@ -35,12 +35,16 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
   \entry "0.1.0" "Added support for the Tengwar Elfica font"
   \entry "0.1.1" "Added support for inlined raw tengwar"  
   \entry "0.1.2" "Added support for non-breaking spaces"  
+  \entry "0.1.3" "Added support for new unicode charsets"   
+  \entry "0.1.4" "Added support for the Tengwar Telcontar font"
+  \entry "0.1.5" "Added a few labial exotic combinations. Reworked median point behaviour, and ng."
+  \entry "0.1.6" "Added gasdil handling."
 \end
 
 \language "Sindarin"
 \writing  "Tengwar"
 \mode     "Sindarin Tengwar - General Use"
-\version  "0.1.2"
+\version  "0.1.6"
 \authors  "J.R.R. Tolkien, impl. Talagan (Benjamin Babut)"
 
 \world      arda
@@ -51,7 +55,15 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 \charset  tengwar_ds_eldamar  false
 \charset  tengwar_ds_annatar  false
 \charset  tengwar_ds_elfica   false
+
+\charset  tengwar_guni_sindarin false
+\charset  tengwar_guni_parmaite false
+\charset  tengwar_guni_eldamar  false
+\charset  tengwar_guni_annatar  false
+\charset  tengwar_guni_elfica   false
+
 \charset  tengwar_freemono    false
+\charset  tengwar_telcontar   false
 
 \raw_mode "raw-tengwar"
 
@@ -62,6 +74,16 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     \value U_UP_O_DOWN 2
   \end
 
+  \beg option apostrophe APOSTROPHE_IGNORED
+    \value APOSTROPHE_IGNORED 0
+    \value APOSTROPHE_GASDIL  1
+  \end
+  
+  \beg option hyphen HYPHEN_WORD_BREAKER
+    \value HYPHEN_WORD_BREAKER  0
+    \value HYPHEN_WORD_JOINER   1
+  \end
+
   \beg option consonant_modification_style CONSONANT_MODIFICATION_STYLE_BAR
     \value CONSONANT_MODIFICATION_STYLE_WAVE 0
     \value CONSONANT_MODIFICATION_STYLE_BAR 1
@@ -85,6 +107,17 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
   \** Work exclusively downcase **\
   \downcase
   
+  \if "hyphen == HYPHEN_WORD_JOINER"
+    \** Replace hyphen by median point **\
+    \substitute "-" "·"  
+  \else
+    \** Replace hyphen by glaemscribe's word breaker **\
+    \substitute "-" "|"
+  \endif
+  
+  \** Add keyboard friendly word joiner **\
+  \substitute "*" "·"
+  
   \** Simplify trema vowels **\
   \substitute ä a
   \substitute ë e
@@ -107,6 +140,17 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
   \** Special case of starting 'i' before vowels, replace i by j **\     
   \rxsubstitute "\\bi([aeouyáāâéēêíīîóōôúūûýȳŷ])" "j\\1"
   
+  \** Special case for ng : before the vast majority of consonnants, treat as ŋ **\
+  \** Don't include r / l / lh / w **\
+  \rxsubstitute "ng([tpckbdfðvnmhs])" "ŋ\\1"
+  
+  \** Avoid mutated ng of being treated as strong middle word n|g (ex : i·ngelaidh [iŋɛlaið] ) **\
+  \substitute "·ng" "·ŋ"  
+  \** But avoid losing the strong g in nasal mutation of g (ex : in·Gelydh [iŋgɛlyð]] ) **\
+  \substitute "n·g" "·ŋg"
+  \** Use median dot as word joiner **\
+  \substitute "·" ""
+  
   \** Preprocess numbers **\
   \elvish_numbers "\\eval numbers_base" "\\eval reverse_numbers"
 \end
@@ -189,8 +233,12 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     {V_D_WN}n{K} --> CALMA {NASAL} {_V_D_WN_}
 
     \** 2ND LINE **\
-    {L2}        === d     * b     * g     * ng                    \** * g **\
-    {_L2_}      === ANDO  * UMBAR * UNGWE * UNGWE {NASAL}      \** * s **\
+    \** 
+        /ŋg/ : this is ng in middle of words + might be found at word start. 
+        See also final/initial ng_ / _ng below 
+    **\
+    {L2}        === d     * b     * g     * (ng,ngg,ŋg,ñg)              
+    {_L2_}      === ANDO  * UMBAR * UNGWE * UNGWE {NASAL}      
 
     {V_D_WN}[{L2}] --> 2,1 --> [{_L2_}]{_V_D_WN_}
 
@@ -210,13 +258,13 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
     \** 4TH LINE **\
     {L4}        === (đ,ð,ðh,dh)   * (v,bh,f_) \** Some noldorin variants here ... **\
-    {_L4_}        === ANTO          * AMPA 
+    {_L4_}      === ANTO          * AMPA 
 
     {V_D_WN}[{L4}] --> 2,1 --> [{_L4_}]{_V_D_WN_}
 
     \** 5TH LINE **\
-    {L5}        === n * m * _ng * _mh
-    {_L5_}      === NUMEN * MALTA * NWALME * MALTA_W_HOOK 
+    {L5}        === n     *   m       * (_ng,ng_,ŋ,ñ)   * _mh  \** weak ng at initial and final **\
+    {_L5_}      === NUMEN *   MALTA   * NWALME          * MALTA_W_HOOK 
 
     {V_D_WN}[{L5}] --> 2,1 --> [{_L5_}]{_V_D_WN_}
 
@@ -226,7 +274,7 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     \** 6TH LINE **\
 
     \** 7TH LINE **\
-    {L7}        === r_    * r     * l     * ll                    * w     
+    {L7}        === r_    * r     * l     * ll               * w     
     {_L7_}      === ORE   * ROMEN * LAMBE * LAMBE {GEMINATE} * VALA
         
     {V_D_WN}[{L7}] --> 2,1 --> [{_L7_}]{_V_D_WN_}
@@ -252,29 +300,47 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
     \** 
         Ok here come the labialized consonants which are really tricky
-        The fonts generally do not handle well the u curl + tehtar, this should be one more argument for
+        The fonts generally do not handle well the wa-tehta curl + tehtar, this should be one more argument for
         adopting open type anchors with which we can stack diacritics (see the sarati modes).
         For here, we cheat. Either we don't have any tehta on the tengwa, and it's easy.
         Or, we put the two signs in their small versions, side by side.
         We give an option not to use that trick, if the option is not set, we simply do not use
-        the u-curl at all when there's a tehta on the tengwa.
+        the wa-tehta curl at all when there's a tehta on the tengwa.
     **\
     
     \if "labialized_consonants_u_curl == LABIALIZED_U_CURL_NO_TEHTAR || labialized_consonants_u_curl == LABIALIZED_U_CURL_ALWAYS"
-      dw   --> ANDO  SEV_TEHTA  
-      gw   --> UNGWE SEV_TEHTA  
-      lw   --> LAMBE SEV_TEHTA
-      nw   --> NUMEN SEV_TEHTA   
-      rw   --> ROMEN SEV_TEHTA   
+      bw   --> UMBAR WA_TEHTA
+      dw   --> ANDO  WA_TEHTA  
+      gw   --> UNGWE WA_TEHTA  
+      lw   --> LAMBE WA_TEHTA
+      nw   --> NUMEN WA_TEHTA   
+      rw   --> ROMEN WA_TEHTA   
+      (ng,ngg,ŋg,ñg)w --> UNGWE {NASAL} WA_TEHTA
+      (_ng,ng_,ŋ,ñ)w  --> NWALME WA_TEHTA
     \endif
 
     \if "labialized_consonants_u_curl == LABIALIZED_U_CURL_ALWAYS"    
-      {V_D}dw   --> ANDO  SEV_TEHTA {_V_D_}
-      {V_D}gw   --> UNGWE SEV_TEHTA {_V_D_}
-      {V_D}lw   --> LAMBE SEV_TEHTA {_V_D_}   
-      {V_D}nw   --> NUMEN SEV_TEHTA {_V_D_}
-      {V_D}rw   --> ROMEN SEV_TEHTA {_V_D_}
+      {V_D}bw               --> UMBAR WA_TEHTA {_V_D_}
+      {V_D}dw               --> ANDO  WA_TEHTA {_V_D_}
+      {V_D}gw               --> UNGWE WA_TEHTA {_V_D_}
+      {V_D}lw               --> LAMBE WA_TEHTA {_V_D_}   
+      {V_D}nw               --> NUMEN WA_TEHTA {_V_D_}
+      {V_D}rw               --> ROMEN WA_TEHTA {_V_D_}
+      {V_D}(ng,ngg,ŋg,ñg)w  --> UNGWE {NASAL} WA_TEHTA {_V_D_}
+      {V_D}(_ng,ng_,ŋ,ñ)w   --> NWALME WA_TEHTA {_V_D_}
     \endif
+    
+    \if "apostrophe == APOSTROPHE_IGNORED"
+      ' --> {NULL}
+      ’ --> {NULL}
+    \else
+      \** use gasdil **\
+      ' --> HALLA
+      ’ --> HALLA
+    \endif
+    
+    \** Forced gasdil **\
+    ° --> HALLA
   \end
   
   \beg rules punctuation
@@ -297,11 +363,6 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     - --> {NULL} 
     – --> PUNCT_TILD  
     — --> PUNCT_TILD
-
-    \** Apostrophe **\
-
-    ' --> {NULL}
-    ’ --> {NULL}
     
     \** NBSP **\
     {NBSP} --> NBSP
@@ -323,7 +384,7 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     > --> PUNCT_PAREN_R
 
     \** Not universal between fonts ... **\
-    $ --> BOOKMARK_SIGN
+    $ --> ELVISH_PAREN
     ≤ --> RING_MARK_L \** Ring inscription left beautiful stuff **\
     ≥ --> RING_MARK_R \** Ring inscription right beautiful stuff **\
   \end

data/glaemresources/modes/{telerin.glaem → telerin-tengwar-glaemscrafu.glaem}

@@ -34,12 +34,14 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
   \entry "0.1.1" "Added support for inlined raw tengwar"  
   \entry "0.1.2" "Added support for non-breaking spaces"   
   \entry "0.1.3" "Correcting visibility options to conform to new glaeml args strict syntax"  
+  \entry "0.1.4" "Added support for new unicode charsets"   
+  \entry "0.1.5" "Added support for the Tengwar Telcontar font"
 \end
 
 \language "Telerin"
 \writing  "Tengwar"
 \mode     "Telerin Tengwar - G*"
-\version  "0.1.3"
+\version  "0.1.5"
 \authors  "Talagan (Benjamin Babut), based on J.R.R Tolkien"
 
 \world      arda
@@ -50,7 +52,15 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 \charset  tengwar_ds_eldamar  false
 \charset  tengwar_ds_annatar  false
 \charset  tengwar_ds_elfica   false
+
+\charset  tengwar_guni_sindarin false
+\charset  tengwar_guni_parmaite false
+\charset  tengwar_guni_eldamar  false
+\charset  tengwar_guni_annatar  false
+\charset  tengwar_guni_elfica   false
+
 \charset  tengwar_freemono    false
+\charset  tengwar_telcontar   false
 
 \raw_mode "raw-tengwar"
 
@@ -237,10 +247,10 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     [ {L1} ] {V_D_WN}         --> [ {_L1_} ] {_V_D_WN_}
     [ {L1_GEMS} ] {V_D_WN}    --> [ {_L1_GEMS_} ] {_V_D_WN_}
                             
-    ts{V_D_WN}          --> TINCO ALVEOLAR_SIGN {_V_D_WN_} 
-    ps{V_D_WN}          --> PARMA ALVEOLAR_SIGN {_V_D_WN_}
-    {K}s{V_D_WN}        --> CALMA ALVEOLAR_SIGN {_V_D_WN_}   
-    x{V_D_WN}           --> CALMA ALVEOLAR_SIGN {_V_D_WN_}   \** render ks for x **\
+    ts{V_D_WN}          --> TINCO SARINCE {_V_D_WN_} 
+    ps{V_D_WN}          --> PARMA SARINCE {_V_D_WN_}
+    {K}s{V_D_WN}        --> CALMA SARINCE {_V_D_WN_}   
+    x{V_D_WN}           --> CALMA SARINCE {_V_D_WN_}   \** render ks for x **\
                             
     \** ===================== **\
     \** 2ND LINE RULES **\
@@ -369,7 +379,7 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     > --> PUNCT_PAREN_R  
 
     \** Not universal between fonts ... **\
-    $ --> BOOKMARK_SIGN
+    $ --> ELVISH_PAREN
     ≤ --> RING_MARK_L \** Ring inscription left beautiful stuff **\
     ≥ --> RING_MARK_R \** Ring inscription right beautiful stuff **\
  

data/glaemresources/modes/{westron.glaem → westron-tengwar-glaemscrafu.glaem}

@@ -30,13 +30,15 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
   \entry "0.1.0" "Added support for the Tengwar Elfica font"
   \entry "0.1.1" "Added support for inlined raw tengwar"  
   \entry "0.1.2" "Added support for non-breaking spaces"   
+  \entry "0.1.3" "Added support for new unicode charsets"   
+  \entry "0.1.4" "Added support for the Tengwar Telcontar font"
 \end
 
 \**  Westron mode for glaemscribe (MAY BE INCOMPLETE) **\
 \language Westron
 \writing  Tengwar
 \mode     "Westron Tengwar - G*"
-\version  "0.1.2"
+\version  "0.1.4"
 \authors  "Talagan (Benjamin Babut), based on J.R.R. Tolkien"
 
 \world      arda
@@ -49,7 +51,15 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 \charset  tengwar_ds_eldamar  false
 \charset  tengwar_ds_annatar  false
 \charset  tengwar_ds_elfica   false
+
+\charset  tengwar_guni_sindarin false
+\charset  tengwar_guni_parmaite false
+\charset  tengwar_guni_eldamar  false
+\charset  tengwar_guni_annatar  false
+\charset  tengwar_guni_elfica   false
+
 \charset  tengwar_freemono    false
+\charset  tengwar_telcontar   false
 
 \beg      options
 
@@ -293,13 +303,13 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     nz{V_D}           --> ESSE_NUQUERNA {NASAL} {_V_D_}   
     nz                --> ESSE_NUQUERNA {NASAL}                
 
-    ts                --> TINCO ALVEOLAR_SIGN
-    ps                --> PARMA ALVEOLAR_SIGN
-    (ks,cs,x)         --> QUESSE ALVEOLAR_SIGN
+    ts                --> TINCO SARINCE
+    ps                --> PARMA SARINCE
+    (ks,cs,x)         --> QUESSE SARINCE
 
-    ts{V_D}           --> TINCO ALVEOLAR_SIGN {_V_D_}  
-    ps{V_D}           --> PARMA ALVEOLAR_SIGN {_V_D_}
-    (ks,cs,x){V_D}    --> QUESSE ALVEOLAR_SIGN {_V_D_}	
+    ts{V_D}           --> TINCO SARINCE {_V_D_}  
+    ps{V_D}           --> PARMA SARINCE {_V_D_}
+    (ks,cs,x){V_D}    --> QUESSE SARINCE {_V_D_}	
 
     h{V_D}            --> HYARMEN {_V_D_}
     h                 --> HYARMEN
@@ -366,7 +376,7 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
     > --> PUNCT_PAREN_R  
 
     \** Not universal between fonts ... **\
-    $ --> BOOKMARK_SIGN
+    $ --> ELVISH_PAREN
     ≤ --> RING_MARK_L \** Ring inscription left beautiful stuff **\
     ≥ --> RING_MARK_R \** Ring inscription right beautiful stuff **\
   \end

data/lib/api/charset.rb

@@ -30,11 +30,11 @@ module Glaemscribe
       attr_reader   :virtual_chars
             
       class Char
-        attr_accessor :line
-        attr_accessor :code
-        attr_accessor :names
-        attr_accessor :str
-        attr_accessor :charset
+        attr_accessor :line     # Line num in the sourcecode
+        attr_accessor :code     # Position in unicode
+        attr_accessor :names    # Names
+        attr_accessor :str      # How does this char resolve as a string
+        attr_accessor :charset  # Pointer to parent charset
         
         def initialize
           @names = {}
@@ -43,9 +43,13 @@ module Glaemscribe
         def virtual?
           false
         end
+        
+        def sequence?
+          false
+        end
       end
       
-      class VirtualChar
+      class VirtualChar # Could have had inheritance here ... 
         attr_accessor :line
         attr_accessor :names
         attr_accessor :classes
@@ -121,6 +125,45 @@ module Glaemscribe
         def virtual?
           true
         end
+        
+        def sequence?
+          false
+        end
+      end
+      
+      class SequenceChar
+        attr_accessor :line     # Line of code
+        attr_accessor :names    # Names
+        attr_accessor :sequence # The sequence of chars
+        attr_accessor :charset  # Pointer to parent charset
+        
+        def virtual?
+          false
+        end
+        
+        def sequence?
+          true
+        end   
+        
+        def str
+          # A sequence char should never arrive unreplaced
+          VIRTUAL_CHAR_OUTPUT
+        end
+        
+        def finalize          
+          if @sequence.count == 0
+            @charset.errors << Glaeml::Error.new(@line, "Sequence for sequence char is empty.")    
+          end
+          
+          @sequence.each{ |symbol|
+            # Check that the sequence is correct
+            found = @charset[symbol]
+            if !found
+              @charset.errors << Glaeml::Error.new(@line, "Sequence char #{symbol} cannot be found in the charset.")
+            end
+          }    
+        end
+        
       end
       
       def initialize(name)
@@ -156,10 +199,21 @@ module Glaemscribe
         @chars << c   
       end
       
+      def add_sequence_char(line, names, seq)
+        return if names.empty? || names.include?("?") # Ignore characters with '?'
+        
+        c             = SequenceChar.new
+        c.line        = line
+        c.names       = names
+        c.sequence    = seq.split.reject{|token| token.empty? }    
+        c.charset     = self
+        @chars << c
+      end
+      
       def finalize
         @errors         = []
         @lookup_table   = {}
-        @virtual_chars  = []
+        @virtual_chars  = [] # A convenient filtered array
         
         @chars.each { |c|
           c.names.each { |cname|
@@ -179,6 +233,12 @@ module Glaemscribe
           end
         }
         
+        @chars.each{|c|
+          if c.class == SequenceChar
+            c.finalize
+          end
+        }
+        
         API::Debug::log("Finalized charset '#{@name}', #{@lookup_table.count} symbols loaded.")
       end
       

data/lib/api/charset_parser.rb

@@ -47,6 +47,13 @@ module Glaemscribe
           names  = char_element.args[1..-1].map{|cname| cname.strip }.reject{ |cname| cname.empty? }
           @charset.add_char(char_element.line,code,names)
         }
+               
+        doc.root_node.gpath("seq").each{ |seq_elemnt| 
+          names       = seq_elemnt.args
+          child_node  = seq_elemnt.children.first
+          seq         = (child_node && child_node.text?)?(child_node.args.first):("")
+          @charset.add_sequence_char(seq_elemnt.line,names,seq)
+        }
         
         doc.root_node.gpath("virtual").each { |virtual_element|
           names     = virtual_element.args

data/lib/api/constants.rb

@@ -23,11 +23,10 @@
 module Glaemscribe
   module API
     WORD_BREAKER        = "|"
-    WORD_BOUNDARY       = "_"
-    
-    SPECIAL_CHAR_UNDERSCORE = '➊'
-    SPECIAL_CHAR_NBSP       = '➋'
     
+    WORD_BOUNDARY_LANG  = "_"
+    WORD_BOUNDARY_TREE  = "\u0000"
+        
     UNKNOWN_CHAR_OUTPUT = "☠"      
     VIRTUAL_CHAR_OUTPUT = "☢" # When transcribing a virtual char...
   end