v5 v6: ao, segment numbering, fixed width, leading zeros

* alphanumeric sorting

8 files changed, 162 insertions, 58 deletions

diff --git a/data/doc/sisu/CHANGELOG_v5 b/data/doc/sisu/CHANGELOG_v5
index 9d14921a..45d9587f 100644
--- a/data/doc/sisu/CHANGELOG_v5
+++ b/data/doc/sisu/CHANGELOG_v5
@@ -44,6 +44,7 @@ http://www.jus.uio.no/sisu/pkg/src/sisu_5.4.3.orig.tar.xz
     (fix uppercase first character where spaces used)
   * quote block, insert text indent markup, exclude book index
   * doc objects, layout group open & close identify is_for
+  * segment numbering, fixed width, leading zeros (for alphanumeric sorting)
 
 * code cosmetic rearrangement
   * ao*, param, hub, options, composite
diff --git a/data/doc/sisu/CHANGELOG_v6 b/data/doc/sisu/CHANGELOG_v6
index 72369d35..732d0d99 100644
--- a/data/doc/sisu/CHANGELOG_v6
+++ b/data/doc/sisu/CHANGELOG_v6
@@ -34,6 +34,7 @@ http://www.jus.uio.no/sisu/pkg/src/sisu_6.0.9.orig.tar.xz
     (fix uppercase first character where spaces used)
   * quote block, insert text indent markup, exclude book index
   * doc objects, layout group open & close identify is_for
+  * segment numbering, fixed width, leading zeros (for alphanumeric sorting)
 
 * code cosmetic rearrangement
   * ao*, param, hub, options, composite
diff --git a/lib/sisu/v5/ao_doc_str.rb b/lib/sisu/v5/ao_doc_str.rb
index 5b93819e..931e51d3 100644
--- a/lib/sisu/v5/ao_doc_str.rb
+++ b/lib/sisu/v5/ao_doc_str.rb
@@ -506,8 +506,9 @@ module SiSU_AO_DocumentStructureExtract
                 tags: tags,
                 quote: quotes?,
               }
-              SiSU_AO_DocumentStructure::ObjectPara.new.paragraph(h)
+              t_o=SiSU_AO_DocumentStructure::ObjectPara.new.paragraph(h)
             end
+            t_o=SiSU_AO_DocumentStructureExtract::Structure.new(@md).structure_markup(t_o) #must happen earlier, node info etc. require
           end
         elsif @@flag[:code]==:off
           if t_o =~/^(?:code(?:\.[a-z][0-9a-z_]+)?\{|```[ ]+code(?:\.[a-z][0-9a-z_]+)?)/
@@ -1146,66 +1147,74 @@ module SiSU_AO_DocumentStructureExtract
     end
   end
   class Structure                                          # this must happen early
-    def initialize(md,dob)
-      @md,@dob=md,dob
+    def initialize(md)
+      @md=md
     end
-    def structure
-      structure_markup
-      @dob
+    def structure(data)
+      data.compact.each do |dob|
+        structure_markup(dob)
+      end
     end
-    def structure_markup                                   #build structure where structure provided only in meta header
-      @dob=if @dob.is==:para \
-      && (((@dob.hang !~/[1-9]/) && (@dob.indent !~/[1-9]/)) \
-      || (@dob.hang != @dob.indent)) \
-      and not @dob.bullet_
-        @dob=case @dob.obj
+    def structure_markup(dob)                                   #build structure where structure provided only in meta header
+      dob=if dob.is==:para \
+      && (((dob.hang !~/[1-9]/) && (dob.indent !~/[1-9]/)) \
+      || (dob.hang != dob.indent)) \
+      and not dob.bullet_
+        dob=case dob.obj
         when /^#{@md.lv0}/
           h={
+            is: :heading,
             lv: 'A',
             ln: 0,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv1}/
           h={
+            is: :heading,
             lv: 'B',
             ln: 1,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv2}/
           h={
+            is: :heading,
             lv: 'C',
             ln: 2,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv3}/
           h={
+            is: :heading,
             lv: 'D',
             ln: 3,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv4}/
           h={
+            is: :heading,
             lv: '1',
             ln: 4,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv5}/
           h={
+            is: :heading,
             lv: '2',
             ln: 5,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv6}/
           h={
+            is: :heading,
             lv: '3',
             ln: 6,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
-        else @dob
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
+        else dob
         end
-      else @dob
+      else dob
       end
-      @dob
+      dob
     end
   end
   class OCN
diff --git a/lib/sisu/v5/ao_numbering.rb b/lib/sisu/v5/ao_numbering.rb
index cd631796..c4b6558f 100644
--- a/lib/sisu/v5/ao_numbering.rb
+++ b/lib/sisu/v5/ao_numbering.rb
@@ -63,10 +63,23 @@
 module SiSU_AO_Numbering
   class Numbering
     attr_accessor :obj,:osp,:ocn,:lv,:name,:index,:comment
+    @@segments_count=0
     def initialize(md,data)
       @md,@data=md,data
       @obj=@type=@ocn=@lv=@name=@index=@comment=nil
     end
+    def number_of_segments?
+      if @@segments_count==0
+        @data.each do |dob|
+          if dob.is ==  :heading \
+          and dob.lv == '1'
+            @@segments_count += 1
+          end
+        end
+        @@segments_count
+      else @@segments_count
+      end
+    end
     def numbering_song
       data=@data
       data=number_plaintext_para(data)
@@ -152,7 +165,6 @@ module SiSU_AO_Numbering
       data=data.compact
       data.each do |dob| #@md.seg_names << [additions to segment names]
         title_no=nil
-        dob=SiSU_AO_DocumentStructureExtract::Structure.new(@md,dob).structure_markup #must happen earlier, node info etc. require
         if dob.is ==:heading \
         && dob.autonum_ \
         and defined? @md.make.num_top \
@@ -280,6 +292,33 @@ module SiSU_AO_Numbering
       end
       @tuned_file=@tuned_file.flatten
     end
+    def leading_zeros_fixed_width_number(possible_seg_name)
+      if possible_seg_name =~/^([\d]+?\.|[\d]+)$/m          #!~/[.,:-]+/
+        possible_seg_name=possible_seg_name.
+          gsub(/\.$/,'')
+        nl=possible_seg_name.to_s.length
+        zero='0'
+        zeros_fixed_width=number_of_segments?.to_s.length
+        zero_width=(zeros_fixed_width - nl)
+        zero_width == 0 \
+        ? Mx[:auto_seg_prefix] +
+          possible_seg_name.to_s
+        : Mx[:auto_seg_prefix] +
+          zero*zero_width +
+          possible_seg_name.to_s
+      end
+    end
+    def auto_seg_name(possible_seg_name)
+      if possible_seg_name =~/^([\d]+?\.|[\d]+)$/m          #!~/[.,:-]+/
+        leading_zeros_fixed_width_number(possible_seg_name)
+      elsif possible_seg_name =~/^[\d.,:-]+$/m
+        possible_seg_name=possible_seg_name.
+          gsub(/(?:[:,-]|\W)/,'.').
+          gsub(/\.$/,'')
+        #Mx[:auto_seg_prefix] + possible_seg_name
+      else possible_seg_name
+      end
+    end
     def name_para_seg_filename(data)                                                   #segment naming, remaining
       # paragraph name/numbering rules
       # manual naming overrides, manual naming may be
@@ -314,8 +353,10 @@ module SiSU_AO_Numbering
             @md.set_heading_seg=true
           end
           if dob.name !~/^\S+/ \
-          and dob.obj =~/^\s*(?:\S+\s+)?([\d.,:-]+)/m      #heading starts with a recognised numeric or word followed by a recognised numerical construct, use that as name
+          and dob.obj =~/^\s*(?:\S+\s+)?([\d.,:-]+)/m      #heading starts with a recognised numeric or word followed by a recognised numeric construct, use that as name
             possible_seg_name=$1
+            possible_seg_name=
+              auto_seg_name(possible_seg_name)
             possible_seg_name=possible_seg_name.
               gsub(/(?:[:,-]|\W)/,'.').
               gsub(/\.$/,'')
@@ -339,13 +380,13 @@ module SiSU_AO_Numbering
           end
           if dob.ln==4 \
           and not dob.name                                 #if still no segment name, provide a numerical one
-            pf='_'                                         #pg='' #may use e.g. '' or '~' or '_'
-            segn_auto="#{pf}#{art_filename_auto.to_s}"
+            possible_seg_name=
+              auto_seg_name(art_filename_auto)
             if @md.seg_names.is_a?(Array) \
-            and not @md.seg_names.include?(segn_auto)
-             dob.name=segn_auto
+            and not @md.seg_names.include?(possible_seg_name)
+             dob.name=possible_seg_name
              dob.tags=set_tags(dob.tags,dob.name)
-              @md.seg_names << segn_auto
+              @md.seg_names << possible_seg_name
             else puts 'segment name (numbering) error'
             end
             art_filename_auto+=1
diff --git a/lib/sisu/v5/constants.rb b/lib/sisu/v5/constants.rb
index 42ab53a3..6b44ec68 100644
--- a/lib/sisu/v5/constants.rb
+++ b/lib/sisu/v5/constants.rb
@@ -104,6 +104,7 @@ Xx={
   html_relative1:            '※',
 }
 Mx={
+  auto_seg_prefix:           's',
   ocn_id_char:               'o',
   note:                      'note_',
   note_ref:                  'noteref_',
diff --git a/lib/sisu/v6/ao_doc_str.rb b/lib/sisu/v6/ao_doc_str.rb
index 0c07eee7..2a96abb7 100644
--- a/lib/sisu/v6/ao_doc_str.rb
+++ b/lib/sisu/v6/ao_doc_str.rb
@@ -506,8 +506,9 @@ module SiSU_AO_DocumentStructureExtract
                 tags: tags,
                 quote: quotes?,
               }
-              SiSU_AO_DocumentStructure::ObjectPara.new.paragraph(h)
+              t_o=SiSU_AO_DocumentStructure::ObjectPara.new.paragraph(h)
             end
+            t_o=SiSU_AO_DocumentStructureExtract::Structure.new(@md).structure_markup(t_o) #must happen earlier, node info etc. require
           end
         elsif @@flag[:code]==:off
           if t_o =~/^(?:code(?:\.[a-z][0-9a-z_]+)?\{|```[ ]+code(?:\.[a-z][0-9a-z_]+)?)/
@@ -1146,66 +1147,74 @@ module SiSU_AO_DocumentStructureExtract
     end
   end
   class Structure                                          # this must happen early
-    def initialize(md,dob)
-      @md,@dob=md,dob
+    def initialize(md)
+      @md=md
     end
-    def structure
-      structure_markup
-      @dob
+    def structure(data)
+      data.compact.each do |dob|
+        structure_markup(dob)
+      end
     end
-    def structure_markup                                   #build structure where structure provided only in meta header
-      @dob=if @dob.is==:para \
-      && (((@dob.hang !~/[1-9]/) && (@dob.indent !~/[1-9]/)) \
-      || (@dob.hang != @dob.indent)) \
-      and not @dob.bullet_
-        @dob=case @dob.obj
+    def structure_markup(dob)                                   #build structure where structure provided only in meta header
+      dob=if dob.is==:para \
+      && (((dob.hang !~/[1-9]/) && (dob.indent !~/[1-9]/)) \
+      || (dob.hang != dob.indent)) \
+      and not dob.bullet_
+        dob=case dob.obj
         when /^#{@md.lv0}/
           h={
+            is: :heading,
             lv: 'A',
             ln: 0,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv1}/
           h={
+            is: :heading,
             lv: 'B',
             ln: 1,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv2}/
           h={
+            is: :heading,
             lv: 'C',
             ln: 2,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv3}/
           h={
+            is: :heading,
             lv: 'D',
             ln: 3,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv4}/
           h={
+            is: :heading,
             lv: '1',
             ln: 4,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv5}/
           h={
+            is: :heading,
             lv: '2',
             ln: 5,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
         when /^#{@md.lv6}/
           h={
+            is: :heading,
             lv: '3',
             ln: 6,
           }
-          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,@dob)
-        else @dob
+          SiSU_AO_DocumentStructure::ObjectHeading.new.heading(h,dob)
+        else dob
         end
-      else @dob
+      else dob
       end
-      @dob
+      dob
     end
   end
   class OCN
diff --git a/lib/sisu/v6/ao_numbering.rb b/lib/sisu/v6/ao_numbering.rb
index 791f5bfc..d6faf3a6 100644
--- a/lib/sisu/v6/ao_numbering.rb
+++ b/lib/sisu/v6/ao_numbering.rb
@@ -63,10 +63,23 @@
 module SiSU_AO_Numbering
   class Numbering
     attr_accessor :obj,:osp,:ocn,:lv,:name,:index,:comment
+    @@segments_count=0
     def initialize(md,data)
       @md,@data=md,data
       @obj=@type=@ocn=@lv=@name=@index=@comment=nil
     end
+    def number_of_segments?
+      if @@segments_count==0
+        @data.each do |dob|
+          if dob.is ==  :heading \
+          and dob.lv == '1'
+            @@segments_count += 1
+          end
+        end
+        @@segments_count
+      else @@segments_count
+      end
+    end
     def numbering_song
       data=@data
       data=number_plaintext_para(data)
@@ -152,7 +165,6 @@ module SiSU_AO_Numbering
       data=data.compact
       data.each do |dob| #@md.seg_names << [additions to segment names]
         title_no=nil
-        dob=SiSU_AO_DocumentStructureExtract::Structure.new(@md,dob).structure_markup #must happen earlier, node info etc. require
         if dob.is ==:heading \
         && dob.autonum_ \
         and defined? @md.make.num_top \
@@ -280,6 +292,33 @@ module SiSU_AO_Numbering
       end
       @tuned_file=@tuned_file.flatten
     end
+    def leading_zeros_fixed_width_number(possible_seg_name)
+      if possible_seg_name =~/^([\d]+?\.|[\d]+)$/m          #!~/[.,:-]+/
+        possible_seg_name=possible_seg_name.
+          gsub(/\.$/,'')
+        nl=possible_seg_name.to_s.length
+        zero='0'
+        zeros_fixed_width=number_of_segments?.to_s.length
+        zero_width=(zeros_fixed_width - nl)
+        zero_width == 0 \
+        ? Mx[:auto_seg_prefix] +
+          possible_seg_name.to_s
+        : Mx[:auto_seg_prefix] +
+          zero*zero_width +
+          possible_seg_name.to_s
+      end
+    end
+    def auto_seg_name(possible_seg_name)
+      if possible_seg_name =~/^([\d]+?\.|[\d]+)$/m          #!~/[.,:-]+/
+        leading_zeros_fixed_width_number(possible_seg_name)
+      elsif possible_seg_name =~/^[\d.,:-]+$/m
+        possible_seg_name=possible_seg_name.
+          gsub(/(?:[:,-]|\W)/,'.').
+          gsub(/\.$/,'')
+        #Mx[:auto_seg_prefix] + possible_seg_name
+      else possible_seg_name
+      end
+    end
     def name_para_seg_filename(data)                                                   #segment naming, remaining
       # paragraph name/numbering rules
       # manual naming overrides, manual naming may be
@@ -314,8 +353,10 @@ module SiSU_AO_Numbering
             @md.set_heading_seg=true
           end
           if dob.name !~/^\S+/ \
-          and dob.obj =~/^\s*(?:\S+\s+)?([\d.,:-]+)/m      #heading starts with a recognised numeric or word followed by a recognised numerical construct, use that as name
+          and dob.obj =~/^\s*(?:\S+\s+)?([\d.,:-]+)/m      #heading starts with a recognised numeric or word followed by a recognised numeric construct, use that as name
             possible_seg_name=$1
+            possible_seg_name=
+              auto_seg_name(possible_seg_name)
             possible_seg_name=possible_seg_name.
               gsub(/(?:[:,-]|\W)/,'.').
               gsub(/\.$/,'')
@@ -339,13 +380,13 @@ module SiSU_AO_Numbering
           end
           if dob.ln==4 \
           and not dob.name                                 #if still no segment name, provide a numerical one
-            pf='_'                                         #pg='' #may use e.g. '' or '~' or '_'
-            segn_auto="#{pf}#{art_filename_auto.to_s}"
+            possible_seg_name=
+              auto_seg_name(art_filename_auto)
             if @md.seg_names.is_a?(Array) \
-            and not @md.seg_names.include?(segn_auto)
-             dob.name=segn_auto
+            and not @md.seg_names.include?(possible_seg_name)
+             dob.name=possible_seg_name
              dob.tags=set_tags(dob.tags,dob.name)
-              @md.seg_names << segn_auto
+              @md.seg_names << possible_seg_name
             else puts 'segment name (numbering) error'
             end
             art_filename_auto+=1
diff --git a/lib/sisu/v6/constants.rb b/lib/sisu/v6/constants.rb
index 43fff554..487d96a0 100644
--- a/lib/sisu/v6/constants.rb
+++ b/lib/sisu/v6/constants.rb
@@ -104,6 +104,7 @@ Xx={
   html_relative1:            '※',
 }
 Mx={
+  auto_seg_prefix:           's',
   ocn_id_char:               'o',
   note:                      'note_',
   note_ref:                  'noteref_',