From 6a8a7c3cf32fa1920bfcd62c0702b14dd84e6d65 Mon Sep 17 00:00:00 2001
From: Ralph Amissah <ralph@amissah.com>
Date: Sat, 21 Sep 2013 00:05:24 -0400
Subject: v4: dal, objects, store book index as hash (simplify downstream
 processing)

* merge v5
* db, book index, convert hash back to string (associated downstream processing
  change)
---
 lib/sisu/v4/dal_doc_str.rb | 54 ++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 52 insertions(+), 2 deletions(-)

(limited to 'lib/sisu/v4/dal_doc_str.rb')

diff --git a/lib/sisu/v4/dal_doc_str.rb b/lib/sisu/v4/dal_doc_str.rb
index 2a3e6b83..e4a0be2b 100644
--- a/lib/sisu/v4/dal_doc_str.rb
+++ b/lib/sisu/v4/dal_doc_str.rb
@@ -168,6 +168,55 @@ module SiSU_DAL_DocumentStructureExtract
       end
       [str,tags]
     end
+    def rgx_idx_ocn_seg
+      @rgx_idx_ocn_seg=/(.+?)\s*[+](\d+)/
+    end
+    def construct_idx_array_and_hash(idxraw)
+      idx_array_raw=idxraw.scan(/[^;]+/)
+      idx_hash,idx_array,idx_lst={},[],[]
+      idx_array_raw.each do |idx|
+        idx_lst=case idx
+        when /\S+?\s*:/
+          idx_couplet_tmp=[]
+          idx_couplet=idx.scan(/\s*[^:]+\s*/)
+          if idx_couplet[1] =~/[|]/
+            idx_couplet_tmp << idx_couplet[0] << idx_couplet[1].scan(/\s*[^|]+\s*/)
+          else
+            idx_couplet_tmp << idx_couplet[0] << [idx_couplet[1]]
+          end
+          idx_couplet=idx_couplet_tmp
+        else [idx]
+        end
+        term_nodes=[]
+        idx_lst.each do |term_node|
+          case term_node
+          when String
+            term_node=term_node[0].chr.capitalize + term_node[1,term_node.length]
+            term_node=(term_node =~/.+?[+]\d+/) \
+            ? term_node
+            : (term_node + '+0')
+            term_nodes << term_node
+            use,plus=rgx_idx_ocn_seg.match(term_node)[1,2]
+            @use=use.strip
+            idx_hash[@use]={ sub: [], plus: plus } unless idx_hash[@use] and defined? idx_hash[@use]
+          when Array
+            subterm_nodes=[]
+            term_node.each do |subterm_node|
+              subterm_node=(subterm_node =~/.+?[+]\d+/) \
+              ? subterm_node
+              : (subterm_node + '+0')
+              subterm_nodes << subterm_node
+              sub,sub_plus=rgx_idx_ocn_seg.match(subterm_node)[1,2]
+              idx_hash[@use]={ sub: [], plus: 0 } unless idx_hash[@use] and defined? idx_hash[@use]
+              idx_hash[@use][:sub] << {sub.strip => { plus: sub_plus }}
+            end
+            term_nodes << subterm_nodes
+          end
+        end
+        idx_array << term_nodes
+      end
+      { hash: idx_hash, array: idx_array }
+    end
     def identify_parts
       tuned_file=[]
       @tuned_block,@tuned_code=[],[]
@@ -205,9 +254,10 @@ module SiSU_DAL_DocumentStructureExtract
             idx=if t_o=~/^=\{(.+)\}\s*$\Z/m; m=$1
               m=m.split(/\n/).join(' ').
                 gsub(/\s+([|:;])\s+/,'\1').
-                gsub(/\s+([+])\s+/,'\1')
+                gsub(/\s+([+]\d+)\s+/,'\1')
               t_o=t_o.gsub(/\n=\{.+\}\s*$\Z/m,'')
-              m
+              idx_array_and_hash=construct_idx_array_and_hash(m)
+              idx_array_and_hash[:hash]
             else nil
             end
           end
-- 
cgit v1.2.3