xml_row_finder 0.3.1 → 0.3.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 805ba05e167f14ca224f83611d91ada484c31b7c95b329468f4e2fce47302e0e
-  data.tar.gz: f70f9dd37c9e9f4be41d2d0c5d4408c70049d810273c84eeec1b451c40ad0df4
+  metadata.gz: 1407d0f852c5a904600ef49f43c93b7a8e5cc4a459bb88d9d2e0bc4c26928887
+  data.tar.gz: 3867560288f6ed062b2c59e78bea3e30b7ddaf5217d3a181f864b4fbf28a202f
 SHA512:
-  metadata.gz: 2edc584e01ae048ffb4d7b6ee88d4b08a1dd79ce9b593cecf0b24f9fe606809eb1ae50f6e85d5dee70f75a172dcce4eced9e2d3934f90bb23b77321c00fa04e6
-  data.tar.gz: 00ddeb4c7b968e5e31acf8e3accd7d466d9d9941e0d53d7892a9ffb6b82965b2b97f48d7d3cb32c6bb5013466888a12e384a3bd27c52486bddf79b75178758dc
+  metadata.gz: c19790d1e1c125b0be88831e1a1ab870662b03ad35859fc68b2431390854c0f4e507a8629198fefb1bfd03ed460e75ceb78e278f5de4be43e9dd9a1ab4c03734
+  data.tar.gz: d9d5321b0f639868aeca083728f7ebaa6da2e1d101d609aa05076e9e96c0c7e33e2545db0e803e5efd53808514c028b2190a425e08267e36e42cd6b6118adb3a

data/lib/xml_row_finder.rb

@@ -13,26 +13,9 @@ class XMLRowFinder
 
     @debug = debug
 
-    doc = if raws =~ /^http/ then
+    doc = Nokorexi.new(raws, filter: true).to_doc
 
-      nki = Nokorexi.new(url=raws) do |doc1|
-        doc1.xpath('//*[@onclick]').each do |e|
-          e.attributes['onclick'].value = ''
-        end
-
-        doc1.xpath('//*[@onmousedown]').each do |e|
-          e.attributes['onmousedown'].value = ''
-        end
-
-      end
-
-      nki.to_doc
-
-    else
-      Rexle.new(raws)
-    end
-
-    @doc = Rexle.new(doc.xml)
+    @doc = Rexle.new(doc.root.xml)
 
     a = []
 
@@ -60,18 +43,26 @@ class XMLRowFinder
       [@doc2.xpath(xpath2).length, xpath2]
     end
 
-    @xpath = a5.reverse.detect {|num, xpath2| num > 1}.last
+    puts 'a5: ' + a5.inspect if @debug
+    rows_xpath = a5.reverse.detect {|num, xpath2| num > 1}.last
+    doc3 = Document.new @doc.root.xml
+    @rows = XPath.match(doc3, rows_xpath)
+    @xpath = rows_xpath
+    #@xpath = BacktrackXPath.new(@rows.first).to_xpath.gsub("[@class='']",'')
 
-    last_row = @doc2.xpath(@xpath).last
+    last_row = XPath.match(doc3, @xpath).last
+    puts '@xpath: ' + @xpath.inspect
 
     # find the container element
     xpath = @xpath[/^[^\[]+/]
     axpath = xpath.split('/')
-    e = doc.element xpath
 
-    until (e.xml.include? last_row) do
+    e = XPath.first(doc3, xpath)
+    puts 'e: ' + e.to_s
+
+    until (e.nil? or e.to_s.include?(last_row.to_s)) do
       axpath.pop
-      e = doc.element axpath.join('/')
+      e = XPath.first(doc3, axpath.join('/'))
     end
 
     @cont_xpath = axpath.join('/')
@@ -94,7 +85,7 @@ class XMLRowFinder
   # object returned: An array of Nokogiri XML Element object
   #
   def rows()
-    @doc2.xpath @xpath
+    @rows
   end
 
   # returns the xpath pointing to the rows

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: xml_row_finder
 version: !ruby/object:Gem::Version
-  version: 0.3.1
+  version: 0.3.2
 platform: ruby
 authors:
 - James Robertson
@@ -35,28 +35,28 @@ cert_chain:
   a/pAsvo0jT6QTnSB7xzsx8LSFDT5tfHKR9Dcn1Y3R06fsh02JvwxaSAMgDBM2aFb
   2A7/BQ1hD7SU82VTxB1gFIHl
   -----END CERTIFICATE-----
-date: 2022-01-24 00:00:00.000000000 Z
+date: 2022-01-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokorexi
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '0.5'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.5.4
+        version: 0.6.0
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.6'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '0.5'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.5.4
+        version: 0.6.0
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.6'
 description: 
 email: digital.robertson@gmail.com
 executables: []