ioblockreader 1.0.1.20130611 → 1.0.2.20130613

data/AUTHORS

@@ -2,3 +2,4 @@
 
 * 1.0.0.20130611
 * 1.0.1.20130611
+* 1.0.2.20130613

data/ChangeLog

@@ -1,5 +1,10 @@
 = IOBlockReader Release History
 
+== 1.0.2.20130613 (Beta)
+
+* Added each_block interface
+* Make internal string use encoding of the IO for Ruby >= 1.9
+
 == 1.0.1.20130611 (Beta)
 
 * Adapted test suite for Ruby 1.8

data/README.md

@@ -34,6 +34,8 @@ end
 
 ### IOBlockReader.init(io, options = {})
 
+Get an IOBlockReader instance on an IO.
+
 Parameters:
 * **io** ( _IO_ ): The IO object used to give the String interface
 * **options** (<em>map< Symbol, Object ></em>): Additional options:
@@ -50,6 +52,8 @@ content = IOBlockReader.init(file, :block_size => 32768, :blocks_in_memory => 5)
 
 ### IOBlockReader#\[\](range)
 
+Access a part of the data in the IO as a String.
+
 Parameters:
 * **range** ( _Fixnum_ or _Range_ ): Range to extract
 
@@ -64,6 +68,8 @@ substring = content[10..20]
 
 ### IOBlockReader#index(token, offset = 0, max_size_regexp = 32)
 
+Search for a token or a list of tokens.
+
 Parameters:
 * **token** ( _String_ , _Regexp_ or <em>list< Object ></em>): Token to be found. Can be a list of tokens.
 * **offset** ( _Fixnum_ ): Offset starting the search [optional = 0]
@@ -94,6 +100,30 @@ i, token_index = content.index( [ 'search string', 'another string' ] )
 i, token_index = content.index( [ 'search string', /another f.....g string/ ], 20, 22)
 ```
 
+### IOBlockReader#each_block(range = 0)
+
+Iterate over blocks in the data.
+
+Parameters:
+* **range** ( _Range_ or _Fixnum_ ): The boundaries of the iteration, or the starting index [default = 0]
+* _Block_ : Code called for each block encountered
+  * Parameters:
+  * **data** ( _String_ ): The data
+
+Example:
+```
+# Iterate all over the IO
+content.each_block do |data|
+  puts "Got a block of #{data.size} bytes"
+end
+
+# Iterate on just a part
+content.each_block(10..50) do |data|
+  puts "Got a block of #{data.size} bytes"
+end
+
+```
+
 ## Contact
 
 Want to contribute? Have any questions? [Contact Muriel!](muriel@x-aeon.com)

data/ReleaseInfo

@@ -2,7 +2,7 @@
 # This file has been generated by RubyPackager during a delivery.
 # More info about RubyPackager: http://rubypackager.sourceforge.net
 {
-  :version => '1.0.1.20130611',
+  :version => '1.0.2.20130613',
   :tags => [ 'Beta' ],
   :dev_status => 'Beta'
 }

data/lib/ioblockreader/datablock.rb

@@ -19,26 +19,30 @@ module IOBlockReader
     attr_reader :data
 
     # Constructor
-    def initialize
+    #
+    # Parameters::
+    # * *io* (_IO_): IO to read from
+    def initialize(io)
+      @io = io
       @offset = nil
       @last_access_time = nil
       @data = ''
+      @data.force_encoding(@io.external_encoding) if (@data.respond_to?(:force_encoding))
     end
 
     # Fill the data block for a given IO
     #
     # Parameters::
-    # * *io* (_IO_): IO to read from
     # * *offset* (_Fixnum_): Offset of this block in the IO
     # * *size* (_Fixnum_): Size of the block to be read
-    def fill(io, offset, size)
+    def fill(offset, size)
       @offset = offset
       @last_access_time = @@access_time_sequence
       @@access_time_sequence += 1
       #puts "[IOBlockReader] - Read #{size} @#{@offset}"
-      io.seek(@offset)
-      io.read(size, @data)
-      @last_block = io.eof?
+      @io.seek(@offset)
+      @io.read(size, @data)
+      @last_block = @io.eof?
     end
 
     # Is this block the last of its IO stream?

data/lib/ioblockreader/ioblockreader.rb

@@ -199,6 +199,77 @@ module IOBlockReader
       end
     end
 
+    # Iterate over blocks in the data.
+    # ! Do not use negative integers in the range.
+    #
+    # Parameters::
+    # * *range* (_Range_ or _Fixnum_): The boundaries of the iteration, or the starting index [default = 0]
+    # * _Block_: Code called for each block encountered
+    #   * Parameters::
+    #   * *data* (_String_): The data
+    def each_block(range = 0)
+      #puts "[IOBlockReader] - each_block(#{range})"
+      # Parse parameters
+      begin_offset = range
+      end_offset = nil
+      if (range.is_a?(Range))
+        begin_offset = range.first
+        end_offset = range.last
+      end
+
+      current_block_index, begin_offset_in_first_block = begin_offset.divmod(@block_size)
+      end_offset_block_index, end_offset_in_last_block = ((end_offset == nil) ? [nil, nil] : end_offset.divmod(@block_size))
+      # Make sure first block is loaded
+      if ((current_block = @blocks[current_block_index]) == nil)
+        read_needed_blocks([current_block_index], current_block_index, current_block_index)
+        current_block = @blocks[current_block_index]
+      else
+        current_block.touch
+      end
+      if (current_block_index == end_offset_block_index)
+        # We have a Range in the same block
+        if ((begin_offset_in_first_block == 0) and
+            (end_offset_in_last_block == current_block.data.size-1))
+          yield(current_block.data)
+        else
+          yield(current_block.data[begin_offset_in_first_block..end_offset_in_last_block])
+        end
+      else
+        # We need to loop, but consider first block differently as it might be partially given
+        if (begin_offset_in_first_block == 0)
+          yield(current_block.data)
+        else
+          yield(current_block.data[begin_offset_in_first_block..-1])
+        end
+        if (!current_block.last_block?)
+          # Now loop on all subsequent blocks unless we get to the last one
+          finished = false
+          while (!finished)
+            # Read next block
+            current_block_index += 1
+            if ((current_block = @blocks[current_block_index]) == nil)
+              read_needed_blocks([current_block_index], current_block_index, current_block_index)
+              current_block = @blocks[current_block_index]
+            else
+              current_block.touch
+            end
+            if (end_offset_block_index == current_block_index)
+              # We arrived on the last block of the Range
+              if (end_offset_in_last_block == current_block.data.size-1)
+                yield(current_block.data)
+              else
+                yield(current_block.data[0..end_offset_in_last_block])
+              end
+              finished = true
+            else
+              yield(current_block.data)
+              finished = current_block.last_block?
+            end
+          end
+        end
+      end
+    end
+
     private
 
     # Set the new cache block
@@ -246,8 +317,8 @@ module IOBlockReader
       indexes_needing_loading.each do |block_index|
         # Have to load this block
         block_to_fill = removed_blocks.pop
-        block_to_fill = DataBlock.new if (block_to_fill == nil)
-        block_to_fill.fill(@io, block_index * @block_size, @block_size)
+        block_to_fill = DataBlock.new(@io) if (block_to_fill == nil)
+        block_to_fill.fill(block_index * @block_size, @block_size)
         @blocks[block_index] = block_to_fill
       end
     end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: ioblockreader
 version: !ruby/object:Gem::Version
-  version: 1.0.1.20130611
+  version: 1.0.2.20130613
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-06-11 00:00:00.000000000 Z
+date: 2013-06-13 00:00:00.000000000 Z
 dependencies: []
 description: Ruby library giving block-buffered and cached read over IO objects with
   a String-like interface. Ideal to parse big files as Strings, limiting memory consumption.