file_data 5.2.3 → 6.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 441d9a1d1481ac066af3b8c0bcd2bae00e7b54b88c322ef251a7579ee79f270b
-  data.tar.gz: af5797dd26aefac7eddee225a81012b7942283bee76e5add1911d67cb5158d4b
+  metadata.gz: 4f9ac51b8c0f2d6388b07ab1f6614de4a1fac558c21689fa61d198d70c18a8b4
+  data.tar.gz: 4da0fada39336b0346d0846ee6c04a6962c0c4f34909463ded2c0aa2c97511fe
 SHA512:
-  metadata.gz: ea6f2869ffa5b4f2a54d5767b7affc853c546e65866d55d6b655a6ddfddb00f732370abeea3e431cc3a951fb8cf4789d155d9321974181b84cefe8619591db3d
-  data.tar.gz: efb5a79f2b6297ef5b1f85a7acd6f6d33e8b313a47e6c3cfc0b69080819d6fca5d116158341ede91c5e6b3e0a34c7f641fc27f1fd63c08c9bf28ec11f89ce15b
+  metadata.gz: 74284cdbc2af387ad6e0b058d25ec38ade9bb5021eaec62be513b923ad6a38398ae199698d1aa78226306e3a60856a1c1658b59fe4d460fe7e600345ef555dba
+  data.tar.gz: 3f7a040e99bc6a47da21c759c687f831014ba7e1823f7b63d8e8af2bfb1ab79fb8abbfd5f49059fdb9ad77aa0c0358585f1a935204104597ea580786a46ee6ec

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    file_data (5.2.3)
+    file_data (6.0.0)
 
 GEM
   remote: https://rubygems.org/
@@ -34,7 +34,6 @@ GEM
     docile (1.1.5)
     fakefs (0.10.2)
     ffi (1.9.25)
-    ffi (1.9.25-x64-mingw32)
     highline (2.0.0)
     json (2.6.0)
     method_source (0.9.0)

data/README.md

@@ -8,69 +8,42 @@ file_data
 
 Ruby library that reads file metadata.
 
-The api provides a basic usage and an advanced usage. The basic usage will reopen and reparse the file every time it is called which is no problem when reading a single value but can be a performance drain for multiple values. The advanced usage allows the user to grab more than one value without having to read the file more than once.
-
-## Basic Usage
+## Basic Usage for an Exif File
 
 ```ruby
-filepath = '...' # Path to a jpeg or mpeg4 file
-
-# Get the date when the file content originated. When a photo was taken, when a movie was recorded, etc
-FileData::FileInfo.origin_date(filepath)
-
-# Get the date when the file was considered to be created. This is usually tied in some way to when the file itself was created on a disk somewhere (not usually as useful as origin date)
-FileData::FileInfo.creation_date(filepath)
-```
-
-## Advanced Usage
+require 'file_data'
 
-Varies by file format type. Currently there are low level classes for parsing exif and mpeg4 metadata
+## Step 1: Read in the exif data using either a file path or a stream
 
-## Exif documentation
+# Using a file path...
+file_path = '/home/user/desktop/my_file.jpg' # Path to an exif file
+exif_data = FileData::Exif.all_data(file_path) # read in all of the exif data from the file path
 
-Exif data is hierarchical and consists of tag/value pairs. The first level is whether or not the tag applies to the image or the image's thumbnail. Next a tag may be one of several sections and within the section it will have a unique numeric value. So given this terminology a unique key for an exif tag would be something like image/section1/123 for a tag that applies to the image in section 1 with a tag id of 123.
-
-To read exif data a stream of the jpeg data should be used as input. For performance reasons all of the data that is desired should be extracted in a single method from FileData::Exif. All methods will manipulate the stream position after they are called and the user should not count on the stream position being at a specific location after a FileData::Exif method call.
+# Or using a stream...
+exif_data = File.open(file_path, 'rb') do |f|
+  FileData::Exif.all_data(f)
+end
 
-Examples:
+## Step 2: Data is divided into image data and thumbnail data. Pick which you want to work with.
 
-```ruby
-## Get Exif data from a file path or stream
+# Both objects are hash-like and should respond to all hash method except .length which instead will return the value of :Image_Structure_Length,
+image_data = exif_data.image
+thumbnail_data = exif_data.thumbnail
 
-# The file path or stream is passed as a parameter to class methods on FileData::Exif
-# rather than having the file path or stream set in the constructor and the methods
-# be instance methods to ensure that the user falls into a "pit of success" by only
-# opening the file once to get tag values rather than potentially multiple times
+## Step 3: Extract the tag values
 
-# Get image exif data from a file path...
-hash = FileData::Exif.image_data_only('/path/to/file.jpg')
-
-# Get thumbnail exif data from a file stream... 
-File.open('/path/to/file.jpg', 'rb') do |f|
-  hash = FileData::Exif.thumbnail_data_only(f)
-end
+### Step 3A: Extract tags with a known name (ones that are listed in the "Known Tag Keys" section below)
 
-## Extract values depending on how much data is returned
+# Convenience methods are added for the names after the last underscore in the known tag names (casing and underscores are ignored in the convenince method names)
 
-# Image and thumbnail data are returned as a hash with keys
-# being the full tag name if it exists in FileData::ExifTags (see below listing)
-# or "#{ifd_id}-#{tag_id}" if there is no existing name for the tag in FileData::ExifTags
-image_hash = FileData::Exif.image_data_only(file_path_or_stream)
-image_description_value = image_hash[:Other_ImageDescription]
-unknown_gps_tag_value = image_hash["34853-99999"]
+bits_per_sample = image_data.BitsPerSample # Gets :Image_Structure_BitsPerSample from the :Tiff section
+bits_per_sample = image_data.bits_per_sample # Also gets :Image_Structure_BitsPerSample from the :Tiff section
 
-thumbnail_hash = FileData::Exif.thumbnail_data_only(file_path_or_stream)
-image_width_tag_value = thumbnail_hash[:Image_Structure_Width]
-unknown_gps_tag_value = thumbnail_hash["Tiff-99999"]
+### Step 3B: Extract tags without a known name (ones NOT listed in the "Known Tag Keys" section below)
 
-all_data = FileData::Exif.all_data(file_path_or_stream)
-image_hash = all_data.image
-thumbnail_hash = all_data.thumbnail
+# Use the format "#{ifd_id}-#{tag_id}" for the unknown tag to key into the data
+unknown_gps_tag_value = image_data["34853-99999"]
 
-# For extracting only a specific tag the value is returned and the search key must be
-# an array matching the keys needed to get traverse FileData:ExifTags (see below listing)
-image_description_value = FileData::Exif.only_image_tag(file_path_or_stream, [:Tiff, 270])
-unknown_exif_tag_value = FileData::Exif.only_thumbnail_tag(file_path_or_stream, [34665, 2])
 ```
 
 ## Known Tag Keys

data/lib/file_data/formats/exif/exif_data.rb

@@ -1,3 +1,5 @@
+require 'set'
+
 module FileData
   # Container for Exif tag values
   class ExifData
@@ -5,12 +7,52 @@ module FileData
     SECTIONS.each { |section| define_method(section[1]) { @hash[section[0]] } }
 
     def initialize
-      @hash = SECTIONS.each_with_object({}) { |pair, hash| hash[pair[0]] = {} }
+      @hash = SECTIONS.each_with_object({}) { |pair, hash| hash[pair[0]] = ExifHash.new }
     end
 
     def add_tag(index, ifd_id, tag_id, tag_value)
-      name = ExifTags.get_tag_name(ifd_id, tag_id)
-      @hash[index][name] = tag_value
+      name_info = ExifTags.get_tag_name(ifd_id, tag_id)
+      @hash[index][name_info.name] = tag_value
+    end
+  end
+
+  # Hash with convenience methods for accessing known Exif tag values by name
+  class ExifHash < BasicObject
+    all_tags = ExifTags.tag_groups.values.map{|x| x.values}.flatten
+    tags_map = all_tags.each_with_object({}) do |tag, hash|
+      hash[tag.to_s.split('_').last.upcase] = tag
+    end
+
+    define_method(:method_missing) do |method_name, *args, &block|
+      known_name = tags_map[method_name.to_s.tr('_', '').upcase]
+      
+      if known_name.nil?
+        @hash.send(method_name, *args, &block)
+      else
+        @hash[known_name]
+      end
+    end
+
+    define_method(:respond_to_missing?) do |method_name, include_private|
+      known_name = tags_map[method_name.to_s.tr('_', '').upcase]
+      
+      if known_name.nil?
+        @hash.respond_to?(known_name) || super
+      else
+        true
+      end
+    end
+
+    def initialize
+      @hash = {}
+    end
+
+    def [](key)
+      @hash[key]
+    end
+
+    def []=(key, value)
+      @hash[key] = value
     end
   end
 end

data/lib/file_data/formats/exif/exif_tags.rb

@@ -1,11 +1,14 @@
 module FileData
+  TagNameInfo = Struct.new(:name, :is_known);
+
   # Contains tag number to name information taken from the exif spec
   class ExifTags
     singleton_class.class_eval { attr_accessor :tag_groups }
     @tag_groups = {}
 
     def self.get_tag_name(ifd_id, tag_id)
-      get_known_name(ifd_id, tag_id) || "#{ifd_id}-#{tag_id}".to_sym
+      known_name = get_known_name(ifd_id, tag_id)
+      TagNameInfo.new(known_name || "#{ifd_id}-#{tag_id}".to_sym, known_name != nil)
     end
 
     def self.get_known_name(ifd_id, tag_id)

data/lib/file_data/version.rb

@@ -1,3 +1,3 @@
 module FileData
-  VERSION = '5.2.3'.freeze
+  VERSION = '6.0.0'.freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: file_data
 version: !ruby/object:Gem::Version
-  version: 5.2.3
+  version: 6.0.0
 platform: ruby
 authors:
 - Scott
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2021-10-21 00:00:00.000000000 Z
+date: 2021-11-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler