ipynbdiff 0.3.3 → 0.3.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 80e13b121205b2811ffb5d508e0978284562ced1b359156530b32ec2e10de6d4
-  data.tar.gz: 9a64f732b7e2ca0143bab893eb8e7713761f533d79798935d137d88d58f401e6
+  metadata.gz: d87c49e29a069b5064985b1b1097ac4d02c375161d9fe64df8e70f0155419220
+  data.tar.gz: 97931332962139f7f129a546bb7a7f9b670d779dcb032b27fbb1a3faa6f7dc85
 SHA512:
-  metadata.gz: 27f12477595add7a90f8f8d3ebf3b46b32ae24ec5c25a9417b03ff52d0be47ca68db501a2feca3337a0e8cf8ccee257cceccb7b9cbbf6716680641fb00a6a7f7
-  data.tar.gz: b1c3cd5e12a5f949be5e14c1dd2b69b93304b4b507787c91f8d7cc383855bcf52dabf2edbfe4a67e4410e20d575686f71dbfe5d2b7b4853678acf978555c4cd4
+  metadata.gz: 86c4be0fc53e0db2172221a17dd07e0c3f37492ffc699075b0de74dea3265feebee2207ef4e9170ead59be15c42b858918fbd2e56c7ed109cb674fbf7eb7dbbb
+  data.tar.gz: 10fdb124114a73425edc57f5c9fb80fbd546f6b8e7d58929a327f1be268774d8409238630286aa050e41384bf53c951d2b8c67aaeb76835bc3e0f37672775d48

data/README.md

@@ -1,3 +1,57 @@
-# rb-ipynbdiff: Better Jupyter Notebook diffs, in Ruby
+# IpynbDiff: Better diff for Jupyter Notebooks
 
-This is a port of [ipynbdiff](https://gitlab.com/gitlab-org/incubation-engineering/mlops/ipynbdiff)
+This is a simple diff tool that cleans up jupyter notebooks, transforming each [notebook](example/1/from.ipynb)
+into a [readable markdown file](example/1/from_html.md), keeping the output of cells, and running the
+diff after. Markdowns are generated using an opinionated Jupyter to Markdown conversion. This means
+that the entire file is readable on the diff.
+
+The result are diffs that are much easier to read:
+
+| Diff | IpynbDiff - HTML | IpynbDiff - Percent |  
+| ------ | ------ | ------ |
+| [Here](example/diff.txt) | [Here](example/ipynbdiff.txt) | [Here](example/ipynbdiff_percent.txt) |
+| ![](example/img/diff.png) | ![](example/img/ipynbdiff_html.png) | ![](example/img/ipynbdiff_percent.png) | 
+
+
+This started as a port of This is a port of [ipynbdiff](https://gitlab.com/gitlab-org/incubation-engineering/mlops/ipynbdiff), 
+but now has extended functionality although not working as git driver.
+
+## Usage
+
+### Generating diffs
+
+```ruby
+IpynbDiff.diff(from_path, to_path, options)
+```
+
+Options:
+
+```ruby
+@default_transform_options = {
+  preprocess_input: TRUE, # Whether the input should be transformed
+  write_output_to: nil, # Pass a path to save the output to a file
+  format: :text, # These are the formats Diffy accepts https://github.com/samg/diffy
+  sources_are_files: FALSE, # Weather to use the from/to as string or path to a file 
+  transform_options: @default_transform_options, # See below for transform options
+  diff_opts: {
+    include_diff_info: FALSE # These are passed to Diffy https://github.com/samg/diffy
+  }
+} 
+```
+
+### Transforming the notebooks
+
+It might be necessary to have the transformed files in addition to the diff.
+
+```ruby
+IpynbDiff.transform(notebook, options)
+```
+
+Options:
+
+```ruby
+@default_transform_options = {
+    include_metadata: FALSE, # Whether to include or not the notebook metadata (kernel, language, etc)
+    cell_decorator: :html # :html is useful to add styling with css, :percent is better for text format
+}
+```

data/ipynbdiff.gemspec

@@ -2,7 +2,7 @@
 
 Gem::Specification.new do |s|
   s.name        = 'ipynbdiff'
-  s.version     = '0.3.3'
+  s.version     = '0.3.4'
   s.summary     = 'Human Readable diffs for Jupyter Notebooks'
   s.description = 'Better diff for Jupyter Notebooks by first preprocessing them and removing clutter'
   s.authors     = ['Eduardo Bonet']

data/lib/ipynbdiff.rb

@@ -2,15 +2,20 @@
 
 # Human Readable Jupyter Diffs
 module IpynbDiff
-  require 'transformer.rb'
+  require 'transformer'
   require 'diffy'
 
-  @default_options = {
+  @default_transform_options = {
+    include_metadata: FALSE,
+    cell_decorator: :html
+  }
+
+  @default_diff_options = {
     preprocess_input: TRUE,
     write_output_to: nil,
     format: :text,
     sources_are_files: FALSE,
-    include_metadata: TRUE,
+    transform_options: @default_transform_options,
     diff_opts: {
       include_diff_info: FALSE
     }
@@ -19,7 +24,7 @@ module IpynbDiff
   def self.prepare_input(to_prepare, options)
     prepared = to_prepare
     prepared = File.read(prepared) if options[:sources_are_files]
-    prepared = Transformer.transform(prepared, include_metadata: options[:include_metadata]) if options[:preprocess_input]
+    prepared = transform(prepared, options[:transform_options]) if options[:preprocess_input]
 
     prepared
   end
@@ -29,7 +34,7 @@ module IpynbDiff
     to_notebook,
     options = {}
   )
-    options = @default_options.merge(options)
+    options = @default_diff_options.merge(options)
 
     from = from_notebook && prepare_input(from_notebook, options) || ''
     to = to_notebook && prepare_input(to_notebook, options) || ''
@@ -40,4 +45,9 @@ module IpynbDiff
 
     d
   end
+
+  def self.transform(notebook, options)
+    options = @default_transform_options.merge(options)
+    Transformer.new(**options).transform(notebook)
+  end
 end

data/lib/transformer.rb

@@ -6,88 +6,101 @@ module IpynbDiff
     require 'json'
     require 'yaml'
 
-    def self.transform(notebook, include_metadata: TRUE)
+    @cell_decorator = :html
+    @include_metadata = TRUE
+
+    def initialize(include_metadata: TRUE, cell_decorator: :html)
+      @include_metadata = include_metadata
+      @cell_decorator = cell_decorator
+    end
+
+    def transform(notebook)
       notebook_json = JSON.parse(notebook)
-      transformed_blocks = notebook_json['cells'].map { |cell| transform_cell(cell, notebook_json) }
-      transformed_blocks.prepend(transform_metadata(notebook_json)) if include_metadata
+      transformed_blocks = notebook_json['cells'].map do |cell|
+        decorate_cell(transform_cell(cell, notebook_json), cell)
+      end
+
+      transformed_blocks.prepend(transform_metadata(notebook_json)) if @include_metadata
       transformed_blocks.join("\n")
     end
 
-    def self.transform_cell(cell, notebook)
+    def decorate_cell(rows, cell)
+      tags = cell['metadata']&.fetch('tags', [])
+      type = cell['cell_type']
+
+      case @cell_decorator
+      when :html
+        rows.prepend(%(<div class="cell #{type}" data-id="#{cell['id']}" data-tags="#{tags&.join(' ')}">\n\n))
+            .append("\n</div>\n")
+      when :percent
+        rows.prepend(%(%% Cell type:#{type} id:#{cell['id']} tags:#{tags&.join(',')}\n\n))
+      else
+        rows
+      end.join('')
+    end
+
+    def transform_cell(cell, notebook)
       cell['cell_type'] == 'code' ? transform_code_cell(cell, notebook) : transform_text_cell(cell)
     end
 
-    def self.transform_code_cell(cell, notebook)
-      tags = cell['metadata']&.fetch('tags', [])&.join(' ') || ''
+    def decorate_output(output_rows, output)
+      if @cell_decorator == :html
+        output_rows.prepend(%(\n<div class="output #{output['output_type']}">\n\n)).append("\n</div>\n")
+      else
+        output_rows.prepend(%(\n%%%% Output: #{output['output_type']}\n\n))
+      end
+    end
 
+    def transform_code_cell(cell, notebook)
       [
-        %(<div class="cell code" data-id="#{cell['id']}" data-tags="#{tags}">\n\n),
         %(``` #{notebook['metadata']['kernelspec']['language']}\n),
         *cell['source'],
         "\n```\n",
-        *cell['outputs'].map { |output| transform_output(output) },
-        "\n</div>\n"
-      ].join('')
+        *cell['outputs'].map { |output| transform_output(output) }
+      ]
     end
 
-    def self.format_traceback(traceback)
+    def format_traceback(traceback)
       traceback.map do |t|
         t.split("\n").map do |line|
-          line.gsub(/\[[0-9][0-9;]*m/, '').sub("\u001B", '    ').gsub(/\u001B/, '').rstrip
+          line.gsub(/\[[0-9][0-9;]*m/, '').sub("\u001B", '    ').gsub(/\u001B/, '').rstrip << "\n"
         end
-      end.join("\n")
+      end
     end
 
-    def self.transform_execute_result(output)
-      [
-        %(\n<div class="output execute_result">\n\n),
-        *output['data']['text/plain'].map { |line| "    #{line}" },
-        "\n\n</div>\n"
-      ].join('')
+    def transform_execute_result(output)
+      output['data']['text/plain'].map { |line| "    #{line}" }.append("\n")
     end
 
-    def self.transform_image_result(output)
+    def transform_image_result(output)
       if output['data'].key?('image/png')
-        [
-          %(\n<div class="output display_data">\n\n),
-          "![](data:image/png;base64,#{output['data']['image/png'].gsub("\n", '')})",
-          "\n\n</div>\n"
-        ].join('')
+        ["![](data:image/png;base64,#{output['data']['image/png'].gsub("\n", '')})", "\n"]
       end
     end
 
-    def self.transform_error_result(output)
-      [
-        %(\n<div class="output error">\n\n),
-        format_traceback(output['traceback']),
-        "\n\n</div>\n"
-      ].join('')
+    def transform_error_result(output)
+      format_traceback(output['traceback'])
     end
 
-    def self.transform_output(output)
-      case output['output_type']
-      when 'execute_result'
-        transform_execute_result(output)
-      when 'display_data'
-        transform_image_result(output)
-      when 'error'
-        transform_error_result(output)
-      end
-    end
+    def transform_output(output)
+      transformed =
+        case output['output_type']
+        when 'execute_result'
+          transform_execute_result(output)
+        when 'display_data'
+          transform_image_result(output)
+        when 'error'
+          transform_error_result(output)
+        end
 
-    def self.transform_text_cell(cell)
-      tags = cell['metadata']&.fetch('tags', [])&.join(' ') || ''
-      id = cell['id']
-      cell_type = cell['cell_type']
+      decorate_output(transformed, output).join('') if transformed
+    end
 
-      [
-        %(<div class="cell #{cell_type}" data-id="#{id}" data-tags="#{tags}">\n\n),
-        *cell['source'],
-        "\n\n</div>\n"
-      ].join('')
+    def transform_text_cell(cell)
+      cell['source'].append("\n")
     end
 
-    def self.transform_metadata(notebook_json)
+    def transform_metadata(notebook_json)
       {
         'jupyter' => {
           'kernelspec' => notebook_json['metadata']['kernelspec'],

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: ipynbdiff
 version: !ruby/object:Gem::Version
-  version: 0.3.3
+  version: 0.3.4
 platform: ruby
 authors:
 - Eduardo Bonet
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-10-13 00:00:00.000000000 Z
+date: 2021-10-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: diffy