git_statistics 0.2.0 → 0.3.0

data/.travis.yml

@@ -1,7 +1,14 @@
 language: ruby
+
 rvm:
   - 1.9.2
   - 1.9.3
 
+bundler_args: --without development darwin
+
+before_install:
+  - sudo apt-get update
+  - sudo apt-get install libicu-dev
+
 notifications:
   email: false

data/CHANGELOG.md

@@ -0,0 +1,6 @@
+# CHANGELOG
+
+* [0.3.0 - September 12, 2012](https://github.com/kevinjalbert/git_statistics/compare/v0.2.0...v0.3.0)
+* [0.2.0 - September 6, 2012](https://github.com/kevinjalbert/git_statistics/compare/v0.1.2...v0.2.0)
+* [0.1.2 - April 12, 2012](https://github.com/kevinjalbert/git_statistics/compare/v0.1.1...v0.1.2)
+* [0.1.1 - April 12, 2012](https://github.com/kevinjalbert/git_statistics/compare/0e82e507e5b64a1140623c702015b97b1b3f7f81...v0.1.1)

data/Gemfile

@@ -4,9 +4,20 @@ gem "json"
 gem "trollop"
 gem "grit"
 gem "github-linguist"
+gem "os"
 
 group :test do
   gem "simplecov"
   gem "rspec"
   gem "rake"
 end
+
+group :development do
+  gem "guard"
+  gem "guard-rspec"
+end
+
+group :darwin do
+  gem "rb-fsevent"
+  gem "terminal-notifier-guard"
+end

data/Gemfile.lock

@@ -15,13 +15,21 @@ GEM
       diff-lcs (~> 1.1)
       mime-types (~> 1.15)
       posix-spawn (~> 0.3.6)
+    guard (1.3.2)
+      listen (>= 0.4.2)
+      thor (>= 0.14.6)
+    guard-rspec (1.2.1)
+      guard (>= 1.1)
     json (1.7.5)
+    listen (0.5.0)
     mime-types (1.19)
     multi_json (1.3.6)
+    os (0.9.6)
     posix-spawn (0.3.6)
     pygments.rb (0.2.13)
       rubypython (~> 0.5.3)
     rake (0.9.2.2)
+    rb-fsevent (0.9.1)
     rspec (2.11.0)
       rspec-core (~> 2.11.0)
       rspec-expectations (~> 2.11.0)
@@ -37,6 +45,8 @@ GEM
       multi_json (~> 1.0)
       simplecov-html (~> 0.5.3)
     simplecov-html (0.5.3)
+    terminal-notifier-guard (1.5.3)
+    thor (0.16.0)
     trollop (2.0)
 
 PLATFORMS
@@ -45,8 +55,13 @@ PLATFORMS
 DEPENDENCIES
   github-linguist
   grit
+  guard
+  guard-rspec
   json
+  os
   rake
+  rb-fsevent
   rspec
   simplecov
+  terminal-notifier-guard
   trollop

data/Guardfile

@@ -0,0 +1,7 @@
+# A sample Guardfile
+# More info at https://github.com/guard/guard#readme
+guard :rspec, :version => 2, :all_on_start => false do
+  watch(%r{^spec/.+_spec\.rb$})
+  watch(%r{^lib/(.+)\.rb$})     { |m| "spec/lib/#{m[1]}_spec.rb" }
+  watch('spec/spec_helper.rb')  { "spec" }
+end

data/README.md

@@ -26,11 +26,16 @@ This gem will analyze every commit within a git repository using `git log` and [
 * Total file renames
 * Total file copies
 
-This gem also uses [github/linguist](https://github.com/github/linguist) to determine the langugae of each individual file within commits. This augments the reported statistics by breaking down the author's statistics by languages.
+This gem also uses [github/linguist](https://github.com/github/linguist) to determine the language of each individual file within commits. This augments the reported statistics by breaking down the author's statistics by languages.
 
 This gem also has the ability to save the acquired data into a JSON file (in either a compressed or pretty format). If a saved file is present for the repository you can use the gem to load the data from the file, thus saving time for re-displaying the statistics using a different set of display flags (what statistic to sort on, number of authors to show, consider merges, etc...). In the event that a repository updates with new commits the gem allows you to update the saved file with the new commits.
 
-## Contributing
+# Example Output
+The following is the output produced by *git_statistics* when used on the [pengwynn/octokit](https://github.com/pengwynn/octokit) (at commit [95a9de3](https://github.com/pengwynn/octokit/commit/95a9de325bee4ca03c9c1d61de2d643666c90037)) git repository. In this output we show the top three authors in rankings based on number of commits (merge commits are excluded from these results).
+
+![screenshot](http://cloud.github.com/downloads/kevinjalbert/git_statistics/pengwynn_octokit_output.png)
+
+# Contributing
 
 1. Fork it
 2. Create your feature branch (`git checkout -b my-new-feature`)

data/git_statistics.gemspec

@@ -19,4 +19,5 @@ Gem::Specification.new do |gem|
   gem.add_dependency('trollop')
   gem.add_dependency('grit')
   gem.add_dependency('github-linguist')
+  gem.add_dependency('os')
 end

data/lib/git_statistics/collector.rb

@@ -6,16 +6,7 @@ module GitStatistics
     def initialize(verbose)
       @commits = Commits.new
       @verbose = verbose
-
-      # Connect to git repository if it exists
-      directory = Pathname.new(Dir.pwd)
-      while @repo == nil && !directory.root? do
-        begin
-          @repo = Grit::Repo.new(directory)
-        rescue
-          directory = directory.parent
-        end
-      end
+      @repo = Utilities.get_repository
 
       # Abort if no git repository is found
       if @repo == nil
@@ -23,86 +14,134 @@ module GitStatistics
       end
     end
 
-    def collect(branch, since="")
+    def collect(branch, time_since="", time_until="")
+      # Create pipe for git log to acquire branches
+      pipe = open("|git --no-pager branch --no-color")
+
       # Collect branches to use for git log
-      branches = collect_branches
+      branches = collect_branches(pipe)
       branches = ["", ""] if branch
 
       # Create pipe for the git log to acquire commits
       pipe = open("|git --no-pager log #{branches.join(' ')} --date=iso --reverse"\
                   " --no-color --find-copies-harder --numstat --encoding=utf-8 "\
-                  "--summary #{since} --format=\"%H,%an,%ae,%ad,%p\"")
+                  "--summary #{time_since} #{time_until} "\
+                  "--format=\"%H,%an,%ae,%ad,%p\"")
 
       # Use a buffer approach to queue up lines from the log for each commit
       buffer = []
       pipe.each do |line|
 
-        line = clean_string(line)
+        line = Utilities.clean_string(line)
 
         # Extract the buffer (commit) when we match ','x5 in the log format (delimeter)
         if line.split(',').size == 5
+
+          # Sometimes 'git log' doesn't populate the buffer, try fallback option if so
+          buffer = fall_back_collect_commit(line.split(',').first) if buffer.size == 1
+
           extract_commit(buffer) if not buffer.empty?
           buffer = []
         end
 
-        buffer << line.strip
+        buffer << line
       end
 
       # Extract the last commit
       extract_commit(buffer) if not buffer.empty?
     end
 
-    def collect_branches
-      # Create pipe for git log to acquire branches
-      pipe = open("|git --no-pager branch --no-color")
+    def fall_back_collect_commit(sha)
+
+      # Create pipe for the git log to acquire commits
+      pipe = open("|git --no-pager show #{sha} --date=iso --reverse"\
+                  " --no-color --find-copies-harder --numstat --encoding=utf-8 "\
+                  "--summary --format=\"%H,%an,%ae,%ad,%p\"")
+
+      buffer = []
+      pipe.each do |line|
+        buffer << Utilities.clean_string(line)
+      end
 
-      # Acquire all availble branches from repository
+      # Check that the buffer has valid information (i.e., sha was valid)
+      if buffer.empty?
+        return nil
+      elsif buffer.first.split(',').first == sha
+        return buffer
+      else
+        return nil
+      end
+    end
+
+    def collect_branches(pipe)
+      # Acquire all available branches from repository
       branches = []
       pipe.each do |line|
 
         # Remove the '*' leading the current branch
         line = line[1..-1] if line[0] == '*'
-        branches << clean_string(line)
+        branches << Utilities.clean_string(line)
       end
 
       return branches
     end
 
-    def extract_commit(buffer)
-      # Acquire general commit information
-      commit_info = buffer[0].split(',')
-      sha = commit_info[0]
+    def acquire_commit_data(line)
+      # Split up formated line
+      commit_info = line.split(',')
 
       # Initialize commit data
-      data = (@commits[sha] ||= Hash.new(0))
+      data = (@commits[commit_info[0]] ||= Hash.new(0))
       data[:author] = commit_info[1]
       data[:author_email] = commit_info[2]
       data[:time] = commit_info[3]
       data[:files] = []
 
-      # Flag commit as merge if nessecary (determined if two parents)
+      # Flag commit as merge if necessary (determined if two parents)
       if commit_info[4] == nil or commit_info[4].split(' ').size == 1
         data[:merge] = false
       else
         data[:merge] = true
       end
 
-      puts "Extracting #{sha}" if @verbose
+      return {:sha => commit_info[0], :data => data}
+    end
+
+    def extract_commit(buffer)
+      # Acquire general commit information
+      commit_data = acquire_commit_data(buffer[0])
+
+      puts "Extracting #{commit_data[:sha]}" if @verbose
+
+      # Abort if the commit sha extracted form the buffer is invalid
+      if commit_data[:sha].scan(/[\d|a-f]{40}/)[0] == nil
+        puts "Invalid buffer containing commit information"
+        return
+      end
 
       # Identify all changed files for this commit
-      files = identify_changed_files(buffer)
+      files = identify_changed_files(buffer[2..-1])
+
+      # No files were changed in this commit, abort commit
+      if files == nil
+        puts "No files were changed"
+        return
+      end
 
       # Acquire blob for each changed file and process it
       files.each do |file|
-        blob = get_blob(sha, file)
+        blob = get_blob(commit_data[:sha], file)
 
-        # Only process blobs, otherwise log problematic file/blob
+        # Only process blobs, or log the submodules and problematic files
         if blob.instance_of?(Grit::Blob)
-          process_blob(data, blob, file)
+          process_blob(commit_data[:data], blob, file)
+        elsif blob.instance_of?(Grit::Submodule)
+          puts "Ignoring submodule #{blob.name}"
         else
           puts "Problem processing file #{file[:file]}"
         end
       end
+      return commit_data[:data]
     end
 
     def get_blob(sha, file)
@@ -110,7 +149,7 @@ module GitStatistics
       file = file[:file].split(File::Separator)
 
       # Acquire blob of the file for this specific commit
-      blob = find_blob_in_tree(sha, @repo.tree(sha), file)
+      blob = Utilities.find_blob_in_tree(@repo.tree(sha), file)
 
       # If we cannot find blob in current commit (deleted file), check previous commit
       if blob == nil || blob.instance_of?(Grit::Tree)
@@ -118,94 +157,60 @@ module GitStatistics
         return nil if prev_commit == nil
 
         prev_tree = @repo.tree(prev_commit.id)
-        blob = find_blob_in_tree(prev_commit.id, prev_tree, file)
+        blob = Utilities.find_blob_in_tree(prev_tree, file)
       end
       return blob
     end
 
     def identify_changed_files(buffer)
-      # If the buffer is larger than 2 lines then we have per-file details to process
-      changed_files = []
-      if buffer.size > 2
-
-        # For each modification extract the details
-        buffer[2..-1].each do |line|
+      return buffer if buffer == nil
 
-          # Extract changed file information if it exists
-          data = extract_change_file(line)
-          if data != nil
-            changed_files << data
-            next  # This line is processed, skip to next
-          end
+      # For each modification extract the details
+      changed_files = []
+      buffer.each do |line|
 
-          # Extract details of create/delete files if it exists
-          data = extract_create_delete_file(line)
-          if data != nil
-            augmented = false
-            # Augment changed file with create/delete information if possible
-            changed_files.each do |file|
-              if file[:file] == data[:file]
-                file[:status] = data[:status]
-                augmented = true
-                break
-              end
-            end
-            changed_files << data if !augmented
-            next  # This line is processed, skip to next
-          end
+        # Extract changed file information if it exists
+        data = extract_change_file(line)
+        if data != nil
+          changed_files << data
+          next  # This line is processed, skip to next
+        end
 
-          # Extract details of rename/copy files if it exists
-          data = extract_rename_copy_file(line)
-          if data != nil
-            augmented = false
-            # Augment changed file with rename/copy information if possible
-            changed_files.each do |file|
-              if file[:file] == data[:new_file]
-                file[:status] = data[:status]
-                file[:old_file] = data[:old_file]
-                file[:similar] = data[:similar]
-                augmented = true
-                break
-              end
+        # Extract details of create/delete files if it exists
+        data = extract_create_delete_file(line)
+        if data != nil
+          augmented = false
+          # Augment changed file with create/delete information if possible
+          changed_files.each do |file|
+            if file[:file] == data[:file]
+              file[:status] = data[:status]
+              augmented = true
+              break
             end
-            changed_files << data if !augmented
-            next  # This line is processed, skip to next
           end
+          changed_files << data if !augmented
+          next  # This line is processed, skip to next
         end
-      end
-      return changed_files
-    end
-
-    def find_blob_in_tree(sha, tree, file)
-      # Check If cannot find tree in commit or if we found a submodule as the changed file
-      if tree == nil
-        return nil
-      elsif tree.instance_of?(Grit::Submodule)
-        return tree
-      end
-
-      # If the blob is within the current directory (tree)
-      if file.size == 1
-        blob = tree / file.first
-
-        # Check if blob is nil (could not find changed file in tree)
-        if blob == nil
 
-          # Try looking for submodules as they cannot be found using tree / file notation
-          tree.contents.each do |content|
-            if file.first == content.name
-              return nil
+        # Extract details of rename/copy files if it exists
+        data = extract_rename_copy_file(line)
+        if data != nil
+          augmented = false
+          # Augment changed file with rename/copy information if possible
+          changed_files.each do |file|
+            if file[:file] == data[:new_file]
+              file[:status] = data[:status]
+              file[:old_file] = data[:old_file]
+              file[:similar] = data[:similar]
+              augmented = true
+              break
             end
           end
-
-          # Exit through recusion with the base case of a nil tree/blob
-          return find_blob_in_tree(sha, blob, file)
+          changed_files << data if !augmented
+          next  # This line is processed, skip to next
         end
-        return blob
-      else
-        # Explore deeper in the tree to find the blob of the changed file
-        return find_blob_in_tree(sha, tree / file.first, file[1..-1])
       end
+      return changed_files
     end
 
     def process_blob(data, blob, file)
@@ -221,36 +226,21 @@ module GitStatistics
       data[:additions] += file[:additions]
       data[:deletions] += file[:deletions]
 
-      # Handle submodule if present, otherwise acquire specifics on blob
-      if blob.instance_of?(Grit::Submodule)
-        file_hash[:language] = "Submodule"
+      # Acquire specifics on blob
+      file_hash[:binary] = blob.binary?
+      file_hash[:image] = blob.image?
+      file_hash[:vendored] = blob.vendored?
+      file_hash[:generated] = blob.generated?
+
+      # Identify the language of the blob if possible
+      if blob.language == nil
+        file_hash[:language] = "Unknown"
       else
-        file_hash[:binary] = blob.binary?
-        file_hash[:image] = blob.image?
-        file_hash[:vendored] = blob.vendored?
-        file_hash[:generated] = blob.generated?
-
-        # Identify the language of the blob if possible
-        if blob.language == nil
-          file_hash[:language] = "Unknown"
-        else
-          file_hash[:language] = blob.language.name
-        end
+        file_hash[:language] = blob.language.name
       end
       data[:files] << file_hash
-    end
 
-    def clean_string(file_name)
-      #if file_name.include?("foo")
-      #blob = @repo.tree("1ec5c2674fd792e8f9ddbff5afcacc3e1f7c506d") / "actionpack" / "test" / "fixtures" / "public" / "foo"
-      #ap "=-=-=-=-=-=-="
-      #ap file_name
-      #ap "--------------------"
-      #ap blob.contents[2].name
-      #ap "=-=-=-=-=-=-="
-      #end
-      # Clean up a string and force utf-8 encoding
-      return file_name.strip.gsub('"', '').gsub("\\\\", "\\").force_encoding("utf-8")
+      return data
     end
 
     def extract_change_file(line)
@@ -258,11 +248,11 @@ module GitStatistics
       changes = line.scan(/^([-|\d]+)\s+([-|\d]+)\s+(.+)\s+=>\s+(.+)/)[0]
       if changes != nil and changes.size == 4
         # Split up the file into the old and new file
-        split_file = split_old_new_file(changes[2], changes[3])
+        split_file = Utilities.split_old_new_file(changes[2], changes[3])
         return {:additions => changes[0].to_i,
                 :deletions => changes[1].to_i,
-                :file => clean_string(split_file[:new_file]),
-                :old_file => clean_string(split_file[:old_file])}
+                :file => Utilities.clean_string(split_file[:new_file]),
+                :old_file => Utilities.clean_string(split_file[:old_file])}
       end
 
       # Use regex to detect a changed file | 1  2  /path/test/file.txt
@@ -270,7 +260,7 @@ module GitStatistics
       if changes != nil and changes.size == 3
         return {:additions => changes[0].to_i,
                 :deletions => changes[1].to_i,
-                :file => clean_string(changes[2])}
+                :file => Utilities.clean_string(changes[2])}
       end
       return nil
     end
@@ -279,8 +269,8 @@ module GitStatistics
       # Use regex to detect a create/delete file | create mode 100644 /path/test/file.txt
       changes = line.scan(/^(create|delete) mode \d+ ([^\\\n]*)/)[0]
       if changes != nil and changes.size == 2
-        return {:status => clean_string(changes[0]),
-                :file => clean_string(changes[1])}
+        return {:status => Utilities.clean_string(changes[0]),
+                :file => Utilities.clean_string(changes[1])}
       end
       return nil
     end
@@ -290,118 +280,13 @@ module GitStatistics
       changes = line.scan(/^(rename|copy)\s+(.+)\s+=>\s+(.+)\s+\((\d+)/)[0]
       if changes != nil and changes.size == 4
         # Split up the file into the old and new file
-        split_file = split_old_new_file(changes[1], changes[2])
-        return {:status => clean_string(changes[0]),
-                :old_file => clean_string(split_file[:old_file]),
-                :new_file => clean_string(split_file[:new_file]),
+        split_file = Utilities.split_old_new_file(changes[1], changes[2])
+        return {:status => Utilities.clean_string(changes[0]),
+                :old_file => Utilities.clean_string(split_file[:old_file]),
+                :new_file => Utilities.clean_string(split_file[:new_file]),
                 :similar => changes[3].to_i}
       end
       return nil
     end
-
-    def split_old_new_file(old, new)
-      # Split the old and new chunks up (separted by the =>)
-      split_old = old.split('{')
-      split_new = new.split('}')
-
-      # Handle recombine the file splits into their whole paths)
-      if split_old.size == 1 && split_new.size == 1
-        old_file = split_old[0]
-        new_file = split_new[0]
-      elsif split_new.size == 1
-        old_file = split_old[0] + split_old[1] + split_new[0]
-        new_file = split_old[0] + split_new[0]
-      elsif split_old.size == 1
-        old_file = split_old[0] + split_new[1]
-        new_file = split_old[0] + split_new[0] + split_new[1]
-      else
-        old_file = split_old[0] + split_old[1] + split_new[1]
-        new_file = split_old[0] + split_new[0] + split_new[1]
-      end
-
-      # Return files, yet remove the '//' if present from combining splits
-      return {:old_file => old_file.gsub('//', '/'),
-              :new_file => new_file.gsub('//', '/')}
-    end
-
-    def print_summary(sort_type, email, n=0)
-      # Default to a 0 if given a negative number to display
-      n = 0 if n < 0
-
-      # Acquire data based on sorty type and top # to show
-      data = @commits.author_top_n_type(sort_type, n)
-      if data == nil
-        raise "Parameter for --sort is not valid"
-      end
-
-      # Acquire formatting pattern for output
-      author_length = find_longest_author(data)
-      language_length = find_longest_language(data)
-      pattern = "%-#{author_length}s | %-#{language_length}s | %7s | %9s | %9s | %7s | %7s | %7s | %6s | %6s |"
-
-      # Print query/header information
-      print_header(pattern, sort_type, n, author_length, language_length)
-
-      # Print per author information
-      data.each do |key,value|
-        puts pattern % [key, "", value[:commits], value[:additions],
-                        value[:deletions], value[:create], value[:delete],
-                        value[:rename], value[:copy], value[:merges]]
-        print_language_data(pattern, value)
-      end
-
-      # Reprint query/header for repository information
-      print_header(pattern, sort_type, n, author_length, language_length)
-      data = @commits.totals
-      puts pattern % ["Repository Totals", "", data[:commits],
-                      data[:additions], data[:deletions], data[:create],
-                      data[:delete], data[:rename], data[:copy], data[:merges]]
-      print_language_data(pattern, data)
-    end
-
-    def print_language_data(pattern, data)
-      # Print information of each language for the data
-      data[:languages].each do |key,value|
-        puts pattern % ["", key, "", value[:additions], value[:deletions],
-                        value[:create], value[:delete], value[:rename],
-                        value[:copy], value[:merges]]
-      end
-    end
-
-    def print_header(pattern, sort_type, n, author_length, language_length)
-      total_authors = @commits.author_list.length
-
-      # Print summary information of displayed results
-      if n > 0 and n < total_authors
-        puts "\nTop #{n} authors(#{total_authors}) sorted by #{sort_type.to_s}\n"
-      else
-        puts "\nAll authors(#{total_authors}) sorted by #{sort_type.to_s}\n"
-      end
-
-      # Print column headers
-      puts "-"*87 + "-"*author_length + "-"*language_length
-      puts pattern % ['Name/Email', 'Language', 'Commits', 'Additions', 'Deletions', 'Creates', 'Deletes', 'Renames', 'Copies', 'Merges']
-      puts "-"*87 + "-"*author_length + "-"*language_length
-    end
-
-    def find_longest_author(data)
-      # Find the longest author name/email (for string formatting)
-      total_authors = @commits.author_list.length
-      author_length = 17
-      data.each do |key,value|
-        author_length = key.length if key.length > author_length
-      end
-      return author_length
-    end
-
-    def find_longest_language(data)
-      # Find the longest language name (for string formatting)
-      total_language = @commits.language_list.length
-      language_length = 9
-      @commits.language_list.each do |key,value|
-        language_length = key.length if key.length > language_length
-      end
-      return language_length
-    end
   end
 end