Fixes, optimizations and pretty up

Author: shayonj

.gitignore

@@ -15,4 +15,5 @@
 
 node_modules
 
-.db
+*.db
+.DS_Store

README.md

@@ -4,6 +4,8 @@
 
 You can now easily run, any kind of analytics on your Git directory using the SQLite database.
 
+![plot](./internal/screenshot.png)
+
 ## Features ✨
 
 - Synchronize Git repository data into a SQLite database.
@@ -64,11 +66,13 @@ Once your repository data is synchronized into a SQLite database, you can run va
 6. **Authors Who Have Worked on a Specific File**
 
    ```sql
-    SELECT DISTINCT commits.author
-    FROM commits
-    JOIN commit_files ON commits.commit_hash = commit_files.commit_hash
-    JOIN files ON commit_files.file_id = files.file_id
-    WHERE files.file_path like '%sqlite3%'
+   SELECT files.file_path, commits.author, COUNT(*) as times_contributed
+   FROM commits
+   JOIN commit_files ON commits.commit_hash = commit_files.commit_hash
+   JOIN files ON commit_files.file_id = files.file_id
+   WHERE files.file_path LIKE '%connection_adapters/sqlite%'
+   GROUP BY files.file_path, commits.author
+   ORDER BY times_contributed DESC;
    ```
 
 ## Installation 📥

internal/screenshot.jpg

@@ -1,9 +1,25 @@
 # frozen_string_literal: true
 
+require "logger"
 require "branch_base/database"
 require "branch_base/repository"
 require "branch_base/sync"
 require "branch_base/cli"
 
 module BranchBase
+  def self.logger
+    @logger ||=
+      Logger
+        .new($stdout)
+        .tap do |log|
+          log.progname = "BranchBase"
+
+          log.level = ENV["DEBUG"] ? Logger::DEBUG : Logger::INFO
+
+          log.formatter =
+            proc do |severity, datetime, progname, msg|
+              "#{datetime}: #{severity} - #{progname}: #{msg}\n"
+            end
+        end
+  end
 end

internal/screenshot.png

@@ -5,25 +5,33 @@
 
 module BranchBase
   class CLI < Thor
-    desc "sync REPO_PATH [BRANCH_OR_TAG]",
-         "Synchronize a specific branch or tag of the Git repository with the SQLite database"
+    desc "sync REPO_PATH", "Synchronize a Git directory to a SQLite database"
     def sync(repo_path)
+      BranchBase.logger.info("Starting sync process for #{repo_path}...")
+
       full_repo_path = File.expand_path(repo_path)
 
       unless File.directory?(File.join(full_repo_path, ".git"))
-        puts "The specified path is not a valid Git repository: #{full_repo_path}"
+        BranchBase.logger.error(
+          "The specified path is not a valid Git repository: #{full_repo_path}",
+        )
         exit(1)
       end
 
       repo_name = File.basename(full_repo_path)
-      db_filename = "#{repo_name}_git_data.db"
+      db_directory = full_repo_path
+      db_filename = File.join(db_directory, "#{repo_name}_git_data.db")
 
       database = Database.new(db_filename)
       repository = Repository.new(full_repo_path)
+      start_time = Time.now
       sync = Sync.new(database, repository)
 
       sync.run
-      puts "Repository data synced successfully for"
+      elapsed_time = Time.now - start_time
+      BranchBase.logger.info(
+        "Repository data synced successfully in #{db_filename} in #{elapsed_time.round(2)} seconds",
+      )
     end
   end
 end

lib/branch_base.rb

@@ -2,6 +2,7 @@
 
 module BranchBase
   class Sync
+    # TODO acctualy see if bulk inserts are faster
     BATCH_SIZE = 1000
 
     def initialize(database, repository)
@@ -38,6 +39,10 @@ def sync_repository
     end
 
     def sync_branches(repo_id)
+      BranchBase.logger.debug(
+        "Syncing branches for repository ID: #{@repo.path}",
+      )
+
       batched_branches = []
 
       @repo.branches.each do |branch|
@@ -127,6 +132,10 @@ def insert_commit_files(commit, repo_id)
     end
 
     def insert_commit_parents(commit)
+      BranchBase.logger.debug(
+        "Inserting parent commits for repository: #{@repo.path}",
+      )
+
       commit.parent_ids.each do |parent_id|
         @db.execute(
           "INSERT INTO commit_parents (commit_hash, parent_hash) VALUES (?, ?)",
@@ -147,13 +156,27 @@ def insert_branches(batched_branches)
     end
 
     def insert_commits(batched_commits)
+      BranchBase.logger.debug(
+        "Inserting commits for repository ID: #{@repo.path}",
+      )
+
+      return if batched_commits.empty?
+
       @db.transaction do
-        batched_commits.each do |data|
-          @db.execute(
-            "INSERT INTO commits (commit_hash, repo_id, author, committer, message, timestamp) VALUES (?, ?, ?, ?, ?, ?)",
-            data,
-          )
-        end
+        values_clause =
+          batched_commits
+            .map do |commit_data|
+              "(#{commit_data.map { |value| "'#{value}'" }.join(", ")})"
+            end
+            .join(", ")
+
+        sql = <<~SQL
+          INSERT INTO commits
+          (commit_hash, repo_id, author, committer, message, timestamp)
+          VALUES #{values_clause};
+        SQL
+
+        @db.execute(sql)
       end
     end
 

lib/branch_base/cli.rb

@@ -1,6 +1,5 @@
 # frozen_string_literal: true
-require "branch_base/database"
-require "rspec"
+require "spec_helper"
 
 RSpec.describe(BranchBase::Database) do
   let(:database) { BranchBase::Database.new(":memory:") }

lib/branch_base/sync.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
-require "branch_base/repository"
-require "rugged"
-require "rspec"
+require "spec_helper"
 require "test_helper"
 
 RSpec.describe(BranchBase::Repository) do
@@ -24,7 +22,7 @@
       repository.walk { |commit| commit_messages << commit.message.strip }
       expect(commit_messages).to contain_exactly(
         "Add contributing guidelines",
-        "Initial commit"
+        "Initial commit",
       )
     end
   end

spec/branch_base/cli_spec.rb

@@ -1,9 +1,6 @@
 # frozen_string_literal: true
-require "branch_base/sync"
-require "branch_base/database"
-require "branch_base/repository"
 require "test_helper"
-require "rspec"
+require "spec_helper"
 
 RSpec.describe(BranchBase::Sync) do
   let(:db) { BranchBase::Database.new(":memory:") }

spec/branch_base/database_spec.rb

@@ -13,6 +13,8 @@
 # the additional setup, and require it from the spec files that actually need
 # it.
 #
+require "branch_base"
+
 # See https://rubydoc.info/gems/rspec-core/RSpec/Core/Configuration
 RSpec.configure do |config|
   # rspec-expectations config goes here. You can use an alternate