hubba 0.6.0 → 0.6.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 63505774303ed6881c5650478eb76a1df7a3d412
-  data.tar.gz: 517173da9e008bd5d9ea74d2e9d8d1622edf7fdf
+  metadata.gz: 122527cdc3cddb68e2c54a6e51fcfb77c2c91c0c
+  data.tar.gz: 6bb52baff0536f55ad969885fc8500b031457e44
 SHA512:
-  metadata.gz: 8a806d4ee314cc13c1605781cbc6af7192600d2c34ec9ebccc949541a800a41d93d43def28600aa0d27e7526a47afa1e319bc2184ec4cee1c917196cc080ca91
-  data.tar.gz: bcc42747f508c0acab55a8f652be1103afef20d9409172081bba34a530a9bbc441a2d5ec109bcd4b43e3049a3b3f74fc1c478d17e9c84a7e7ce41e7dabd11014
+  metadata.gz: df4e036a205be30371ec9de85e8085717235922fcc21e22f9f516cc40481a2a8cf6a9b0edd0a37cb24759059a08b5ea74957f41f89139bf9e1a860ed20029e3b
+  data.tar.gz: 698fc0d240aa6c6f8b125d1ae004beacf5ae97c220eb42d4fd542a49a656ece101e3a9fd9e740eaefb63e7f03c132458872f4ea5949cd30e2eceb9ccc3bb9e54

data/Manifest.txt

@@ -6,6 +6,8 @@ lib/hubba.rb
 lib/hubba/client.rb
 lib/hubba/config.rb
 lib/hubba/github.rb
+lib/hubba/reports.rb
+lib/hubba/reposet.rb
 lib/hubba/stats.rb
 lib/hubba/version.rb
 test/helper.rb

data/lib/hubba.rb

@@ -1,6 +1,24 @@
 # 3rd party (our own)
 require 'webclient'
 
+
+###############
+## helpers
+
+def save_json( path, data )    ## data - hash or array
+  File.open( path, 'w:utf-8' ) do |f|
+    f.write( JSON.pretty_generate( data ))
+  end
+end
+
+def save_yaml( path, data )
+  File.open( path, 'w:utf-8' ) do |f|
+    f.write( data.to_yaml )
+  end
+end
+
+
+
 # our own code
 require 'hubba/version'   # note: let version always go first
 require 'hubba/config'
@@ -8,6 +26,11 @@ require 'hubba/client'
 require 'hubba/github'
 require 'hubba/stats'
 
+## "higher level" porcelain services / helpers for easy (re)use
+require 'hubba/reposet'
+
+require 'hubba/reports'
+
 
 ############
 # add convenience alias for alternate spelling - why? why not?

data/lib/hubba/client.rb

@@ -57,11 +57,11 @@ def get( request_uri )
   # => "text/html; charset=UTF-8"
 
   # Iterate all response headers.
-  puts "HTTP HEADERS:"
-  res.headers.each do |key, value|
-    puts "  #{key}: >#{value}<"
-  end
-  puts
+  # puts "HTTP HEADERS:"
+  # res.headers.each do |key, value|
+  #  puts "  #{key}: >#{value}<"
+  # end
+  # puts
 
   # => "location => http://www.google.com/"
   # => "content-type => text/html; charset=UTF-8"

data/lib/hubba/config.rb

@@ -1,8 +1,13 @@
 module Hubba
 
 class Configuration
-  def data_dir()         @data_dir || './data'; end
-  def data_dir=( value ) @data_dir = value;     end
+  def data_dir()          @data_dir || './data'; end
+  def data_dir=( value )  @data_dir = value;     end
+
+  ###  todo/check: rename to/use tmp_dir - why? why not?
+  def cache_dir()         @cache_dir || './cache'; end
+  def cache_dir=( value ) @cache_dir = value;      end
+
 
   # try default setup via ENV variables
   def token()            @token || ENV[ 'HUBBA_TOKEN' ]; end

data/lib/hubba/github.rb

@@ -87,19 +87,47 @@ end
 # more
 def update( obj )
   if obj.is_a?( Stats )
-    full_name = obj.full_name
+    stats = obj
+    full_name = stats.full_name
     puts "[update 1/2] fetching repo >#{full_name}<..."
     repo    = repo( full_name )
     puts "[update 2/2] fetching repo >#{full_name}< commits ..."
     commits = repo_commits( full_name )
 
-    obj.update( repo, commits )
+    stats.update( repo, commits )
   else
     raise ArgumentError, "unknown source object passed in - expected Hubba::Stats; got #{obj.class.name}"
   end
 end
 
 
+def update_stats( h )    ## todo/fix: change to Reposet - why? why not???
+    h.each do |org_with_counter,names|
+
+      ## remove optional number from key e.g.
+      ##   mrhydescripts (3)    =>  mrhydescripts
+      ##   footballjs (4)       =>  footballjs
+      ##   etc.
+
+      org = org_with_counter.sub( /\([0-9]+\)/, '' ).strip
+
+      ## puts "  -- #{key_with_counter} [#{key}] --"
+
+      names.each do |name|
+        full_name = "#{org}/#{name}"
+
+        ## puts "  fetching stats #{count+1}/#{repo_count} - >#{full_name}<..."
+        stats = Stats.new( full_name )
+        stats.read
+
+        update( stats )   ## fetch & update stats
+
+        stats.write
+      end
+    end
+end
+
+
 private
 def get( request_uri )
   @client.get( request_uri )

data/lib/hubba/reports.rb

@@ -0,0 +1,249 @@
+module Hubba
+
+
+class Report
+  def initialize( stats_or_hash_or_path=Hubba.stats )
+    ## puts "[debug] Report#initialize:"
+    ## pp  stats_or_hash_or_path   if stats_or_hash_or_path.is_a?( String )
+
+    @stats = if stats_or_hash_or_path.is_a?( String ) ||
+                stats_or_hash_or_path.is_a?( Hash )
+                  hash_or_path = stats_or_hash_or_path
+                  Hubba.stats( hash_or_path )
+             else
+                  stats_or_hash_or_path  ## assume Summary/Stats - todo/fix: double check!!!
+             end
+  end
+
+  def save( path )
+    buf = build
+    puts "writing report >#{path}< ..."
+    File.open( path, "w:utf-8" ) do |f|
+      f.write( buf )
+    end
+  end
+end  ## class Report
+
+
+
+class ReportSummary < Report
+
+def build
+## create a (summary report)
+##
+##  add stars, last_updates, etc.
+##  org description etc??
+
+## note: orgs is orgs+users e.g. geraldb, yorobot etc
+buf = String.new('')
+buf << "# #{@stats.repos.size} repos @ #{@stats.orgs.size} orgs\n"
+buf << "\n"
+
+
+@stats.orgs.each do |org|
+  name  = org[0]
+  repos = org[1]
+  buf << "### #{name} _(#{repos.size})_\n"
+  buf << "\n"
+
+  ### add stats for repos
+  entries = []
+  repos.each do |repo|
+    entries << "**#{repo.name}** ★#{repo.stats.stars} (#{repo.stats.size} kb)"
+  end
+
+  buf << entries.join( ' · ' )   ## use interpunct? - was: • (bullet)
+  buf << "\n"
+  buf << "\n"
+end
+
+buf
+end  # method build
+end  # class ReportSummary
+
+
+
+class ReportStars < Report
+
+def build
+
+##  add stars, last_updates, etc.
+##  org description etc??
+
+## note: orgs is orgs+users e.g. geraldb, yorobot etc
+buf = String.new('')
+buf << "# #{@stats.repos.size} repos @ #{@stats.orgs.size} orgs\n"
+buf << "\n"
+
+
+repos = @stats.repos.sort do |l,r|
+  ## note: use reverse sort (right,left) - e.g. most stars first
+  r.stats.stars <=> l.stats.stars
+end
+
+## pp repos
+
+repos.each_with_index do |repo,i|
+  buf << "#{i+1}. ★#{repo.stats.stars} **#{repo.full_name}** (#{repo.stats.size} kb)\n"
+end
+
+buf
+end  # method build
+end  # class ReportStars
+
+
+
+class ReportTimeline < Report
+
+def build
+## create a (timeline report)
+
+## note: orgs is orgs+users e.g. geraldb, yorobot etc
+buf = String.new('')
+buf << "# #{@stats.repos.size} repos @ #{@stats.orgs.size} orgs\n"
+buf << "\n"
+
+
+repos = @stats.repos.sort do |l,r|
+  ## note: use reverse sort (right,left) - e.g. most stars first
+  ## r[:stars] <=> l[:stars]
+
+  ## sort by created_at (use julian days)
+  r.stats.created.jd <=> l.stats.created.jd
+end
+
+
+## pp repos
+
+
+last_year  = -1
+last_month = -1
+
+repos.each_with_index do |repo,i|
+  year       = repo.stats.created.year
+  month      = repo.stats.created.month
+
+  if last_year != year
+    buf << "\n## #{year}\n\n"
+  end
+
+  if last_month != month
+    buf << "\n### #{month}\n\n"
+  end
+
+  last_year  = year
+  last_month = month
+
+  buf << "- #{repo.stats.created_at.strftime('%Y-%m-%d')} ★#{repo.stats.stars} **#{repo.full_name}** (#{repo.stats.size} kb)\n"
+end
+
+buf
+end  # method build
+end  # class ReportTimeline
+
+
+
+class ReportTrending < Report
+
+def build
+
+## note: orgs is orgs+users e.g. geraldb, yorobot etc
+buf = String.new('')
+buf << "# #{@stats.repos.size} repos @ #{@stats.orgs.size} orgs\n"
+buf << "\n"
+
+###
+## todo:
+##   use calc per month (days: 30)
+##   per week is too optimistic (e.g. less than one star/week e.g. 0.6 or something)
+
+repos = @stats.repos.sort do |l,r|
+  ## note: use reverse sort (right,left) - e.g. most stars first
+  ## r[:stars] <=> l[:stars]
+
+  ## sort by created_at (use julian days)
+  ## r[:created_at].to_date.jd <=> l[:created_at].to_date.jd
+
+  res = r.diff <=> l.diff
+  res = r.stats.stars <=> l.stats.stars  if res == 0
+  res = r.stats.created.jd <=> l.stats.created.jd  if res == 0
+  res
+end
+
+
+## pp repos
+
+
+repos.each_with_index do |repo,i|
+  if repo.diff == 0
+    buf << "-  -/- "
+  else
+    buf << "- #{repo.diff}/month "
+  end
+
+  buf <<  " ★#{repo.stats.stars} **#{repo.full_name}** (#{repo.stats.size} kb) - "
+  buf <<  "#{repo.stats.history_str}\n"
+end
+
+
+buf
+end  # method build
+end  # class ReportTrending
+
+
+
+class ReportUpdates < Report
+
+def build
+
+## note: orgs is orgs+users e.g. geraldb, yorobot etc
+buf = String.new('')
+buf << "# #{@stats.repos.size} repos @ #{@stats.orgs.size} orgs\n"
+buf << "\n"
+
+repos = @stats.repos.sort do |l,r|
+  r.stats.committed.jd <=> l.stats.committed.jd
+end
+
+## pp repos
+
+
+buf << "committed / pushed / updated / created\n\n"
+
+today = Date.today
+
+repos.each_with_index do |repo,i|
+
+  days_ago = today.jd - repo.stats.committed.jd
+
+  diff1 = repo.stats.committed.jd - repo.stats.pushed.jd
+  diff2 = repo.stats.committed.jd - repo.stats.updated.jd
+  diff3 = repo.stats.pushed.jd    - repo.stats.updated.jd
+
+  buf <<  "- (#{days_ago}d) **#{repo.full_name}** ★#{repo.stats.stars} - "
+  buf <<  "#{repo.stats.committed} "
+  buf <<  "("
+  buf <<  (diff1==0 ? '=' : "#{diff1}d")
+  buf <<  "/"
+  buf <<  (diff2==0 ? '=' : "#{diff2}d")
+  buf <<  ")"
+  buf <<  " / "
+  buf <<  "#{repo.stats.pushed} "
+  buf <<  "("
+  buf <<  (diff3==0 ? '=' : "#{diff3}d")
+  buf <<  ")"
+  buf <<  " / "
+  buf <<  "#{repo.stats.updated} / "
+  buf <<  "#{repo.stats.created} - "
+  buf <<  "‹#{repo.stats.last_commit_message}›"
+  buf <<  " (#{repo.stats.size} kb)"
+  buf <<  "\n"
+end
+
+
+buf
+end  # method build
+end  # class ReportUpdates
+
+
+end # module Hubba

data/lib/hubba/reposet.rb

@@ -0,0 +1,170 @@
+module Hubba
+
+
+def self.update_stats( host_or_path='./repos.yml' )  ## move to reposet e.g. Reposet#update_status!!!!
+  h = if hash_or_path.is_a?( String )    ## assume it is a file path!!!
+        path = hash_or_path
+        YAML.load_file( path )
+      else
+        hash_or_path  # assume its a hash / reposet already!!!
+      end
+
+  gh = Github.new
+  gh.update_stats( h )
+end
+
+
+def self.stats( hash_or_path='./repos.yml' )   ## use read_stats or such - why? why not?
+  h = if hash_or_path.is_a?( String )    ## assume it is a file path!!!
+        path = hash_or_path
+        YAML.load_file( path )
+      else
+        hash_or_path  # assume its a hash / reposet already!!!
+      end
+
+  Summary.new( h )  ## wrap in "easy-access" facade / wrapper
+end
+
+
+
+class Summary    # todo/check: use a different name e.g (Data)Base, Census, Catalog, Collection, Index, Register or such???
+
+class Repo  ## (nested) class
+
+  attr_reader :owner,
+              :name
+
+  def initialize( owner, name )
+    @owner = owner   ## rename to login, username - why? why not?
+    @name  = name    ## rename to reponame ??
+  end
+
+  def full_name() "#{owner}/#{name}"; end
+
+  def stats
+    ## note: load stats on demand only (first access) for now - why? why not?
+    @stats ||= begin
+                 stats = Stats.new( full_name )
+                 stats.read
+                 stats
+               end
+  end
+
+  def diff
+    @diff ||= stats.calc_diff_stars( samples: 3, days: 30 )
+  end
+end  # (nested) class Repo
+
+
+attr_reader :orgs, :repos
+
+def initialize( hash )
+  @orgs     = []    # orgs and users -todo/check: use better name - logins or owners? why? why not?
+  @repos    = []
+  add( hash )
+
+  puts "#{@repos.size} repos @ #{@orgs.size} orgs"
+end
+
+#############
+## private helpes
+def add( hash )   ## add repos.yml set
+  hash.each do |org_with_counter, names|
+    ## remove optional number from key e.g.
+    ##   mrhydescripts (3)    =>  mrhydescripts
+    ##   footballjs (4)       =>  footballjs
+    ##   etc.
+    org = org_with_counter.sub( /\([0-9]+\)/, '' ).strip
+    repos = []
+    names.each do |name|
+      repo = Repo.new( org, name )
+      repos << repo
+    end
+    @orgs << [org, repos]
+    @repos += repos
+  end
+end
+end  # class Summary
+
+
+
+## orgs  - include repos form org(anizations) too
+## cache - save json response to cache_dir   - change to/use debug/tmp_dir? - why? why not?
+def self.reposet( *users, orgs: true,
+                          cache: false )
+  # users = [users]   if users.is_a?( String )  ### wrap in array if single user
+
+  gh = Hubba::Github.new
+
+  forks = []
+
+  h = {}
+  users.each do |user|
+    res = gh.user_repos( user )
+    save_json( "#{config.cache_dir}/users~#{user}~repos.json", res.data )  if cache
+
+    repos = []
+    ####
+    #  check for forked repos (auto-exclude personal by default)
+    #   note: forked repos in orgs get NOT auto-excluded!!!
+    res.data.each do |repo|
+      fork = repo['fork']
+      if fork
+        print "FORK "
+        forks << "#{repo['full_name']} (AUTO-EXCLUDED)"
+      else
+        print "     "
+        repos << repo['name']
+      end
+      print repo['full_name']
+      print "\n"
+    end
+
+
+    h[ "#{user} (#{repos.size})" ] = repos.sort
+  end
+
+
+  ## all repos from orgs
+  ##  note: for now only use first (primary user) - why? why not?
+  if orgs
+    user = users[0]
+    res = gh.user_orgs( user )
+    save_json( "#{config.cache_dir}/users~#{user}~orgs.json", res.data )  if cache
+
+
+    logins = res.logins.each do |login|
+       ## next if ['xxx'].include?( login )      ## add orgs here to skip
+
+       res = gh.org_repos( login )
+       save_json( "#{config.cache_dir}/orgs~#{login}~repos.json", res.data )  if cache
+
+       repos = []
+       res.data.each do |repo|
+         fork = repo['fork']
+         if fork
+           print "FORK "
+           forks << repo['full_name']
+           repos << repo['name']
+         else
+           print "     "
+           repos << repo['name']
+         end
+         print repo['full_name']
+         print "\n"
+       end
+
+       h[ "#{login} (#{repos.size})" ] = repos.sort
+    end
+  end
+
+  if forks.size > 0
+    puts
+    puts "#{forks.size} fork(s):"
+    puts forks
+  end
+
+  h
+end  ## method reposet
+end # module Hubba
+

data/lib/hubba/stats.rb

@@ -189,7 +189,7 @@ module Hubba
       ##  todo: check for better way (convert to float upfront - why? why not?)
 
       diff = (diff_stars * days * 1000) / diff_days
-      puts "diff=#{diff}:#{diff.class.name}"    ## check if it's a float
+      ##  puts "diff=#{diff}:#{diff.class.name}"    ## check if it's a float
       (diff.to_f/1000.0)
     end
   end
@@ -284,7 +284,7 @@ module Hubba
     def write
       basename = full_name.gsub( '/', '~' )   ## e.g. poole/hyde become poole~hyde
       data_dir = Hubba.config.data_dir
-      puts "writing stats to #{basename} (#{data_dir})..."
+      puts "  writing stats to #{basename} (#{data_dir})..."
 
       ## todo/fix: add FileUtils.makepath_r or such!!!
       File.open( "#{data_dir}/#{basename}.json", 'w:utf-8' ) do |f|
@@ -301,7 +301,7 @@ module Hubba
       path = "#{data_dir}/#{basename}.json"
 
       if File.exist?( path )
-        puts "reading stats from #{basename} (#{data_dir})..."
+        puts "  reading stats from #{basename} (#{data_dir})..."
         json = File.open( path, 'r:utf-8' ) { |f| f.read }
         @data = JSON.parse( json )
 

data/lib/hubba/version.rb

@@ -1,7 +1,7 @@
 module Hubba
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
   MINOR = 6
-  PATCH = 0
+  PATCH = 1
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: hubba
 version: !ruby/object:Gem::Version
-  version: 0.6.0
+  version: 0.6.1
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-10-08 00:00:00.000000000 Z
+date: 2020-10-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: webclient
@@ -75,6 +75,8 @@ files:
 - lib/hubba/client.rb
 - lib/hubba/config.rb
 - lib/hubba/github.rb
+- lib/hubba/reports.rb
+- lib/hubba/reposet.rb
 - lib/hubba/stats.rb
 - lib/hubba/version.rb
 - test/helper.rb