infod 0.0.2 → 0.0.3

data/infod/mail.rb

@@ -0,0 +1,90 @@
+#watch __FILE__
+class E
+
+  begin 
+    require 'tmail'
+  rescue LoadError => e
+  end
+
+  def triplrTmail &f
+    messagePath = ->id{
+      h = id.h # hash
+      '/msg/' + h[0..1] + '/' + h[2] + '/' + id}
+
+    (TMail::Mail.load node).do{|m|      # load
+      d = m.message_id; return unless d # parse successful?
+      id = d[1..-2]                     # message-ID
+      e = messagePath[id]               # webized ID
+      yield e, DC+'identifier', id      # original ID
+      yield e, DC+'source', self        # original file
+      yield e, Type, E[SIOCt + 'MailMessage']
+      yield e, Type, E[SIOC  + 'Post']
+      yield e, Date, m.date.iso8601 if m.date
+      yield e, Title, m.subject.to_utf8
+      yield e, SIOC+'name', m.friendly_from.to_utf8
+      yield e, Creator, E['/m/'+m.from[0].to_utf8]
+      m.header['x-original-to'].do{|f|
+        yield e, SIOC+'reply_to', E[URI.escape "mailto:#{f}?References=<#{e}>&In-Reply-To=<#{e}>&Subject=#{m.subject.to_utf8}"] }
+      %w{to cc bcc}.map{|to|
+        m.send(to).do{|to| to.map{|to|
+          yield e, To, E['/m/'+to.to_utf8]}}}
+      %w{in_reply_to references}.map{|ref|
+        m.send(ref).do{|refs| refs.map{|r|
+          yield e, SIOC+'reply_of', E[messagePath[r[1..-2]]]}}}
+                           # minimal local markup to use as HTML-literal even if decoupled from specialized view
+      yield e, Content, H([{_: :pre, class: :mail, style: 'white-space: pre-wrap',
+                            c: m.concat_message(e.E,0,&f).gsub(/^\s*(&gt;)(&gt;|\s)*\n/,"").lines.to_a.map{|l| # < skip quoted emptylines  v tag quoted lines
+                              {_: :span, class: ((l.match /(^\s*(&gt;|On[^\n]+(said|wrote))[^\n]*)\n/) ? 'q' : 'u'), c: [ l.chomp, "\n" ]}}},
+                           {_: :style, c: "pre.mail .q {background-color:#0018ff;color:#fff}\npre.mail a {background-color: #91acb3;color:#fff}\npre.mail img {max-width:100%}"}])}
+  rescue Exception => e
+    puts e
+  end
+
+  def triplrMailMessage &f
+    insertDocs :triplrTmail, nil, [To,SIOC+'reply_of'], &f
+  end
+=begin
+ there's another mail library called Mail, as of v2.5.4 takes 50x as long as tmail (apt-get install ruby-tmail)
+HEAD 200 http://m/m/2013/12/01/?nocache=&triplr=triplrMail curl/7.33.0  5.4003388
+HEAD 200 http://m/m/2013/12/01/?nocache=&triplr=triplrTmail curl/7.33.0  0.1198720
+
+almost a copy of above works but identifiers are not wrapped in <> - with caching it might be fast enough..
+
+=end
+
+end
+
+module TMail
+  class Mail
+    def unicode_body
+      unquoted_body.to_utf8
+    end
+    def concat_message i, partCount=0, &f
+      if multipart?
+        parts.map{|part|
+          if part.multipart?   # and even more nested parts..
+            part.concat_message i, partCount, &f
+          elsif !attachment?(part) && part.sub_type != 'html'
+            part.unicode_body.hrefs true
+          else # attachment
+            a = i.a('.attache').mk # create container
+            p = a.as(part['content-type']['name'] || (partCount.to_s + '.' + (E::MIME.invert[part.content_type] || '.bin').to_s))
+            p.w part.body if !p.e # write attachment into message container
+            partCount += 1        # display images
+            yield i.uri, E::SIOC+'attachment', p
+            '<a href="'+p.uri+'">'+(part.main_type=='image' ? '<img src="'+p.uri+'">' : '')+p.uri.label+"</a><br>\n"
+          end
+        }.join
+      else # just a part
+        unicode_body.do{|b|
+          if content_type && content_type.match(/html/)
+           (b.split /<body[^>]*>/)[-1].split(/<\/body>/)[0]
+          else
+            b.hrefs true
+          end}
+      end
+    end 
+  end
+end
+
+class String; def is_binary_data?; true; end; end

data/infod/{Es/man.rb → man.rb}

@@ -54,18 +54,6 @@ class E
           preconv = %w{hu pt tr}.member?(superLang) ? "" : "-k"
           pageCmd = "zcat #{man} | groff #{preconv} -T html -man -P -D -P #{imagePath}"
           page = `#{pageCmd}`#.to_utf8
-
-          [[:name,name],
-           [:acceptLang,acceptLang],
-           [:lang, lang],
-           [:langSH, langSH],
-           [:superLang, superLang],
-           [:roff,man],
-           [:htmlBase,htmlBase.d],
-           [:imagePath,imagePath],
-           [:localizations,localesAvail],
-           [:pageCmd,pageCmd]].map{|p| puts [" "*(13-p[0].size),*p].join ' ' }
-          
           page = Nokogiri::HTML.parse page
           body = page.css('body')[0]
           
@@ -88,9 +76,9 @@ class E
           #  HTMLize hyperlinks
           #  markup commands
           body.xpath('//text()').map{|a|
-            a.replace a.to_s.gsub('&gt;','>').hrefs.gsub /\b([^<>\s(]+)\(/mi, '<b>\1</b>('
-          }
-          
+            a.replace a.to_s.gsub('&gt;','>').hrefs.gsub /\b([^<>\s(]+)\(/mi, '<b>\1</b>('}
+          body.css('font').map{|f|f.remove_attribute 'color'}
+
           qs = r['QUERY_STRING'].do{|q| q.empty? ? '' : '?' + q}
           # href-ize commands
           body.css('b').map{|b|

data/infod/{H/microblog.rb → microblog.rb}

@@ -19,14 +19,14 @@ class E
         yield s, Content,             m[3].hrefs(true)
         yield s, Type,                E[SIOCt+'InstantMessage']
         yield s, Type,                E[SIOC+'Post']
-        yield s, 'hasLink',           (m[3].match(/http:\//) ? 'true' : 'false')
-        yield s, 'hasNum', 'true' if m[3].match(/\d/)} rescue (puts "skipped #{l}")
+        yield s, SIOC+'link',        (m[3].match(/http:\//) ? 'true' : 'false')
+      } rescue nil
     }
   end
 
-  def tw g='m'
+  def tw g
     no.readlines.shuffle.each_slice(22){|s|
-      E['https://twitter.com/search/realtime?q='+s.map{|u|'from:'+u.chomp}.intersperse('+OR+').join].addJSON :triplrTweets, g}
+      E['https://twitter.com/search/realtime?q='+s.map{|u|'from:'+u.chomp}.intersperse('+OR+').join].insertDocs :triplrTweets, g}
   end
 
   def triplrTweets
@@ -47,38 +47,29 @@ class E
     }
   end
 
-  fn 'head/chat',->d,e{
-     t = d.map{|_,r|r[Date]}.flatten.compact.map &:to_time
-     c = d.map{|_,r|r[Content]}.compact.size
-    [{_: :title, c: "#{c} post#{c!=1 && 's'} #{t.min} -> #{t.max}"},
-     (Fn 'head.formats',e)]}
-
   fn 'view/chat',->d,e{
-    Fn'view/chat/base',d,e,->{d.map{|u,r|Fn 'view/chat/item',r,e}}}
+    Fn'baseview/chat',d,e,->{d.map{|u,r|Fn 'itemview/chat',r,e}}}
 
-  fn 'view/chat/item',->r,e{
-    line = r.E.frag
-    r[Type] && r[Type].map(&:uri).include?(SIOCt+'MailMessage') && r[:mail]=true
-    r[Content] && 
-    [{_: :a, id: line},
-     {_: :a, :class => :date, href: r.url, c: r[Date][0].match(/T([0-9:]{5})/)[1]},
+  fn 'itemview/chat',->r,e{
+    r[Type] && [*r[Type]].map{|t|t.respond_to?(:uri) && t.uri}.include?(SIOCt+'MailMessage') && r[:mail]=true
+    r[Content] && r[Date] && r[Date][0] &&
+    [r[Date][0].match(/T([0-9:]{5})/).do{|m|m[1]},
      {_: :span, :class => :nick, c: {_: :a, href: r[Atom+'/link/alternate'].do{|a|a[0].uri}||r.url,
-            c: [r[Atom+"/link/image"].do{|p| {_: :img, class: :a, src: p[0].uri}},
+            c: [r[Atom+"/link/image"].do{|p| {_: :img, src: p[0].uri, style: "#{rand(2).zero? ? 'left' : 'right'}: 0"}},
                 {_: :span, c: r[SIOC+'name']||r[Creator]||'#'}]}},' ',
-        {_: :span, :class => :tw, 
-       c: [r[Atom+'/link/media'].do{|a|
-             a.compact.map{|a|{_: :a, href: r.url, c: {_: :img, src: a.uri}}}},
-           ((r[Title].to_s==r[Content].to_s || r.uri.match(/twitter/)) && '' ||
-            {_: :a, href: r.url, c: r[Title],:class => r[:mail] ? :titleMail : :title}),
+        {_: :span, :class => :tw, # skip redundant title fields
+       c: [((r[Title].to_s == r[Content].to_s || r.uri.match(/twitter/)) && '' ||
+            {_: :a, :class => :title, href: r.url, c: r[Title]}), # skip quoted mail-lines & abbreviate
            r[:mail] ? (r[Content].map{|c|c.lines.to_a.grep(/^[^&@_]+$/)[0..21]}) : r[Content],
-          ]},' ',
-     {_: :a, class: :line, href: '#'+line, c: ' '},
-     "<br>\n"] if r.uri}
-  
-  fn 'view/chat/base',->d,e,c{
-    [(H.once e,'chat.head',(H.css '/css/tw'),{_: :style, c: "body, span.nick span, a {background-color: #{E.c}}\n"}),
-     {:class => :ch, c: c.()},
-     (H.once e,'chat.tail',{id: :b})]}
+          ]},"<br>\n"]}
+
+  F['view/'+SIOCt+'BoardPost']=->d,e{
+    d.map{|u,r|
+      {class: :BoardPost, style: "background-color:#ff4500;color:#fff;float:left;border-radius:.8em;padding:.4em;max-width:42em;margin:.5em",
+        c: F['itemview/chat'][r,e]}}}
+
+  fn 'baseview/chat',->d,e,c{
+    [(H.once e,'chat.head',(H.css '/css/tw'),{_: :style, c: "body {background-color: #{E.c}}\n"}),c.()]}
 
   F['view/'+SIOCt+'InstantMessage']=F['view/chat']
   F['view/'+SIOCt+'MicroblogPost']=F['view/chat']

data/infod/{K.rb → mime.rb}

@@ -1,37 +1,49 @@
 #watch __FILE__
 class E
 
-  FSbase = `pwd`.chomp ;  BaseLen = FSbase.size
-  Prefix = '/@'  # non-HTTP URI path resolution-prefix
-  S      = /\._/ # data path-separator
+  # no link-follow
+  def mime
+    @mime ||=
+      (t = ext.downcase.to_sym
 
-  W3    = 'http://www.w3.org/'
-  Purl  = 'http://purl.org/'
-  FOAF  = "http://xmlns.com/foaf/0.1/"
-  SIOC  = 'http://rdfs.org/sioc/ns#'
-  SIOCt = 'http://rdfs.org/sioc/types#'
-  MIMEtype = 'http://www.iana.org/assignments/media-types/'
-  DC       = Purl + 'dc/terms/'
-  Date     = DC   + 'date'
-  Modified = DC   + 'modified'
-  Title    = DC   + 'title'
-  Name     = FOAF + 'name'
-  To       = SIOC + 'addressed_to'
-  Creator  = SIOC + 'has_creator'
-  Content  = SIOC + 'content'
-  Type     = W3   + "1999/02/22-rdf-syntax-ns#type"
-  RDFs     = W3   + '2000/01/rdf-schema#'
-  HTTP     = W3   + '2011/http#'
-  Header   = W3   + '2011/http-headers#'
-  Posix    = W3   + 'ns/posix/'
-  Stat     = Posix + 'stat#'
-  Label    = RDFs + 'label'
-  EXIF     = 'http://www.w3.org/2003/12/exif/ns#'
-  Audio    = 'http://www.semanticdesktop.org/ontologies/nid3/#'
-  Edit     = 'http://buzzword.org.uk/rdf/personal-link-types#edit'
-  Render = 'render/'
-  Apache = ENV['apache']
-  Nginx  = ENV['nginx']
+       if node.symlink?
+         "inode/symlink"
+       elsif d?
+         "inode/directory"
+       elsif MIME[t]
+         MIME[t]
+       elsif Rack::Mime::MIME_TYPES[t='.'+t.to_s]
+         Rack::Mime::MIME_TYPES[t]
+       elsif base.index('msg.')==0
+         "message/rfc822"
+       elsif e
+         `file --mime-type -b #{sh}`.chomp
+       else
+         "application/octet-stream"
+       end)
+  end
+
+  # recursively-dereferenced links
+  def mimeP
+    @mime ||=
+      (p = realpath
+       unless p
+         nil
+       else
+         t = ((File.extname p).tail || '').downcase.to_sym
+         if p.directory?
+           "inode/directory"
+         elsif MIME[t]
+           MIME[t]
+         elsif Rack::Mime::MIME_TYPES[t='.'+t.to_s]
+           Rack::Mime::MIME_TYPES[t]
+         elsif (File.basename p).index('msg.')==0
+           "message/rfc822"
+         else
+           `file --mime-type -b #{Shellwords.escape p.to_s}`.chomp
+         end
+       end )
+  end
 
   MIME={
     aif: 'audio/aif',
@@ -40,6 +52,7 @@ class E
     avi: 'video/avi',
     e: 'application/json+rdf',
     coffee: 'text/plain',
+    conf: 'text/plain',
     css: 'text/css',
     csv: 'text/comma-separated-values',
     doc: 'application/word',
@@ -70,12 +83,14 @@ class E
     pdf: 'application/pdf',
     pl:  'application/perl',
     png: 'image/png',
+    ps:  'application/postscript',
     py:  'application/python',
     rb:  'application/ruby',
     ru:  'application/ruby',
     rdf: 'application/rdf+xml',
     rtf: 'text/rtf',
     ssv: 'text/semicolon-separated-values',
+    tex: 'text/x-tex',
     textile: 'application/textile',
     tsv: 'text/tab-separated-values',
     ttl: 'text/turtle',
@@ -87,9 +102,6 @@ class E
     xlsx: 'application/excel',
   }
 
-  VideoFile = /(avi|flv|mkv|mpg|mp4|wmv)$/i
-  AudioFile = /(aif|wav|flac|mp3|m4a|aac|ogg)$/i
-
   MIMEsource={
     'application/atom+xml' => [:triplrFeed],
     'application/markdown' => [:triplrMarkdown],
@@ -97,18 +109,21 @@ class E
     'application/rdf+xml'  => [:triplrRDF,:rdfxml],
     'application/json'     => [:triplrJSON],
     'application/pdf'      => [:triplrPDF],
+    'application/postscript'=> [:triplrPS],
     'application/textile'  => [:triplrTextile],
     'application/uri'      => [:triplrUriList],
     'application/word'     => [:triplrWord],
     'audio/mp4'            => [:triplrStdOut,'faad -i',Audio],
     'audio/mpeg'           => [:triplrStdOut,'id3info',Audio,/\((.*?)\)$/],
     'audio'                => [:triplrStdOut,'sndfile-info',Audio],
+    'image'                => [:triplrImage],
     'inode/symlink'        => [:triplrSymlink],
-    'message/rfc822'       => [:triplrMail],
+    'message/rfc822'       => [:triplrMailMessage],
     'text/ansi'            => [:triplrANSI],
     'text/comma-separated-values'=>[:triplrCSV,/,/],
     'text/log'             => [:triplrIRC],
     'text/man'             => [:triplrMan],
+    'text/n3'              => [:triplrRDF, :n3],
     'text/nfo'             => [:triplrHref,'cp437'],
     'text/ntriples'        => [:triplrRDF, :ntriples],
     'text/plain'           => [:triplrHref],
@@ -116,6 +131,7 @@ class E
     'text/semicolon-separated-values'=>[:triplrCSV,/;/],
     'text/tab-separated-values'=>[:triplrCSV,/\t/],
     'text/turtle'          => [:triplrRDF,:turtle],
+    'text/x-tex'           => [:triplrTeX],
   }
 
   # prefer a view even if requested file exists
@@ -124,6 +140,7 @@ class E
     'application/markdown' => true,
     'application/json+rdf' => true,
     'application/org' => true,
+    'application/postscript' => true,
     'application/textile' => true,
     'application/uri' => true,
     'application/word' => true,
@@ -134,6 +151,14 @@ class E
     'text/man'=>true,
     'text/nfo'=>true,
     'text/rtf'=>true,
+    'text/x-tex'=>true,
+  }
+
+  # cache triplr output
+  MIMEcache={
+    'audio' => true,
+    'image' => true,
+#    '' => 
   }
 
   %w{c c++ fortran haskell makefile pascal perl php python ruby}.map{|t|
@@ -141,25 +166,16 @@ class E
       MIMEcook[m+t] = true
   }}
 
-  Abbrev={
-    "dc" => DC,
-    "foaf" => FOAF,
-    "rdf" => W3+"1999/02/22-rdf-syntax-ns#",
-    "rdfs" => RDFs,
-    "sioc" => SIOC,
-    "stat" => Stat,
-  }
-  
-  # literal to pathname types
-  Literal={}
-   [Purl+'dc/elements/1.1/date',
-    Date,
-    DC+'created',
-    Modified,
-   ].map{|f|Literal[f]=true}
+  def render mime,   graph, e
+   E[Render+ mime].y graph, e
+  end
 
-  def == u
-      to_s == u.to_s
+  def triplrMIME &b
+    mimeP.do{|mime|
+      yield uri, E::Type, (E MIMEtype+mimeP)
+      (MIMEsource[mimeP]||
+       MIMEsource[mimeP.split(/\//)[0]]).do{|s|
+        send *s,&b }}
   end
 
 end

data/infod/{N.rb → names.rb}

@@ -1,25 +1,17 @@
 %w{base64 cgi shellwords}.each{|r|require(r)}
 
-def E e
-  E.new e
-end
-
 class E
 
-  def E.[] u; u.E end
+  attr_reader :uri
+  alias_method :url, :uri
 
-  def E arg=nil
-    if arg
-      E.new arg
-    else
-      self
-    end
+  def env r=nil
+    r ? (@r = r
+         self) : @r
   end
-  
-  attr_reader :uri
 
-  def initialize uri
-    @uri = uri.to_s
+  def == u
+      to_s == u.to_s
   end
 
   def basename
@@ -39,11 +31,10 @@ class E
   def ttl; @ttl||= docBase.a('.ttl') end
 
   def docBase
-    uri.split(/#/)[0].E.do{|d|
-      d.dirname.as d.bare }
+    !uri.empty? && uri.split(/#/)[0].do{|u|u.E.do{|d| d.dirname.as d.bare }} || E['']
   end
   
-  # same as above, but w/ URI errors
+  # and w/ URI-lib
   def docBaseURI
     u = URI uri
     s = u.scheme
@@ -79,22 +70,27 @@ class E
   end
   alias_method :dir, :dirname
 
-  # local URL from unlocatable identifier (mail MSGID, etc)
-  def url
-    path? ? uri : Prefix + (CGI.escape uri)
+  # add hostname to URI if missing
+  def hostURL e
+    host = 'http://'+e['SERVER_NAME']
+    if uri.index('/') == 0 
+      host + uri
+    else
+      uri
+    end
   end
 
-  # local URL even if locatable-identifier
+  # locator for local data about global URI
   def localURL e
     # path
-    if uri.index('/') == 0
+    if uri.index('/') == 0 # already a local path
       uri
     # host match
-    elsif uri.index('http://'+e['SERVER_NAME']+'/') == 0
+    elsif e && uri.index('http://'+e['SERVER_NAME']+'/') == 0 
       pathSegment.uri
     # non-local
     else
-      Prefix + (CGI.escape uri)
+      URIURL + (CGI.escape uri)
     end
   end
 
@@ -116,6 +112,10 @@ class E
     uri.expand.E
   end
 
+  def shorten
+    uri.shorten.E
+  end
+
   def prependURI u
     E u.to_s + uri
   end
@@ -129,7 +129,7 @@ class E
   end
 
   def concatURI b
-    u.appendURI b.E.path
+    u.appendURI b.E.shortPath
   end
 
   alias_method :a, :appendURI
@@ -140,6 +140,19 @@ class E
     uri.path?
   end
 
+  def shortPath
+    @shortPath ||=
+      (if path?
+         if uri.match /^\//
+           uri
+         else
+           '/' + uri.shorten
+         end
+       else
+         '/E/' + uri.h.dive[0..5] + (Base64.urlsafe_encode64 uri)
+       end)
+  end
+
   # URI -> path
   def path
     @path ||=
@@ -170,15 +183,21 @@ class E
   end
 
   # literals to URIs
-
+  # currently used for iso8601 dates mapping to paths, so date-range queries can be done w/ just dir/fs tooling
+  # could also use as a "trie" for autocomplete or sorting strings
   def E.literal o
     E['/'].literal o
   end
+  
+  Literal={}
+   [Purl+'dc/elements/1.1/date',
+    Date,DC+'created',DC+'modified',
+   ].map{|f|Literal[f]=true}
 
   def literal o
 
     # already a URI
-    return self if o.class == E
+    return o if o.class == E
 
     # blob for non-strings
     return literalBlob o unless o.class == String
@@ -196,7 +215,7 @@ class E
 
   # pathname for short literals
   def literalURI o
-    E "/l/"+(Literal[uri] && o.gsub(/[\.:\-T+]/,'/'))+'/'+o
+    E "/l/"+o.gsub(/[\.:\-T+]/,'/')+'/'+o if Literal[uri] && o
   end
  
   def literalBlobURI o
@@ -225,13 +244,16 @@ class E
    {'uri' => uri}
   end
 
+  # implementation-specific internal pathnames not on the web
+  F['/E/GET'] = F[E404]
+
 end
 
 class Hash
   def uri
-    self["uri"]
+    self["uri"]||""
   end
-  def url; self.E.url end
+  alias_method :url, :uri
   def label
     self[E::Label] || uri.label
   end
@@ -253,18 +275,25 @@ class String
     self[4..-1]
   end
 
-  # expand qname-style identifier to URI
+  # expand qname/CURIE-style identifier to URI
   Expand={}
   def expand
-    # memoize lookups
    (Expand.has_key? self) ?
     Expand[self] :
    (Expand[self] =
      match(/([^:]+):([^\/].*)/).do{|e|
-       (E::Abbrev[e[1]]||e[1]+':')+e[2]} || 
+      ( E::Prefix[e[1]] || e[1]+':' )+e[2]} || 
      self )
   end
 
+  # shrink URI to qname/CURIE/prefix'd identifier
+  def shorten
+    E::Prefix.map{|p,f|
+      return p + ':' + self[f.size..-1]  if (index f) == 0
+    }
+    self
+  end
+
   def sh
     Shellwords.escape self
   end
@@ -277,29 +306,33 @@ class String
   # path -> URI || literal
   def unpath
 
-    # URI with scheme
+    # HTTP URI
     if m = (match /^\/([a-z]+:)\/+(.*)/)
       (m[1] + '//' + m[2]).E
 
-    # String literal in store
+    # CURIE
+    elsif m = (match /^\/([^\/:]+:[^\/]+)/)
+      m[1].expand.E
+
+    # opaque URI w/ optional extension
+    elsif match /^\/E\/..\//
+      self[9..-1].match(/([^.]+)(.*)/).do{|c|
+        (Base64.urlsafe_decode64 c[1]) + c[2]
+    }.E
+
+    # String literal
     elsif match /^\/E\/blob/
       self.E.r
 
-    # JSON literal in store
+    # JSON literal
     elsif match /^\/E\/json/
       self.E.r true
 
-    # String literal in basename
+    # literal in basename
     elsif match /^\/l\//
       File.basename self
 
-    # URI (opaque)
-    elsif match /^\/E\/..\//
-      self[9..-1].match(/([^.]+)(.*)/).do{|c|
-        (Base64.urlsafe_decode64 c[1]) + c[2]
-      }.E
-
-    # path
+    # plain path
     else
       self.E
     end