iankelling.org Git - iankelling.org/blob - b.rb

   1 # encoding: utf-8
   2 # Copyright (C) 2016 Ian Kelling
   3
   4 # This program is free software: you can redistribute it and/or modify
   5 # it under the terms of the GNU General Public License as published by
   6 # the Free Software Foundation, either version 2 of the License, or
   7 # (at your option) any later version.
   8
   9 # This program is distributed in the hope that it will be useful,
  10 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 # GNU General Public License for more details.
  13
  14 # You should have received a copy of the GNU General Public License
  15 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
  16 module B # blog module
  17   require 'fileutils'
  18   require 'time'
  19   require 'safe_yaml'
  20   require 'pygments'
  21   require 'sqlite3'
  22   require 'redcarpet'
  23   JS_INFO = "<p>All JavaScript has <a href=\"https://www.gnu.org/software/librejs/index.html\">LibreJS</a> support.</p>"
  24
  25   DAY = 60*60*24
  26   DN = 'iankelling'
  27   FQDN = DN + '.org'
  28   DURL = 'https://' + FQDN
  29   DESCRIPTION = "Ian Kelling's personal site and blog on software"
  30   DATE_LEN = 'YYYY-MM-DD'.length
  31   NOW = Time.now.to_f
  32   WAIT_DATE = NOW - 60*60*24*1
  33
  34   def db_init
  35     SQLite3::Database.new('../proposed-comments/comments.sqlite')
  36   end
  37
  38   # from the redcarpet readme, then a bunch of googling to figure
  39   # out what to do on exception.
  40   class HTMLwithPygments < Redcarpet::Render::HTML
  41     def block_code(code, language)
  42       begin
  43         Pygments.highlight(code, lexer: language)
  44       rescue MentosError
  45         # when language detection fails
  46         Pygments.highlight(code, lexer: 'text')
  47       end
  48     end
  49   end
  50
  51   def fwrite(output_path, string)
  52     FileUtils.mkdir_p(File.dirname(output_path))
  53     File.write(output_path, string)
  54   end
  55
  56   def fskel(rel_path, title, content, o={})
  57     head = <<EOF
  58 <link rel="canonical" href="#{DURL}/#{rel_path}">
  59 EOF
  60     if rel_path =~ %r{^blog/.}
  61       head += <<EOF
  62 <link rel="alternate" type="application/atom+xml" title="#{DN}" href="#{DURL}/feed.xml">
  63 EOF
  64     end
  65     o[:head] = head
  66     fwrite(rel_path, skel(title, content, o))
  67   end
  68   def skel(title, content, o={})
  69     # got meta viewport from jekyll's default later. It's for better
  70     # mobile viewing.
  71     output = <<EOF
  72 <!DOCTYPE html>
  73 <html lang="en-US">
  74   <head>
  75     <meta charset="utf-8">
  76     <title>#{title}</title>
  77     <link rel="stylesheet" href="/main.css">
  78     <link rel="stylesheet" href="/common.css">
  79     <link rel="shortcut icon" href="/assets/favicon.png" />
  80     <meta name="description" content="#{o[:description] || DESCRIPTION}">
  81 #{o[:head]}
  82     <meta name="viewport" content="width=device-width, initial-scale=1">
  83   </head>
  84   <body>
  85     <header>
  86       <h3><a href="/">iankelling.org</a>#{o[:header]}</h3>
  87     </header>
  88     <div class="main-content-stripe">
  89       <div class="content">
  90         #{content}
  91       </div>
  92     </div>
  93     <div id="comment-stripe">
  94 #{o[:comments]}
  95     </div>
  96     <footer>
  97 #{o[:footer]}
  98       <p>This site has a <a href="/git/?p=iankelling.org;a=summary">git repo</a>. Code is gpl, content is <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/"><img id="cc-by-sa" alt="Creative Commons License" src="/assets/cc-by-sa-4.0-80x15.png" /></a> unless otherwise noted</p>
  99       <p><address><a href="ian@iankelling.org">ian@iankelling.org</a> let me know what you think</address></p>
 100     </footer>
 101   </body>
 102 </html>
 103 EOF
 104     output
 105   end
 106
 107   def stdpage(page_name, content)
 108     fskel("#{page_name}.html",
 109           "#{DN}/#{page_name}",
 110           content,
 111           header: " / <a href=\"/#{page_name}.html\">#{page_name}</a>")
 112   end
 113
 114   def md_to_html(md)
 115     # Using redcarpet over kramdown because syntax highlighting is
 116     # simpler. kramdown uses some crap highlighter by default,
 117     # supports using rouge, but then the classes are all screwy
 118     # for what pygments css expects, rouge has a pygments compatibility mode,
 119     # but that is a pita to get working, then it doesn't even work right.
 120     # kramdown is jekyll's default markdown parser, but it doesn't use
 121     # it for code blocks, it strips them out using custom templating
 122     # extension class, then uses rouge, then wraps it in some
 123     # custom html for pygments compatibility. It's a complicated mess.
 124     Redcarpet::Markdown.new(HTMLwithPygments, fenced_code_blocks: true).render(md)
 125   end
 126
 127   def comment_html(comment, date)
 128     inner = Redcarpet::Markdown.new(Redcarpet::Render::HTML, fenced_code_blocks: true).render(<<EOF)
 129 #{comment}
 130 <span class="comment-date">#{Time.at(date).strftime("%b %-d '%y")}</span>
 131 EOF
 132     # I tried putting the time,  %I:%M %p UTC, but it looks kinda
 133     # clunky, going against my simple theme.
 134     <<EOF
 135 <div class="comment">
 136   #{inner}
 137 </div>
 138 EOF
 139   end
 140
 141   def post(file, build_time=false)
 142     content = File.read(file)
 143     content =~ %r{\A(---\s*\n.*?\n?)^((---)\s*$\n?)}m # yaml front matter
 144     # stuff after last match. jekyll uses $POSTMATCH,
 145     # but it's nil for me, I don't know what magic they are using.,
 146     # but only $' is listed here http://ruby-doc.org/core-2.3.1/doc/globals_rdoc.html,
 147     content = $'
 148
 149     front = SafeYAML.load(Regexp.last_match(1))
 150     title = front['title']
 151     $page_title = "#{title} | #{DN}"
 152     header_rel = ' / <a href="/blog.html">blog</a> /'
 153
 154     footer_extra = <<-EOF
 155 <p><a class="icon-rss" href="/feed.xml">Subscribe</a></p>
 156 EOF
 157     footer_extra += JS_INFO if content =~ /<script/
 158
 159
 160
 161     b = File.basename(file,'.md')
 162     # date is in the format: YYYY-MM-DD-
 163     date = Time.parse(b[0..DATE_LEN])
 164     rel_path = "blog/#{b[(DATE_LEN + 1)..-1]}.html"
 165     comments = $db.execute <<-SQL, [WAIT_DATE]
 166       select comment, date from c
 167       where page = '#{rel_path}' and (
 168        state = 'picked' or state = 'known'
 169        or (state = 'waiting' and date < ?))
 170 SQL
 171     # get earliest comment. earlier ones stored in git will also be
 172     # published. This get's us easily sharable comments, and allows us
 173     # to expire unpublished comments and ip addresses which are PII and
 174     # should never be kept around indefinitely.
 175     sql_start_date = $db.execute('select min(date) from c')[0][0] || NOW
 176     comment_file_dir = "../comments/#{rel_path}"
 177     old_comments = Dir["#{comment_file_dir}/*"].reduce([]) do |memo, f|
 178       dt = File.basename(f).to_f
 179       if dt < sql_start_date
 180         memo << [File.read(f), dt]
 181       else
 182         FileUtils.rm(f) if build_time
 183         memo
 184       end
 185     end
 186     if build_time
 187       FileUtils.mkdir_p comment_file_dir
 188       comments.each do |c, c_date|
 189         # fyi: there is an extremely small chance of 2 comments having
 190         # the same floating point time and thus overwriting each other.
 191         # Small enough that it won't happen at my site's scale.
 192         File.write(File.join(comment_file_dir, c_date.to_s), c)
 193       end
 194       # https://piwik.org/docs/privacy/ says keep logs for 3-6 months
 195       $db.execute("delete from c where date < #{NOW - DAY*180}")
 196     end
 197     comments = old_comments + comments
 198     pending_comments = $db.execute(<<-SQL, [WAIT_DATE])[0][0]
 199       select count(*) from c
 200       where page = '#{rel_path}' and
 201        (state = 'waiting' and date > ? or state = 'suspect')
 202 SQL
 203
 204     feed_html = md_to_html(content)
 205     page_html = <<-EOF
 206       <header class="post-header">
 207         <h1 class="post-title">#{title}</h1>
 208         <p class="post-date">#{date.strftime("%b %-d, %Y")}</p>
 209       </header>
 210       #{feed_html}
 211 EOF
 212     com_list = ''
 213     comments.each { |c, date| com_list += comment_html(c, date) }
 214     if pending_comments > 0
 215       if pending_comments >= 2
 216         text = "are #{pending_comments} new comments"
 217       else
 218         text = 'is 1 new comment'
 219       end
 220       com_list +=
 221         comment_html("Note: there #{text} pending approval.", NOW)
 222     end
 223     com_section = <<-EOF
 224       <form class="comment" action="/comment.rb" method="post">
 225         <input class="misc" type="text" name="url">
 226         <input name="goto" type="hidden" value="#{rel_path}">
 227         <textarea rows="10" name="comment" placeholder="markdown" maxlength="1000"></textarea>
 228         <input type="submit" value="Leave a comment">
 229       </form>
 230       <div id="comments">
 231          #{com_list}
 232       </div>
 233 EOF
 234     links = front['comment_links']
 235     if links
 236       link_html = links.map { |name,url| "<a href=\"#{url}\">#{name}</a>" }
 237                     .join(', ')
 238       com_section += (<<EOF)
 239 <p>More comments at #{link_html}</p>
 240 EOF
 241     end
 242
 243     blog_toc_entry = "<li><a href=\"#{rel_path}\">#{title}</a></li>"
 244
 245     com_section = <<EOF
 246 <div id="comment-section">
 247   #{com_section}
 248 </div>
 249 EOF
 250
 251
 252     if front['description']
 253       description = front['description']
 254     else
 255       # the first 300 saves ~ 1 ms
 256       # regex for striping html from liquid template src
 257       description = feed_html[0..300].gsub(/<script.*?<\/script>/m, '').
 258                       gsub(/<!--.*?-->/m, '').gsub(/<style.*?<\/style>/m, '').
 259                       gsub(/<.*?>/m, '')
 260       if description.length > 160
 261         description = description[0..156] + '...'
 262       end
 263     end
 264
 265     head = <<EOF
 266 EOF
 267
 268     fskel(rel_path, title, page_html,
 269           header: header_rel,
 270           footer: footer_extra,
 271           comments: com_section,
 272           description: description)
 273     url="#{DURL}/#{rel_path}"
 274
 275
 276     # following from https://creativecommons.org/choose,
 277     # with the addition of "unless otherwise noted", for js licenses.
 278     feed_copyright = <<-EOF
 279 <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/"><img alt="Creative Commons License" style="border-width:0" src="https://i.creativecommons.org/l/by-sa/4.0/88x31.png" /></a><br /><span xmlns:dct="http://purl.org/dc/terms/" href="http://purl.org/dc/dcmitype/Text" property="dct:title" rel="dct:type">#{title}</span> by <a xmlns:cc="http://creativecommons.org/ns#" href="#{url}" property="cc:attributionName" rel="cc:attributionURL">Ian Kelling</a> unless otherwise noted is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/">Creative Commons Attribution-ShareAlike 4.0 International License</a>.
 280 EOF
 281
 282     feed_entry = <<EOF
 283 <entry>
 284   <title>#{title}</title>
 285   <link rel="alternate" href="#{url}"/>
 286   <id>#{url}</id>
 287   <updated>#{date.to_datetime.rfc3339}</updated>
 288   <content type="html" xml:lang="en-us" xml:base="#{DURL}/blog">
 289     <![CDATA[
 290     #{feed_html}
 291     ]]>
 292   </content>
 293   <rights>
 294     #{feed_copyright}
 295   </rights>
 296 </entry>
 297 EOF
 298     return [feed_entry, blog_toc_entry]
 299   end
 300 end