X-Git-Url: https://pere.pagekite.me/gitweb/text-madewithcc.git/blobdiff_plain/5ae54b9934be83eacfb5b1a66908b452e870c567..96afcb1558dc71921389416cf0e0ba4aec0e8ae3:/fixup.rb

diff --git a/fixup.rb b/fixup.rb
index 595a8c8..d15b243 100644
--- a/fixup.rb
+++ b/fixup.rb
@@ -1,38 +1,52 @@
 #!/usr/bin/ruby
 # coding: utf-8
+LogLevel=1
 raise ArgumentError, 'Source/destination files not specified' if ARGV.size != 2
 srcfile = ARGV[0]
 dstfile = ARGV[1]
 
 data=File.open(srcfile).readlines.map {|l| l.chomp!}
 
-data.map {|lin| lin.gsub!(/<span id="anchor-?\d*"><\/span>/, '')}
+data.map {|lin| lin.gsub!(/\[\]{#anchor-?\d*}/, '')}
+
+def log(level,what)
+  indicators = %w(! • # -)
+  if level >= LogLevel
+    print indicators[level]
+  else
+    print "\n%s %s\n" % ['*' * (level+1), what]
+  end
+end
 
 # There are several titles that are spread in more than one line. Make
 # them into translatable sentences.
+log 0, 'Merging multiline sentences '
 [ ['Made', '', 'with', '', 'Creative', '', 'Commons'],
   ['The New','', 'World of', '', 'Digital', '', 'Commons'],
   ['How', '', 'to Be', '', 'Made with', '', 'Creative', '', 'Commons'],
-  ['Use CC to get attribution and name', 'recognition'],
-  ['Use CC to enable hands-on engagement with', 'your work'],
-  ['Providing a custom service to consumers of', 'your work * \[MARKET-BASED\]*'],
-  ['Charging for the physical copy *', '\[MARKET-BASED\]*'],
-  ['Charging for the in-person version *','\[MARKET-BASED\]*'],
-  ['Charging advertisers or sponsors *', '\[MARKET-BASED\]*'],
-  ['Charging your content creators *', '\[MARKET-BASED\]*'],
-  ['Charging a transaction fee *', '\[MARKET-BASED\]*'],
-  ['Providing a service to your creators*', '\[MARKET-BASED\]*'],
+  ['Providing a custom service to consumers of your work *', '\[MARKET-BASED\]*'],
   ['Memberships and individual donations', '*\[RECIPROCITY-BASED\]*'],
-  ['The pay-what-you-want model', '*\[RECIPROCITY-BASED\]*'],
   ['The', '', 'Creative', '', 'Commons', '', 'Licenses'],
+  ['jonathanmann.net and', '', 'jonathanmann.bandcamp.com'],
   ['PLOS','', '(Public Library of Science)']
 ].each do |str|
   # This should be done more generic, more robust... But before
   # burning brain cells, verify if it's needed!
-  if (data.select {|i| i == str[0]}.size != 1)
-    raise RuntimeError, 'First string appears multiple times: %s (%s)' %
-                        [str.reject{|i| i==''}.join(' '), str[0]]
+  log 1, str
+  matches=[]
+  data.each_with_index do |lin, idx|
+    if lin == str[0]
+      matches << idx
+    end
   end
+  if matches.size == 0
+    raise RuntimeError, 'Anchor string («%s» for «%s») not found' %
+                        [str[0], str.reject {|word| word==''}.join(' ')]
+  elsif matches.size > 1
+    raise RuntimeError, 'Anchor string (%s) appears multiple times: %s' %
+                        [str[0], matches.map {|i| i.to_s}.join(', ')]
+  end
+  log 2, 'Matches %s: %d - %s' % [str.join(' '), matches.size, matches.join(',')]
 
   len = str.size
   at = data.index(str[0])
@@ -62,6 +76,7 @@ end
 #
 # Try to keep this ordered as it appears within the book, as it will
 # help us spot omissions and mistakes!
+log 0, 'Mark up headings'
 [ [2, 'Foreword'],
   [2, 'Introduction'],
   [1, 'Part 1'],
@@ -111,7 +126,7 @@ end
   [4, 'Build a community'],
   [4, 'Give more to the commons than you take'],
   [4, 'Involve people in what you do'],
-  [4, 'Notes'],
+  [3, 'Notes'],
   [2, 'The Creative Commons Licenses'],
   [1, 'Part 2'],
   [1, 'The Case Studies'],
@@ -143,6 +158,7 @@ end
   [2, 'Acknowledgments'],
   
 ].each do |item|
+  log 1, item.join(' -> ')
   at = data.index {|i| i == item[1]}
   if at.nil?
     raise RuntimeError, 'Heading string (level %d) not found: «%s»' % item
@@ -150,39 +166,54 @@ end
   data[at] = '%s %s' % ['#' * item[0], data[at]]
 end
 
+# We have the explicit strings "Part 1" and "Part 2" as structural
+# elements — They are to be generated upon book compilation. Nuke
+# them.
+data.delete("# Part 1")
+data.delete("# Part 2")
 
-
-# Join erroneously split paragraphs
-['content and, in turn, spend money and',
- 'still other',
- 'to the values symbolized by',
- 'the kinds of participative communities that drive open',
- 'At a minimum, a CC-',
- 'easier to trust a',
- 'free download, the',
- 'openness to fans remixing the game—give',
- 'Attribution-',
- 'both journal publishers and researchers. Figshare now provides',
- 'get the “network effect”—',
- 'access to scholarly books. For Frances, the current scholarly-',
- 'for-',
- 'sales',
- 'contributing to the open',
- 'doesn’t seem like it should be sung about',
- 'songwriter, and he has found a way to keep it interesting for',
- 'building trust is the top',
- 'license',
- 'authors and Shuttleworth; Mark remains incredibly proud of this',
- 'BY-SA and opting in others with collecting societies like',
- 'Cecilie Maria, Cedric Howe, Cefn Hoile,',
- 'Braddlee, Drew Spencer, Duncan',
- 'Elizabeth Holloway, Ellen Buecher, Ellen Kaye-',
- 'Helen',
+# Join erroneously split paragraphs: Write the contents of the line
+# _preceding_ the unneeded break, the break will be removed.
+#
+# I'm noting the line number for each _after_ corrections so it's
+# easier to find them; please keep them sorted! :-P
+log 0, 'Join erroneously split paragraphs'
+['content and, in turn, spend money and', # 1595
+ 'still other', # 1662
+ 'content functions as a marketing tool for the paid product or', # 1724
+ 'lowest-common-denominator solutions and', #2035
+ 'to the values symbolized by', # 2145
+ 'the kinds of participative communities that drive open', # 2157
+ 'time', # 2220
+ 'At a minimum, a CC-', # 2375
+ 'easier to trust a', # 2580
+ 'free download, the', # 3086
+ 'openness to fans remixing the game—give', # 3087
+ 'Attribution-', # 3307
+ 'both journal publishers and researchers. Figshare now provides', # 3672
+ 'get the “network effect”—', # 4002
+ 'access to scholarly books. For Frances, the current scholarly-', # 4033
+ 'for-', # 4288
+ 'sales', # 4410
+ 'contributing to the open', # 4438
+ 'doesn’t seem like it should be sung about', # 4616
+ 'songwriter, and he has found a way to keep it interesting for', # 4624
+ 'building trust is the top', # 4793
+ 'version', # 6023
+ 'license', # 6169
+ 'authors and Shuttleworth; Mark remains incredibly proud of this', # 6452
+ 'BY-SA and opting in others with collecting societies like', # 7218
+ 'Cecilie Maria, Cedric Howe, Cefn Hoile,', # 7796
+ 'Braddlee, Drew Spencer, Duncan', # 7839
+ 'Elizabeth Holloway, Ellen Buecher, Ellen Kaye-', # 7844
+ 'Helen', # 7874
 ].each do |line|
+  log 1, line
   at = data.index {|i| i == line}
   if !at.nil? and data[at+1] == ''
     data.delete_at(at+1)
   end
 end
 
+log 0, 'Writing processed file'
 File.open(dstfile, 'w') {|f| f.puts data.join("\n")}