Outils de conversion Trac → Markdown
This commit is contained in:
parent
4436c13004
commit
105f9e9759
|
@ -8,44 +8,67 @@ Si on souhaite convertir des pages du format Trac vers le format Markdown un pet
|
|||
#!/usr/bin/env ruby
|
||||
## Based on: https://gist.github.com/619537
|
||||
|
||||
body = STDIN.read
|
||||
class TracToMarkdown
|
||||
|
||||
# convert inline code blocs
|
||||
body.gsub!(/\{\{\{([^\n]+?)\}\}\}/, '`\1`')
|
||||
def convert(input)
|
||||
output = input.clone
|
||||
|
||||
body.gsub!(/^\s*\*/, '*')
|
||||
# convert inline code blocs
|
||||
output.gsub!(/\{\{\{([^\n]+?)\}\}\}/, '`\1`')
|
||||
|
||||
# add a line break before lists
|
||||
body.gsub!(/(\S)\n\* /, '\1' "\n\n* ")
|
||||
# remove spaces before list items
|
||||
output.gsub!(/^\s*\*/, '*')
|
||||
|
||||
body.gsub!(/^[ ]*\{\{\{/, '{{{')
|
||||
body.gsub!(/(\S)\n\{\{\{/, '\1' "\n\n{{{")
|
||||
body.gsub!(/\{\{\{(.+?)\}\}\}/m) { |m|
|
||||
m.each_line.map { |x|
|
||||
x.gsub(/ *[\{\}]{3}/,"~~~")
|
||||
}.join
|
||||
}
|
||||
# add a line break before lists
|
||||
output.gsub!(/(^[^\* ].*)\n\* /, '\1'+"\n\n* ")
|
||||
|
||||
# convert headings syntax
|
||||
body.gsub!(/\=\=\=\=\=\s(.+?)\s\=\=\=\=\=/, '##### \1')
|
||||
body.gsub!(/\=\=\=\=\s(.+?)\s\=\=\=\=/, '#### \1')
|
||||
body.gsub!(/\=\=\=\s(.+?)\s\=\=\=/, '### \1')
|
||||
body.gsub!(/\=\=\s(.+?)\s\=\=/, '## \1')
|
||||
body.gsub!(/\=\s(.+?)\s\=/, '# \1')
|
||||
output.gsub!(/^[ ]*\{\{\{/, '{{{')
|
||||
output.gsub!(/(\S)\n\{\{\{/, '\1 '+"\n\n{{{")
|
||||
output.gsub!(/\{\{\{(.+?)\}\}\}/m) { |m|
|
||||
m.each_line.map { |x|
|
||||
x.gsub(/ *[\{\}]{3}/,"~~~")
|
||||
}.join
|
||||
}
|
||||
|
||||
# convert links
|
||||
body.gsub!(/\[(https?[^\s\[\]]+)\s([^\[\]]+)\]/, '[\2](\1)')
|
||||
body.gsub!(/([^(])(https?[^\s\[\]]+)/, '\1<\2>')
|
||||
body.gsub!(/\!(([A-Z][a-z0-9]+){2,})/, '\1')
|
||||
# convert headings syntax
|
||||
output.gsub!(/\=\=\=\=\=\s(.+?)\s\=\=\=\=\=/, '##### \1')
|
||||
output.gsub!(/\=\=\=\=\s(.+?)\s\=\=\=\=/, '#### \1')
|
||||
output.gsub!(/\=\=\=\s(.+?)\s\=\=\=/, '### \1')
|
||||
output.gsub!(/\=\=\s(.+?)\s\=\=/, '## \1')
|
||||
output.gsub!(/\=\s(.+?)\s\=/, '# \1')
|
||||
|
||||
body.gsub!(/'''(.+?)'''/, '*\1*')
|
||||
body.gsub!(/''(.+?)''/, '_\1_')
|
||||
body.gsub!(/^\s\d\./, '1.')
|
||||
# convert links
|
||||
output.gsub!(/\[(https?[^\s\[\]]+)\s([^\[\]]+)\]/, '[\2](\1)')
|
||||
output.gsub!(/([^(])(https?[^\s\[\]]+)/, '\1<\2>')
|
||||
output.gsub!(/\!(([A-Z][a-z0-9]+){2,})/, '\1')
|
||||
|
||||
# custom add-on for pedantic satisfaction
|
||||
body.gsub!(/ *(\.){3,}/, '…')
|
||||
output.gsub!(/'''(.+?)'''/, '*\1*')
|
||||
output.gsub!(/''(.+?)''/, '_\1_')
|
||||
output.gsub!(/^\s\d\./, '1.')
|
||||
|
||||
STDOUT.write(body)
|
||||
# custom add-on for pedantic satisfaction
|
||||
# output.gsub!(/ ?(\.){3,}/, '…')
|
||||
|
||||
output
|
||||
end
|
||||
|
||||
def remove_outline(input)
|
||||
input.gsub(/^\[\[PageOutline\]\]\s+/, "")
|
||||
end
|
||||
|
||||
def add_trac_banner(input)
|
||||
banner = "**Cette page a été importée automatiquement de notre ancien wiki mais n'a pas encore été révisée.**"
|
||||
|
||||
banner + "\n\n" + input
|
||||
end
|
||||
|
||||
end
|
||||
|
||||
converter = TracToMarkdown.new
|
||||
|
||||
output = converter.convert(STDIN.read)
|
||||
|
||||
STDOUT.write(output)
|
||||
~~~
|
||||
|
||||
Copiez simplement ce script là où vous le souhaitez, rendez-le exécutable puis :
|
||||
|
@ -55,3 +78,65 @@ $ cat Page.trac | trac2md
|
|||
~~~
|
||||
|
||||
*Note* : le script ne gère pas une conversion parfaite, il faudra relire le document et faire quelques corrections manuelles (titre en "front matter Yaml", etc.). Patches welcome :)
|
||||
|
||||
Voici le script qui a servi à convertir tout le contenu du Trac (nécessite la gem `Mechanize`) :
|
||||
|
||||
~~~{.ruby}
|
||||
require 'mechanize'
|
||||
require 'net/http'
|
||||
require_relative "./trac_to_markdown"
|
||||
|
||||
domain = "http://trac.evolix.net"
|
||||
index_url = "#{domain}/infogerance/wiki/TitleIndex"
|
||||
|
||||
base_pattern = %r(/infogerance/wiki/)
|
||||
links = Mechanize.new.get(index_url).links_with(href: base_pattern)
|
||||
|
||||
skip_pages = [
|
||||
%r(/infogerance/wiki/TracGuide),
|
||||
%r(/infogerance/wiki/TitleIndex),
|
||||
%r(/infogerance/wiki/CamelCase),
|
||||
%r(/infogerance/wiki/RecentChange),
|
||||
%r(/infogerance/wiki/TitleIndex.*),
|
||||
%r(/infogerance/wiki/Wiki.*),
|
||||
%r(/infogerance/wiki/InterWiki),
|
||||
%r(/infogerance/wiki/InterTrac),
|
||||
%r(/infogerance/wiki/InterMapTxt),
|
||||
%r(/infogerance/wiki/SandBox),
|
||||
%r(/infogerance/wiki/Trac[A-Z]\w*),
|
||||
]
|
||||
|
||||
skip_content = /ATTENTION, CETTE PAGE EST OBSOLÈTE/
|
||||
|
||||
save_path = Pathname.new(".")
|
||||
# save_path = Pathname.new("../wiki")
|
||||
|
||||
converter = TracToMarkdown.new
|
||||
|
||||
links.each do |link|
|
||||
page = link.href
|
||||
|
||||
if skip_pages.any? { |skip_page| skip_page.match(page) }
|
||||
puts " #{page} skipped"
|
||||
next
|
||||
end
|
||||
|
||||
file_path = save_path.join(page.gsub(base_pattern, "") + ".md")
|
||||
file_path.dirname.mkpath
|
||||
|
||||
uri = URI.parse(domain + page + "?format=txt")
|
||||
trac_content = Net::HTTP.get(uri).force_encoding("UTF-8")
|
||||
|
||||
if skip_content.match(trac_content)
|
||||
puts " #{page} skipped"
|
||||
next
|
||||
end
|
||||
|
||||
md_content = converter.convert(trac_content)
|
||||
md_content = converter.remove_outline(md_content)
|
||||
md_content = converter.add_trac_banner(md_content)
|
||||
|
||||
file_path.write(md_content)
|
||||
puts "+ #{page} saved"
|
||||
end
|
||||
~~~
|
||||
|
|
Loading…
Reference in a new issue