Discourse/discourse
Discourse/discourse
2
0
Fork 0
mirror of https://github.com/discourse/discourse.git synced 2025-09-06 09:10:25 +08:00
Code Packages Activity Actions 11

fix vbulletin importer to import unreferenced attachments (#12187)

Browse source
This commit is contained in:
Michael Maroszek 2021-04-19 21:05:16 +02:00 • committed by GitHub
parent f74d9ca399
commit 5bec0e5763
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 57 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

57
script/import_scripts/vbulletin.rb
View file

@ -570,6 +570,25 @@ EOM
def import_attachments def import_attachments
puts '', 'importing attachments...' puts '', 'importing attachments...'
mapping = {}
attachments = mysql_query(<<-SQL
SELECT a.attachmentid, a.contentid as postid, p.threadid
FROM #{TABLE_PREFIX}attachment a, #{TABLE_PREFIX}post p
WHERE a.contentid = p.postid
AND contenttypeid = 1 AND state = 'visible'
SQL
)
attachments.each do |attachment|
post_id = post_id_from_imported_post_id(attachment['postid'])
post_id = post_id_from_imported_post_id("thread-#{attachment['threadid']}") unless post_id
if post_id.nil?
puts "Post for attachment #{attachment['attachmentid']} not found"
next
end
mapping[post_id] ||= []
mapping[post_id] << attachment['attachmentid'].to_i
end
current_count = 0 current_count = 0
total_count = mysql_query(<<-SQL total_count = mysql_query(<<-SQL
@ -594,6 +613,10 @@ EOM
matches = attachment_regex.match(s) matches = attachment_regex.match(s)
attachment_id = matches[1] attachment_id = matches[1]
unless mapping[post.id].nil?
mapping[post.id].delete(attachment_id.to_i)
end
upload, filename = find_upload(post, attachment_id) upload, filename = find_upload(post, attachment_id)
unless upload unless upload
fail_count += 1 fail_count += 1
@ -603,6 +626,40 @@ EOM
html_for_upload(upload, filename) html_for_upload(upload, filename)
end end
# make resumed imports faster
if new_raw == post.raw
unless mapping[post.id].nil? || mapping[post.id].empty?
imported_text = mysql_query(<<-SQL
SELECT p.pagetext
FROM #{TABLE_PREFIX}attachment a, #{TABLE_PREFIX}post p
WHERE a.contentid = p.postid
AND a.attachmentid = #{mapping[post.id][0]}
SQL
).first["pagetext"]
imported_text.scan(attachment_regex) do |match|
attachment_id = match[0]
mapping[post.id].delete(attachment_id.to_i)
end
end
end
unless mapping[post.id].nil? || mapping[post.id].empty?
mapping[post.id].each do |attachment_id|
upload, filename = find_upload(post, attachment_id)
unless upload
fail_count += 1
next
end
# internal upload deduplication will make sure that we do not import attachments again
html = html_for_upload(upload, filename)
if !new_raw[html]
new_raw += "\n\n#{html}\n\n"
end
end
end
if new_raw != post.raw if new_raw != post.raw
PostRevisor.new(post).revise!(post.user, { raw: new_raw }, bypass_bump: true, edit_reason: 'Import attachments from vBulletin') PostRevisor.new(post).revise!(post.user, { raw: new_raw }, bypass_bump: true, edit_reason: 'Import attachments from vBulletin')
end end