Project

General

Profile

Bug #617

Search results sometimes produce broken links to works

Added by Luke Murphey over 10 years ago. Updated over 10 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Target version:
Start date:
08/04/2013
Due date:
% Done:

100%


Description

Some search results are not accessible since the verse marker is not recognized.

Steps to reproduce:

Associated revisions

Revision 487 (diff)
Added by Luke Murphey over 10 years ago

Fixed issue where the content search didn't make the indexes correctly. Reference #617.

Revision 487 (diff)
Added by Luke Murphey over 10 years ago

Fixed issue where the content search didn't make the indexes correctly. Reference #617.

Revision 476 (diff)
Added by Luke Murphey over 10 years ago

Fixed issue where the content search didn't make the indexes correctly. Reference #617.

Revision 488 (diff)
Added by Luke Murphey over 10 years ago

Fixed issue where importing of LXX failed to account for the presence of two versions of Esther. Closes #617.

Revision 488 (diff)
Added by Luke Murphey over 10 years ago

Fixed issue where importing of LXX failed to account for the presence of two versions of Esther. Closes #617.

Revision 477 (diff)
Added by Luke Murphey over 10 years ago

Fixed issue where importing of LXX failed to account for the presence of two versions of Esther. Closes #617.

History

#1 Updated by Luke Murphey over 10 years ago

There are two versions of Esther in the LXX. The second instance has the same name as the first which generates confusion. I suspect this is why the references are failing.

#2 Updated by Luke Murphey over 10 years ago

  • Status changed from New to In Progress

#3 Updated by Luke Murphey over 10 years ago

The problem is that there are two Esther's:

select * from reader_division where title = "Esther" 

The solution is to change the title of the second Esther to something like "Esther (Greek)":

update reader_division set title_slug="esther-greek", descriptor="Esther-Greek" where id = 31318

#4 Updated by Luke Murphey over 10 years ago

This indeed fixed the problem. Now I need to update the searches indicies.

#5 Updated by Luke Murphey over 10 years ago

Reimporting with:

python manage.py import_unbound_bible -f ../var/resources/Unbound_Bible/lxx_a_accents/lxx_a_accents_utf8.txt

This was done after deleting the LXX:

delete from reader_work where title_slug = "lxx" 
delete from reader_workalias where title_slug = "lxx" 

#6 Updated by Luke Murphey over 10 years ago

The existing index for LXX was deleted using:

from reader.contentsearch import WorkIndexer
from whoosh.qparser import QueryParser

inx = WorkIndexer.get_index()
parser = QueryParser("content", inx.schema)
inx.delete_by_query(parser.parse(u'work:lxx'))
writer = inx.writer()
writer.commit(optimize=True)

The indexes were rebuilt using:

python manage.py make_search_indexes -w lxx
from reader.models import Work
from reader.contentsearch import WorkIndexer
works = Work.objects.filter(title_slug="lxx")
work = works[0]
WorkIndexer.index_work(work)

#7 Updated by Luke Murphey over 10 years ago

  • Status changed from In Progress to Closed
  • % Done changed from 0 to 100

Applied in changeset r488.

Also available in: Atom PDF