Github/stackdump
1
0
Fork 0
mirror of https://github.com/djohnlewis/stackdump synced 2025-04-07 18:23:27 +00:00
Code Issues Projects Releases Wiki Activity

Added check to ensure the accepted answer to a question actually exists.

Browse Source 
There may be times when it does not exist, e.g. when a question has been merged.
This commit is contained in:
Samuel Lai 2012-08-18 20:18:26 +10:00
parent 9b9b71077c
commit 5fc56e4329
1 changed files with 14 additions and 6 deletions

16
python/src/stackdump/dataproc/import.py

@ -349,7 +349,7 @@ class PostContentHandler(xml.sax.ContentHandler):
for id, q in self.unfinished_questions.items(): for id, q in self.unfinished_questions.items():
if len(q['answers']) >= q['answerCount']: if len(q['answers']) >= q['answerCount']:
if len(q['answers']) > q['answerCount']: if len(q['answers']) > q['answerCount']:
print('Question ID [%s] expected to have %d answers, but got %d instead. Ignoring inconsistency.' % (q['id'], q['answerCount'], len(q['answers']))) print('Question ID [%s] expected to have %d answers, but has %d instead. Ignoring inconsistency.' % (q['id'], q['answerCount'], len(q['answers'])))
try: try:
# question is complete, store it. # question is complete, store it.
@ -418,7 +418,7 @@ class PostContentHandler(xml.sax.ContentHandler):
q['comments'].extend(comments[q['id']]) q['comments'].extend(comments[q['id']])
if len(q['comments']) != q['commentCount']: if len(q['comments']) != q['commentCount']:
print('Post ID [%s] expected to have %d comments, but got %d instead. Ignoring inconsistency.' % (q['id'], q['commentCount'], len(q['comments']))) print('Post ID [%s] expected to have %d comments, but has %d instead. Ignoring inconsistency.' % (q['id'], q['commentCount'], len(q['comments'])))
# add comments to the answers # add comments to the answers
for a in q['answers']: for a in q['answers']:
@ -426,7 +426,7 @@ class PostContentHandler(xml.sax.ContentHandler):
a['comments'].extend(comments[a['id']]) a['comments'].extend(comments[a['id']])
if len(a['comments']) != a['commentCount']: if len(a['comments']) != a['commentCount']:
print('Post ID [%s] expected to have %d comments, but got %d instead. Ignoring inconsistency.' % (a['id'], a['commentCount'], len(a['comments']))) print('Post ID [%s] expected to have %d comments, but has %d instead. Ignoring inconsistency.' % (a['id'], a['commentCount'], len(a['comments'])))
doc = { } doc = { }
@ -474,7 +474,15 @@ class PostContentHandler(xml.sax.ContentHandler):
question_obj = { } question_obj = { }
question_obj['id'] = q['id'] question_obj['id'] = q['id']
if 'acceptedAnswerId' in q: if 'acceptedAnswerId' in q:
# check that the accepted answer is in the question's answers section.
# sometimes they're not, e.g. if the question was merged - the
# acceptedAnswerId would point to an answer in another question
# instead. We don't deal with merged questions yet, so this option
# means questions won't appear to have answers when they don't.
if q['acceptedAnswerId'] in post_ids:
question_obj['acceptedAnswerId'] = q['acceptedAnswerId'] question_obj['acceptedAnswerId'] = q['acceptedAnswerId']
else:
print 'Question [ID# %i] had an unknown answer. Possibly been merged or migrated. Ignoring inconsistency.' % (q['id'], )
question_obj['creationDate'] = q['creationDate'] question_obj['creationDate'] = q['creationDate']
question_obj['score'] = q['score'] question_obj['score'] = q['score']
question_obj['viewCount'] = q['viewCount'] question_obj['viewCount'] = q['viewCount']
@ -515,7 +523,7 @@ class PostContentHandler(xml.sax.ContentHandler):
questions_to_commit = [ ] questions_to_commit = [ ]
for id, q in self.unfinished_questions.items(): for id, q in self.unfinished_questions.items():
print('Question [ID# %d] was expected to have %d answers, but got %d instead. Ignoring inconsistency.' % (q['id'], q['answerCount'], len(q['answers']))) print('Question [ID# %d] was expected to have %d answers, but has %d instead. Ignoring inconsistency.' % (q['id'], q['answerCount'], len(q['answers'])))
try: try:
# question is complete, store it. # question is complete, store it.