Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit1279414

Browse files
committed
Fix misbehavior in contrib/pg_trgm with an unsatisfiable regex.
If the regex compiler can see that a regex is unsatisfiable(for example, '$foo') then it may emit an NFA having no arcs.pg_trgm's packGraph function did the wrong thing in this case;it would access off the end of a work array, and with bad luckcould produce a corrupted output data structure causing moreproblems later. This could end with wrong answers or crashesin queries using a pg_trgm GIN or GiST index with such a regex.Fix by not trying to de-duplicate if there aren't at least 2 arcs.Per bug #17830 from Alexander Lakhin. Back-patch to all supportedbranches.Discussion:https://postgr.es/m/17830-57ff5f89bdb02b09@postgresql.org
1 parenta303108 commit1279414

File tree

3 files changed

+25
-10
lines changed

3 files changed

+25
-10
lines changed

‎contrib/pg_trgm/expected/pg_word_trgm.out

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1044,3 +1044,9 @@ select t,word_similarity('Kabankala',t) as sml from test_trgm2 where t %> 'Kaban
10441044
Waikala | 0.3
10451045
(89 rows)
10461046

1047+
-- test unsatisfiable pattern
1048+
select * from test_trgm2 where t ~ '.*$x';
1049+
t
1050+
---
1051+
(0 rows)
1052+

‎contrib/pg_trgm/sql/pg_word_trgm.sql

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -43,3 +43,6 @@ select t,word_similarity('Baykal',t) as sml from test_trgm2 where 'Baykal' <% t
4343
select t,word_similarity('Kabankala',t)as smlfrom test_trgm2where'Kabankala'<% torder by smldesc, t;
4444
select t,word_similarity('Baykal',t)as smlfrom test_trgm2where t %>'Baykal'order by smldesc, t;
4545
select t,word_similarity('Kabankala',t)as smlfrom test_trgm2where t %>'Kabankala'order by smldesc, t;
46+
47+
-- test unsatisfiable pattern
48+
select*from test_trgm2where t ~'.*$x';

‎contrib/pg_trgm/trgm_regexp.c

Lines changed: 16 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -1935,9 +1935,7 @@ packGraph(TrgmNFA *trgmNFA, MemoryContext rcontext)
19351935
arcsCount;
19361936
HASH_SEQ_STATUSscan_status;
19371937
TrgmState*state;
1938-
TrgmPackArcInfo*arcs,
1939-
*p1,
1940-
*p2;
1938+
TrgmPackArcInfo*arcs;
19411939
TrgmPackedArc*packedArcs;
19421940
TrgmPackedGraph*result;
19431941
inti,
@@ -2009,17 +2007,25 @@ packGraph(TrgmNFA *trgmNFA, MemoryContext rcontext)
20092007
qsort(arcs,arcIndex,sizeof(TrgmPackArcInfo),packArcInfoCmp);
20102008

20112009
/* We could have duplicates because states were merged. Remove them. */
2012-
/* p1 is probe point, p2 is last known non-duplicate. */
2013-
p2=arcs;
2014-
for (p1=arcs+1;p1<arcs+arcIndex;p1++)
2010+
if (arcIndex>1)
20152011
{
2016-
if (packArcInfoCmp(p1,p2)>0)
2012+
/* p1 is probe point, p2 is last known non-duplicate. */
2013+
TrgmPackArcInfo*p1,
2014+
*p2;
2015+
2016+
p2=arcs;
2017+
for (p1=arcs+1;p1<arcs+arcIndex;p1++)
20172018
{
2018-
p2++;
2019-
*p2=*p1;
2019+
if (packArcInfoCmp(p1,p2)>0)
2020+
{
2021+
p2++;
2022+
*p2=*p1;
2023+
}
20202024
}
2025+
arcsCount= (p2-arcs)+1;
20212026
}
2022-
arcsCount= (p2-arcs)+1;
2027+
else
2028+
arcsCount=arcIndex;
20232029

20242030
/* Create packed representation */
20252031
result= (TrgmPackedGraph*)

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp