Regression with large XML data input

Поиск
Список
Период
Сортировка
От Michael Paquier
Тема Regression with large XML data input
Дата
Msg-id aIGknLuc8b8ega2X@paquier.xyz
обсуждение исходный текст
Ответы Re: Regression with large XML data input
Re: Regression with large XML data input
Список pgsql-hackers
Hi all,
(Adding in CC Tom and Eric, as committer and  author.)

A customer has reported a regression with the parsing of rather large
XML data, introduced by the set of backpatches done with f68d6aabb7e2
& friends.

The problem is introduced by the change from
xmlParseBalancedChunkMemory() to xmlNewNode() +
xmlParseInNodeContext() in xml_parse(), to avoid an issue in
xmlParseBalancedChunkMemory() in the range of libxml2 2.13.0-2.13.2
for a bug that has already been fixed upstream, where we use a
temporary root node for the case where parse_as_document is false.

If the input XML data is large enough, one gets a failure at the top
of the latest branches, and it worked properly before.  Here is a
short test case (courtesy of a colleague, case that I've modified
slightly):
CREATE TABLE xmldata (id BIGINT PRIMARY KEY, message XML );
DO $$ DECLARE size_40mb TEXT := repeat('X', 40000000);
BEGIN
 BEGIN
   INSERT INTO xmldata (id, message) VALUES
     ( 1, (('<Root><Item><Name>Test40MB</Name><Content>' || size_40mb || '</Content></Item></Root>')::xml) );
   RAISE NOTICE 'insert 40MB successful';
   EXCEPTION WHEN OTHERS THEN RAISE NOTICE 'Error insert 40MB: %', SQLERRM;
 END;
END $$;

Switching back to the previous code, where we rely on
xmlParseBalancedChunkMemory() fixes the issue.  A quick POC is
attached.  It fails one case in check-world with SERIALIZE because I
am not sure it is possible to pass down some options through
xmlParseBalancedChunkMemory(), still the regression is gone, and I am
wondering if there is not a better solution to be able to dodge the
original problem and still accept this case.  One good thing is that
xmlParseBalancedChunkMemory() is able to return a list of nodes, that
we need for this code path of xml_parse().  So perhaps one solution
would be the addition of a code path with
xmlParseBalancedChunkMemory() depending on the options we want to
process, keeping the code path with the fake "content-root" for the
XML SERIALIZE case.

The patch in question has been applied first to 6082b3d5d3d1 on HEAD
impacting v18~, then it has been backpatched down to all stable
branches, like f68d6aabb7e2, introducing the regression in all the
stable branches since the minor releases done in August 2024, as of:
12.20, 13.16, 14.13, 15.8, 16.4.

Thoughts or comments?
--
Michael

Вложения

В списке pgsql-hackers по дате отправления: