Unexpected sort order on postgres left outer join - postgresql

Background
I'm using Postgres 11 and pgAdmin4 v5.2. The problem I describe below is on my dev machine which has both the postgres server and pgAdmin client.
Questions I've looked at on SO that deal with incorrect ordering seem related to collation-related issues with ordering of text fields, whereas my problem is on an integer field.
Setup
I have a table norm_plans that contains ~5k records.
Column | Type
---------------------------------
canon_id | integer
name | character varying(200)
...
other fields
canon_id is autopopulated using a sequence.
I've created a new table norm_plans_cmp as a copy of norm_plans (CREATE TABLE norm_plans_cmp AS TABLE norm_plans WITH DATA;)
I next insert some new records into norm_plans and update some existing records (fields other than canon_id.
The new records increment the sequence and are assigned canon_id values as expected.
I now want to compare norm_plans against norm_plans_cmp so I perform a left outer join:
select a.*, b.*
from norm_plans a
left outer join norm_plans_cmp b
on a.canon_id = b.canon_id
order by a.canon_id
Problem
I would expect records to be sorted by canon_id. This holds true from 1-2000, but after 2,000 I get canon_ids from 5,001 to 5,111 (which is the last canon_id) and then it again picks up from 2,001. I'm viewing this data from pgAdmin, see screenshot 1 below showing the shift from 2,000 to 5,001, and screenshot 2 showing the transition again from 5,111 back to 2,001.
Additional observations
While incorrect, the ordering seems consistent. Running the query multiple times results in the same (incorrect) ordering.
Despite my question title, I'm not totally sure the left join has anything to do with this.
Running SELECT * ... ORDER BY canon_id on norm_plans or norm_plans_cmp alone also result in incorrect ordering, albeit at different points in the order.
Answers to this SO question suggest index corruption may be a contributing problem, but I have no indexes on either norm_plans or norm_plans_cmp (canon_id is not defined as a PK).
At this point, I'm stumped!

Related

LEFT JOIN returns incorrect result in PostgreSQL

I have two tables: A (525,968 records) and B (517,831 records). I want to generate a table with all the rows from A and the matched records from B. Both tables has column "id" and column "year". The combination of id and year in table A is unique, but not in table B. So I wrote the following query:
SELECT
A.id,
A.year,
A.v1,
B.x1,
B.e1
FROM
A
LEFT JOIN B ON (A.id = B.id AND A.year = B.year);
I thought the result should contain the same total number of records in A, but it only returns about 517,950 records. I'm wondering what the possible cause may be.
Thanks!
First of all, I understand that this is an example, but postgres may hava an issues with capital letters in the table names.
Secondly, it may be a good idea to check how exactly you calculated 525,968 records. The thing is - if you use sime kind of client of database administration / queries - it may show you different / technical information about tables (there may be internal row counters in postgres that may actually differ from the number of records).
And finally to check yourself do something like
SELECT
count("A".id)
FROM
"A"

Postgresql get references from a dictionary

I'm trying to build a request to get the data from a table, but some of those columns have foreign keys I would like to replace by the associated keyword in one request.
Basically there's
table A with column 1:PKA-ID and column 2:name.
table B with column 1:PKB-ID, column 2:FKA-ID, column 3:amount.
I want to get all the lines in table B but with all foreign keys replaced by the associated names in table A.
I started building a request with a subrequest + alias to get that, but ofc I have more than one result per subrequest, yet I can't find a way to link that subrequest to the ID of table B [might be exhausted, dumb or both] from the main request. I did something like that:
SELECT (SELECT "NAME" FROM A JOIN B ON ID = FKA-ID) AS name, amount FROM TABLEB;
it feels so simple of a request yet...
You don't need a join in the subselect.
SELECT pkb_id,
(SELECT name FROM a WHERE a.pka_id = b.fka_id),
amount
FROM b;
(See it live in SQL Fiddle).
The subselect query runs for each and every row of its parent select and has the parent row available from the context.
You can also use a simple join.
SELECT b.pkb_id, a.name, b.amount
FROM b, a
WHERE a.pka_id = b.fka_id;
Note that the join version puts less restrictions on the PostgreSQL query optimizer so in some cases the join version might work faster. (For example, in PostgreSQL 9.6 the join might utilize multiple CPU units, cf. Parallel Query).

Merge SQL to Exclude Duplicate Records So Merge 2nd time Doesn't Fail

I have three tables and only one that I directly control and am doing a MERGE between them. See my abbreviated but working example here (sqlfiddle example).
I am doing a MERGE between table 1 and Table 2 to Table 3. Table 1 has duplicate data which the MERGE (erroneously) can handle on the first run (insert) but fails with this message on the second run (update).
The MERGE statement attempted to UPDATE or DELETE the same row more
than once.
My question is, can the MERGE be written to either use an EXCEPT such as
SELECT AdFull FROM [dbo].[Users] WHERE AdFull IS NOT NULL
EXCEPT
SELECT AdFull FROM [dbo].[Users]
WHERE AdFull IS NOT NULL
GROUP BY AdFull
HAVING COUNT(*) = 1
or a different Join to only show users that are not duplicated? Or even a way to select a specific one of the duplicates?
Answered Questions
MERGE is a working Insert due to the nature of Fiddle. But due (AFAIK) to the stateless nature of fiddle one never sees the error in Fiddle on a second run, because a merge never happens with the data, only inserts.
Ignore Rows: Actually I would eventually like to use an individual duplicate row via divining of one based on a condition. The actual data table I am dealing with away from the fiddle example has more columns and it would be nice to maybe select a specific row in a duplicate set due to a specific condition.
The example doesn't bare it out, but yes the duplicates are due to the computed AdFull column. Think of a system adding a temp employee, that user gets a row. Then the temp employee gets hired on as fulltime, keeps the ad account but then gets another row in the user table. Yes I know it shouldn't happen. So that is how a duplicate comes about.
(Duplicate values Table 3) Table three is a result table that can be cleaned out for any duplicates to start this process afresh.
In your MERGE statement can you do something similar this?
MERGE INTO [dbo].Table3 AS T3
USING
(
SELECT
AdFull,
MAX(StartedOn)
FROM [dbo].Table2 AS [ad]
GROUP BY AdFull
) AS T2
ON (T2.AdFull = T3.AdFull)
WHEN MATCHED THEN UPDATE blah
WHEN NOT MATCHED THEN INSERT blah
Using the MAX aggregate with a GROUP BY should give you only the information from when the temp was hired on. Then if the AdFull matches you can simply UPDATE Table3 with the most recent information and if there is no match then INSERT a new row.
UPDATE: If I fail to mention that MERGE should be used with caution I will take flak from #AaronBertrand.

Postgres LEFT JOIN is creating more rows than in left table

I am running Postgres 9.1.3 32-bit on Windows 7 x64. (Have to use 32 bit because there is no Windows PostGIS release compatible with 64 bit Postgres.) (EDIT: As of PostGIS 2.0, it is compatible with Postgres 64 bit on windows.)
I have a query that left joins a table (consistent.master) with a temporary table, then inserts the resulting data into a third table (consistent.masternew).
Since this is a left join, the resulting table should have the same number of rows as the left table in the query. However, if I run this:
SELECT count(*)
FROM consistent.master
I get 2085343. But if I run this:
SELECT count(*)
FROM consistent.masternew
I get 2085703.
How can masternew have more rows than master? Shouldn't masternew have the same number of rows as master, the left table in the query?
Below is the query. The master and masternew tables should be identically-structured.
--temporary table created here
--I am trying to locate where multiple tickets were written on
--a single traffic stop
WITH stops AS (
SELECT citation_id,
rank() OVER (ORDER BY offense_timestamp,
defendant_dl,
offense_street_number,
offense_street_name) AS stop
FROM consistent.master
WHERE citing_jurisdiction=1
)
--Here's the insert statement. Below you'll see it's
--pulling data from a select query
INSERT INTO consistent.masternew (arrest_id,
citation_id,
defendant_dl,
defendant_dl_state,
defendant_zip,
defendant_race,
defendant_sex,
defendant_dob,
vehicle_licenseplate,
vehicle_licenseplate_state,
vehicle_registration_expiration_date,
vehicle_year,
vehicle_make,
vehicle_model,
vehicle_color,
offense_timestamp,
offense_street_number,
offense_street_name,
offense_crossstreet_number,
offense_crossstreet_name,
offense_county,
officer_id,
offense_code,
speed_alleged,
speed_limit,
work_zone,
school_zone,
offense_location,
source,
citing_jurisdiction,
the_geom)
--Here's the select query that the insert statement is using.
SELECT stops.stop,
master.citation_id,
defendant_dl,
defendant_dl_state,
defendant_zip,
defendant_race,
defendant_sex,
defendant_dob,
vehicle_licenseplate,
vehicle_licenseplate_state,
vehicle_registration_expiration_date,
vehicle_year,
vehicle_make,
vehicle_model,
vehicle_color,
offense_timestamp,
offense_street_number,
offense_street_name,
offense_crossstreet_number,
offense_crossstreet_name,
offense_county,
officer_id,
offense_code,
speed_alleged,
speed_limit,
work_zone,
school_zone,
offense_location,
source,
citing_jurisdiction,
the_geom
FROM consistent.master LEFT JOIN stops
ON stops.citation_id = master.citation_id
In case it matters, I have run a VACUUM FULL ANALYZE and reindexed both tables. (Not sure of exact commands; did it through pgAdmin III.)
A left join does not necessarily have the same number of rows as the number of rows in the left table. Basically, it is like a normal join, except rows of the left table that would not appear in the normal join are also added. So, if you have more than one row in the right table that matches one row in the left table, you can have more rows in your results than the number of rows of the left table.
In order to do what you want to do, you should use a group by, and a count to detect multiples.
select citation_id
from stops join master on stops.citation_id = master.citation_id
group by citation_id
having count(*) > 1
Sometimes you know there are multiples, but don't care. You just want to take the first or top entry.
If so, you can use SELECT DISTINCT ON:
FROM consistent.master LEFT JOIN (SELECT DISTINCT ON (citation_id) * FROM stops) s
ON s.citation_id = master.citation_id
Where citation_id is the column that you want to take the first (any) row for each match.
You might want to ensure this is deterministic and use ORDER BY with some other orderable column:
SELECT DISTINCT ON (citation_id) * FROM stops ORDER BY citation_id, created_at

Cannot sort a row of size 8130, which is greater than the allowable maximum of 8094

SELECT DISTINCT tblJobReq.JobReqId
, tblJobReq.JobStatusId
, tblJobClass.JobClassId
, tblJobClass.Title
, tblJobReq.JobClassSubTitle
, tblJobAnnouncement.JobClassDesc
, tblJobAnnouncement.EndDate
, blJobAnnouncement.AgencyMktgVerbage
, tblJobAnnouncement.SpecInfo
, tblJobAnnouncement.Benefits
, tblSalary.MinRateSal
, tblSalary.MaxRateSal
, tblSalary.MinRateHour
, tblSalary.MaxRateHour
, tblJobClass.StatementEval
, tblJobReq.ApprovalDate
, tblJobReq.RecruiterId
, tblJobReq.AgencyId
FROM ((tblJobReq
LEFT JOIN tblJobAnnouncement ON tblJobReq.JobReqId = tblJobAnnouncement.JobReqId)
INNER JOIN tblJobClass ON tblJobReq.JobClassId = tblJobClass.JobClassId)
LEFT JOIN tblSalary ON tblJobClass.SalaryCode = tblSalary.SalaryCode
WHERE (tblJobReq.JobClassId in (SELECT JobClassId
from tblJobClass
WHERE tblJobClass.Title like '%Family Therapist%'))
When i try to execute the query it results in the following error.
Cannot sort a row of size 8130, which is greater than the allowable maximum of 8094
I checked and didn't find any solution. The only way is to truncate (substring())the "tblJobAnnouncement.JobClassDesc" in the query which has column size of around 8000.
Do we have any work around so that i need not truncate the values. Or Can this query be optimised? Any setting in SQL Server 2000?
The [non obvious] reason why SQL needs to SORT is the DISTINCT keyword.
Depending on the data and underlying table structures, you may be able to do away with this DISTINCT, and hence not trigger this error.
You readily found the alternative solution which is to truncate some of the fields in the SELECT list.
Edit: Answering "Can you please explain how DISTINCT would be the reason here?"
Generally, the fashion in which the DISTINCT requirement is satisfied varies with
the data context (expected number of rows, presence/absence of index, size of row...)
the version/make of the SQL implementation (the query optimizer in particular receives new or modified heuristics with each new version, sometimes resulting in alternate query plans for various constructs in various contexts)
Yet, all the possible plans associated with a "DISTINCT query" involve *some form* of sorting of the qualifying records. In its simplest form, the plan "fist" produces the list of qualifying rows (records) (the list of records which satisfy the WHERE/JOINs/etc. parts of the query) and then sorts this list (which possibly includes some duplicates), only retaining the very first occurrence of each distinct row. In other cases, for example when only a few columns are selected and when some index(es) covering these columns is(are) available, no explicit sorting step is used in the query plan but the reliance on an index implicitly implies the "sortability" of the underlying columns. In other cases yet, steps involving various forms of merging or hashing are selected by the query optimizer, and these too, eventually, imply the ability of comparing two rows.
Bottom line: DISTINCT implies some sorting.
In the specific case of the question, the error reported by SQL Server and preventing the completion of the query is that "Sorting is not possible on rows bigger than..." AND, the DISTINCT keyword is the only apparent reason for the query to require any sorting (BTW many other SQL constructs imply sorting: for example UNION) hence the idea of removing the DISTINCT (if it is logically possible).
In fact you should remove it, for test purposes, to assert that, without DISTINCT, the query completes OK (if only including some duplicates). Once this fact is confirmed, and if effectively the query could produce duplicate rows, look into ways of producing a duplicate-free query without the DISTINCT keyword; constructs involving subqueries can sometimes be used for this purpose.
An unrelated hint, is to use table aliases, using a short string to avoid repeating these long table names. For example (only did a few tables, but you get the idea...)
SELECT DISTINCT JR.JobReqId, JR.JobStatusId,
tblJobClass.JobClassId, tblJobClass.Title,
JR.JobClassSubTitle, JA.JobClassDesc, JA.EndDate, JA.AgencyMktgVerbage,
JA.SpecInfo, JA.Benefits,
S.MinRateSal, S.MaxRateSal, S.MinRateHour, S.MaxRateHour,
tblJobClass.StatementEval,
JR.ApprovalDate, JR.RecruiterId, JR.AgencyId
FROM (
(tblJobReq AS JR
LEFT JOIN tblJobAnnouncement AS JA ON JR.JobReqId = JA.JobReqId)
INNER JOIN tblJobClass ON tblJobReq.JobClassId = tblJobClass.JobClassId)
LEFT JOIN tblSalary AS S ON tblJobClass.SalaryCode = S.SalaryCode
WHERE (JR.JobClassId in
(SELECT JobClassId from tblJobClass
WHERE tblJobClass.Title like '%Family Therapist%'))
FYI, running this SQL command on your DB can fix the problem if it is caused by space that needs to be reclaimed after dropping variable length columns:
DBCC CLEANTABLE (0,[dbo.TableName])
See: http://msdn.microsoft.com/en-us/library/ms174418.aspx
This is a limitation of SQL Server 2000. You can:
Split it into two queries and combine elsewhere
SELECT ID, ColumnA, ColumnB FROM TableA JOIN TableB
SELECT ID, ColumnC, ColumnD FROM TableA JOIN TableB
Truncate the columns appropriately
SELECT LEFT(LongColumn,2000)...
Remove any redundant columns from the SELECT
SELECT ColumnA, ColumnB, --IDColumnNotUsedInOutput
FROM TableA
Migrate off of SQL Server 2000