Here i got noticed that in Postgres string array comparisons we should give array after comparison operator.
For Example:
SELECT campaign_products
FROM contacts
WHERE 'PMP' LIKE ANY(campaign_products) limit 10;
SELECT campaign_products
FROM contacts
WHERE 'as' = ANY(campaign_products);
But when i use like below it giving a syntax error:
SELECT campaign_products
from contacts
where ANY(campaign_products) = 'as';
ERROR: syntax error at or near "ANY"
LINE 1: SELECT campaign_products from contacts where ANY(campaign_pr...
Can some one explain me this and let me know if there is any way to give array left to comparison operator.
It's the SQL standard.
In ISO/IEC 9075-2 (2003), Foundation, it is defined as follows:
8.8 <quantified comparison predicate>
Format
<quantified comparison predicate> ::=
<row value predicand> <quantified comparison predicate part 2>
<quantified comparison predicate part 2> ::=
<comp op> <quantifier> <table subquery>
<quantifier> ::=
<all>
| <some>
<all> ::= ALL
<some> ::=
SOME
| ANY
So it's a question you'd have to take up with the SQL standard.
But seriously, why do you need ANY =? It should not be a problem to switch the sides so that the comparison is syntactically correct.
Think of = ANY or = SOME as operators where the left side is a value and the right side is a collection of values. PostgreSQL extends the SQL standard syntax by allowing arrays on the right hand side.
Related
In my sqlalchemy ( sqlalchemy = "^1.4.36" ) query I have a clause:
.filter( some_model.some_field[2].in_(['item1', 'item2']) )
where some_field is jsonb and the value in some_field value in the db formatted like this:
["something","something","123"]
or
["something","something","0123"]
note: some_field[2] is always digits-only double-quoted string, sometimes with leading zeroes and sometimes without them.
The query works fine for cases like this:
.filter( some_model.some_field[2].in_(['123', '345']) )
and fails when the values in the in_ clause have leading zeroes:
e.g. .filter( some_model.some_field[2].in_(['0123', '0345']) ) fails.
The error it gives:
cursor.execute(statement, parameters)\\npsycopg2.errors.InvalidTextRepresentation: invalid input syntax for type json\\nLINE 3: ...d_on) = 2 AND (app_cache.value_metadata -> 2) IN (\\'0123\\'\\n ^\\nDETAIL: Token \"0123\" is invalid.
Again, in the case of '123' (or any string of digits without leading zero) instead of '0123' the error is not thrown.
What is wrong with having leading zeroes for the strings in the list of in_ clause? Thanks.
UPDATE: basically, sqlachemy's IN_ assumes int input and fails accordingly. There must be some reasoning behind this behavior, can't tell what it is. I removed that filter fromm the query and did the filtering of the ouput in python code afterwards.
The problem here is that the values in the IN clause are being interpreted by PostgreSQL as JSON representations of integers, and an integer with a leading zero is not valid JSON.
The IN clause has a value of type jsonb on the left hand side. The values on the right hand side are not explicitly typed, so Postgres tries to find the best match that will allow them to be compared with a jsonb value. This type is jsonb, so Postgres attempts to cast the values to jsonb. This works for values without a leading zero, because digits in single quotes without leading zeroes are valid representations of integers in JSON:
test# select '123'::jsonb;
jsonb
═══════
123
(1 row)
but it doesn't work for values with leading zeroes, because they are not valid JSON:
test# select '0123'::jsonb;
ERROR: invalid input syntax for type json
LINE 1: select '0123'::jsonb;
^
DETAIL: Token "0123" is invalid.
CONTEXT: JSON data, line 1: 0123
Assuming that you expect some_field[2].in_(['123', '345']) and some_field[2].in_(['0123', '345']) to match ["something","something","123"] and ["something","something","123"] respectively, you can either serialise the values to JSON yourself:
some_field[2].in_([json.dumps(x) for x in ['0123', '345']])
or use the contained_by operator (<# in PostgreSQL), to test whether some_field[2] is present in the list of values:
some_field[2].contained_by(['0123', '345'])
or cast some_field[2] to text (that is, use the ->> operator) so that the values are compared as text, not JSON.
some_field[2].astext.in_(['0123', '345'])
I would like to get the result of a query using rowMode="array" (as this is a potentially very large table and I don't want it formatted to object format) but I couldn't figure out how to pass in a array/list parameter for use in an IN operator.
const events = await t.manyOrNone({text: `select * from svc.events where user_id in ($1:list);`, rowMode: "array"}, [[1,2]]);
However, the above gives an error: syntax error at or near ":"
Removing the :list did not work either:
const events = await t.manyOrNone({text: `select * from svc.events where user_id in ($1);`, rowMode: "array"}, [[1,2]]);
Error: invalid input syntax for integer: "{"1","2"}"
I understand that this might be because I'm forced to use ParameterizedQuery format for rowMode="array" which does not allow those snazzy modifiers like :list, but this then leads to the question, if I were to use ParameterizedQuery format, then how do I natively pass in a Javascript array so that it is acceptable to the driver?
I guess an alternative formulation to this question is: how do I use arrays as parameters for ParameterizedQuery or PreparedStatements...
Answering my own question as I eventually found an answer to this issue: how to pass in arrays as params for use in the IN operator when using rowMode="array" | ParameterizedQuery | PreparedStatements.
Because this query is being parameterized in the server, we cannot use the IN operator, because the IN operator parameterize items using IN ($1, $2, $3...). Instead we need to use the ANY operator, where ANY($1) where for $1 an array is expected.
So the query that will work is:
const events = await t.manyOrNone({text: `select * from svc.events where user_id=ANY($1);`, rowMode: "array"}, [[1,2]]);
I'm getting an error, and I cannot figure out why. I know the error is telling me to cast a type but I'm not sure on what?
What part of CASE is the operator?
ERROR: operator does not exist: character varying = boolean
LINE 6: WHEN lower(foo.name) SIMILAR TO '%(foo
^
HINT: No operator matches the given name and argument type(s). You might need to add explicit type casts.
My query:
SELECT foo.name,
bar.city,
MAX(foo.total - bar.tax_amount),
CASE bar.name
WHEN lower(foo.name) SIMILAR TO '%(foo|bar|baz)%' THEN true
ELSE false
END
....
GROUP BY foo.name, bar.city;
I believe you simply want:
SELECT foo.name,
bar.city,
MAX(foo.total - bar.tax_amount),
(CASE WHEN lower(foo.name) SIMILAR TO '%(foo|bar|baz)%' THEN true
ELSE false
END)
....
GROUP BY foo.name, bar.city;
The CASE expression has two forms. One searches for values for a given expression. This takes a column or expression right after the CASE. The second searches through various expressions, stopping at the first one. This takes the WHEN clause right after the CASE.
Or even more simply:
SELECT foo.name, bar.city,
MAX(foo.total - bar.tax_amount),
(lower(foo.name) SIMILAR TO '%(foo|bar|baz)%')
....
GROUP BY foo.name, bar.city;
The CASE is not needed.
Your CASE expression had a syntax error, like #a_horse commented.
But you don't even need CASE at all in this particular case for a boolean result. Just:
foo.name ~* '(foo|bar|baz)'
That's all.
~* being the case-insensitive regular expression match operator.
Never use SIMILAR TO:
Difference between LIKE and ~ in Postgres
I have a table with a field called tags which can contain any number of strings:
Table "public.page"
Column | Type | Modifiers
----------------------+--------------------------+----------------------------------
tags | text[] | not null default ARRAY[]::text[]
I want to add a string to the tags field - but I can't seem to get the concat function to work for me. I've tried:
update page set tags=concat('My New String',tags);
ERROR: function concat(unknown, text[]) does not exist
LINE 1: update page set tags=concat('My New String',tags) where ...
^
HINT: No function matches the given name and argument types. You might need to add explicit type casts.
and
update page set tags=('My New String'||tags);
ERROR: operator is not unique: unknown || text[]
LINE 1: update page set tags = ('My New String' || tags) where w...
^
HINT: Could not choose a best candidate operator. You might need to add explicit type casts.
Any ideas?
In PostgreSQL's type system, the literal 'My New String' is not a varchar or text value, but a literal of type unknown, which can be processed as any type. (For instance, the literal for a date could be '2013-08-29'; this would not be processed as a varchar and then converted to date, it would be interpreted as a "date literal" at a very low level.)
Often, PostgreSQL can deduce the type automatically, but when it can't, you need to use one of the following to tell it that you want the literal to be treated as text:
text 'My New String' (SQL standard literal syntax)
Cast('My New String' as text) (SQL standard cast syntax, but not really a cast in this context)
'My New String'::text (PostgreSQL non-standard cast syntax, but quite readable)
In your case , the error message operator is not unique: unknown || text[] is saying that there are multiple types that Postgres could interpret the literal as, each with their own definition of the || operator.
You therefore need something like this (I've removed the unnecessary parentheses):
update page set tags = 'My New String'::text || tags;
Did you try || to concatenate?
select array['abc','def']::text[] || 'qwerty'::text;
http://www.postgresql.org/docs/current/static/functions-array.html#ARRAY-OPERATORS-TABLE
Note: this answer was in response to the OP's original (unedited) question. Other answers contain more detail relevant to the updated question.
Currently I'm trying, in a rails 3 application, to select all objects where two arrays have some overlap.
I've tried the following:
Contact.where("possible_unique_keys && ?" c.possible_unique_keys)
Which gives:
array value must start with "{" or dimension information
So I tried to convert the latter record into a postgresql array, like so:
Contact.where("possible_unique_keys && string_to_array(?)", c.possible_unique_keys)
Which gives:
operator does not exist: character varying[] && text[]
How so I get both arrays to a format that will correctly evaluate &&?
I think I've solved it for now, and come up with a way to build a query from a Ruby array to a postgresql array (thanks in part to this bit of code written for rails/postgresql interoperability.
Given a list of strings ["bill", "joe", "stu", "katie", "laura"], we have to do a little bit in the way of acrobatics to get postgresql to recognize them. This is a solution to construct the request.
request = "SELECT * from Contacts where possible_unique_keys && \'{"
list.each do |key|
key.gsub("'", "\\'")
if key == c.possible_unique_keys.last
request << "\"#{key}\"}\'"
else
request << "\"#{key}\", "
end
end
dupes = Contacts.find_by_sql(request)