The data coming into the warehouse is structured like this
{"Client ID":"1234567","client_name":"Kareem" }
when I use the function
JSON_EXTRACT_PATH_TEXT(COLUMN_NAME, 'Client ID')
it give me this error :
Invalid extraction path 'Client ID': invalid token at position 7.
is there another workaround to get the value for this key ? or for using this key with the function to get the value out of the JSON column ?
Per its documentation, the JSON_EXTRACT_PATH_TEXT function follows the standard notation for object keys. Keys that carry spaces are required to be double-quoted, and the same applies here. The following works for your example:
JSON_EXTRACT_PATH_TEXT(COLUMN_NAME, '"Client ID"')
COLUMN_NAME:"Client ID"
worked for me
The following does work, which is supposed to be close to equivalent. My example assumes your values are a string, since that is the input of the function you were trying to use:
WITH x AS (SELECT '{"Client ID":"1234567","client_name":"Kareem" }' as json_string)
SELECT parse_json(json_string):"Client ID"
FROM x;
Related
I cant seem to get the filter clause to retrieve documents from my index using a regex clause. The schema for my index is straight forward, I only have a single field which is both searchable and filterable and is of type Edm.String, called someId (which normally contains a hash value of something) and has sample values like:
someId
k6l7k2oj
k6l55iq8
k6l61ff8
...
I need to be able to extract all values from this field that start with K6 and end with 8. So based on the documentation I am using this in my POST request body
{
"filter": "search.ismatch('/^k6[?]*d$/','someId','simple','all')",
"select":"someId",
"count":"true"
}
and it comes up with nothing.
On the other hand if I simplify and say I only need data where someId starts with K6, I seem to get some success if i just use a wild card.
like this:
{
"filter": "search.ismatch('k6l*','someId','simple','all')",
"select":"someId",
"count":"true"
}
I do get what I am looking for. Question is why does the regex not work with search.isMatch(), what am i missing?
...
Regex is part of the full Lucene syntax; It is not available in the simple syntax. Try changing the third parameter of search.ismatch to 'full'.
Also, did you mean to use search.ismatch or search.ismatchscoring? The latter is functionally equivalent to using the top-level search, searchFields, queryType, and searchMode parameters. The former does not count matches towards relevance scoring.
Your regex does not do what you intend either it seems. I tested your regex with your sample data and it does not match. Try this regex instead:
^k6.{5}8$
It matches a lowercase k6 from the start of the string, followed by 5 characters of anything and finally an 8.
Complete example
{ "filter": "search.ismatch('^k6.{5}8$','someId','full','all')", "select":"someId", "count":"true" }
Thanks to Dan and Bruce.
This exact expression worked for me
{
"filter": "search.ismatch('/k6.{5}8/','someId','full','all')",
"select":"someId",
"count":"true"
}
I'm trying to get a query working using a case statement, and can't figure out how to get the case to return a column value instead of a constant. I have the query working perfectly, except that the column names I'm providing for the results are being quoted or otherwise mishandled by Cake or maybe PDO somewhere down in a layer that I can't dig my way through. I got as far down as bindValue, but none of the documentation I encountered along the way tells me how to do this.
I have found this example comment:
$statement->bindValue(1, 'a title');
$statement->bindValue(2, 5, PDO::INT);
$statement->bindValue('active', true, 'boolean');
$statement->bindValue(5, new \DateTime(), 'date');
but in all these cases, the value provided is a constant. I need to pass in a string that is the name of the column that I want returned.
I tried both 'string' (resulted in quoted column name) and 'integer' (resulted in 0). I tried PDO::FETCH_COLUMN (seemed highly unlikely, but looked like the next best bet from http://php.net/manual/en/pdo.constants.php, and easy to try it...). I tried 'literal', inspired by the way you can put literal strings into expressions (resulted in Error: unknown type "literal"). That error message led me to src/Database/Type.php, but nothing in there helped me either.
So, I'm pretty much stumped. Here's a simple version of the code I have (leaving out a couple of conditions and unrelated columns):
$query = $this->Games->find();
$team_id = $query->newExpr()->addCase(
[$query->newExpr()->eq('Games.status', 'home_default')],
['home_team_id', 'away_team_id'],
['string', 'string']
);
$defaulting = $query
->select([
'id' => $team_id,
'count' => 'COUNT(Games.id)',
])
->where([
'Games.status IN' => ['home_default', 'away_default'],
])
->group('id')
->toArray();
This generates this SQL:
SELECT
(CASE WHEN Games.status = 'home_default'
THEN 'home_team_id' ELSE 'away_team_id' END) AS `id`,
COUNT(Games.id) AS `count`
FROM games Games
WHERE Games.status in ('home_default','away_default')
GROUP BY id
Note that THEN 'home_team_id' ELSE 'away_team_id' END should be simply THEN home_team_id ELSE away_team_id END. This will then allow me to read the list of ids of teams that have defaulted games along with the number of games they defaulted.
By default the values passed to the second argument of QueryExpression::addCase() are being treated as to be converted to literal values, not as identifiers. If you need the latter, then you should use an expression, an IdentifierExpression.
use Cake\Database\Expression\IdentifierExpression;
// ...
$team_id = $query->newExpr()->addCase(
[
$query->newExpr()->eq('Games.status', 'home_default')
],
[
new IdentifierExpression('Games.home_team_id'),
new IdentifierExpression('Games.away_team_id')
]
);
Also ditch the third argument in this case, you don't want the values to be string literals (for expressions the types would be ignored anyways).
I see the document
https://developers.google.com/appengine/docs/python/ndb/keyclass#Key_integer_id
Returns the integer id in the last (kind, id) pair, or None if the key
has an string id or is incomplete.
see I think the id of a key can be a int ; so I write
r = ndb.Key(UserSession, int(id)).get()
if r:
return r.session
but the dev_server.py , will always raise
File "/home/bitcoin/down/google_appengine/google/appengine/datastore/datastore_stub_util.py", line 346, in CheckReference
raise datastore_errors.BadRequestError('missing key id/name')
BadRequestError: missing key id/name
I chanage the int(id) -> str(id)
seems right ;
so my question is , How to use ndb key with integer_id ?
the model is
class UserSession(ndb.Model):
session = ndb.BlobProperty()
The type of the id you use when reading the entity must match the type of the id you used when you wrote the entity. Normally, integer ids are assigned automatically when you write a new entity without specifying an id or key; you then get the id out of the key returned by entity.put(). It is generally not recommended to assign your own integer ids; when the app assigns the keys, the convention is that they should be strings.
There's an easier way to fetch:
UserSession.get_by_id(int(id))
https://developers.google.com/appengine/docs/python/ndb/modelclass#Model_get_by_id
If that doesn't work, I suspect that id is wrong or empty.
There must be something wrong with your variable 'id'.
Your code here should be no problem, and it's better to user long instead of int.
You can try your code on interactive console of development server with specific integer id.
It may be easier to identify your entities in the sessions with their keys instead of their ids. There really is no need to extract the ID from the key to identify the session (other than maybe saving a bit of memory. I think the way your thinking is based on a RDB. I learned that using the key actually makes entity/session identifications easier.
'id' is also a python builtin function. Maybe you are taking that by mistake.
we using mybatis 3.1.1.
we found for oracle the result map returned contains column name in Capital letters and in case of mySql the result map returned contains column name in small letters.
My question is : Is there is any way i can to write some sort of interceptor so that i can modify the result returned by result map.
Thanks.
I'm afraid the answer is that MyBatis doesn't provide any direct way to control the case of the keys in a result map. I asked this question recently on the MyBatis Google Group: https://groups.google.com/forum/?fromgroups#!topic/mybatis-user/tETs_JiugNE
The outcome is dependent on the behavior of the JBDC driver.
It also turns out that doing column aliasing as suggested by #jddsantaella doesn't work in all cases. I've tested MyBatis-3.1.1 with three databases in the past: MySQL, PostgreSQL and H2 and got different answers. With MySQL, the case of the column alias does dictate the case of the key in the hashmap. But with PostgreSQL it is always lowercase and with H2, it is always uppercase. I didn't test whether column aliases will work with Oracle, but by default it appears to return capital letters.
I see two options:
Option 1: Create some helper method that your code will always use to pull the data out of the returned map. For example:
private Object getFromMap(Map<String, Object> map, String key) {
if (map.containsKey(key.toLowerCase())) {
return map.get(key.toLowerCase());
} else {
return map.get(key.toUpperCase());
}
}
Option 2: Write a LowerCaseMap class that extends from java.util.AbstractMap or java.util.HashMap and wrappers all calls to put, putAll and/or get to always be lower case. Then specify that MyBatis should use your specific LowerCaseMap rather than a standard HashMap, when populating the data from the query.
If you like this idea and want help on how to tell MyBatis how to use a different concrete collection class, see my answer to this StackOverflow question: https://stackoverflow.com/a/11596014/871012
What if you modify the query so you get the exactly column name you need? For example:
select my_column as MY_COLUMN from ...
The idea of a LowerCaseMap as the ResultType is sound, but you can probably avoid writing your own. In my case I'm using org.apache.commons.collections4.map.CaseInsensitiveMap:
<select id="getTableValues"
resultType="org.apache.commons.collections.map.CaseInsensitiveMap">
SELECT *
FROM my_table
WHERE seq_val=#{seq_val}
</select>
1) Finding by instance object
Assuming I have the instance object called #topic. I want to retrieve the answers for this given topic. I was thinking I should be able to pass in :topics=>#topic, but i had to do the very ugly query below.
#answers = Answers.where(:topic_ids => {"$in" => [#topic.id]})
2) Getting the string representation of the id. I have a custom function (shown below). But shouldn't this be a very common requirement?
def sid
return id.to_s
end
If your associations are set up correctly, you should be able to do:
#topic.answers
It sounds like the above is what you are looking for. Make sure you have set up your associations correctly. Mongoid is very forgiving when defining associations, so it can seem that they are set up right when there is in fact a problem like mismatched names in references_many and referenced_in.
If there's a good reason why the above doesn't work and you have to use a query, you can use this simple query:
#answers = Answer.where(:topic_ids => #topic.id)
This will match any Answer record whose topic_ids include the supplied ID. The syntax is the same for array fields as for single-value fields like Answer.where(:title => 'Foo'). MongoDB will interpret the query differently depending on whether the field is an array (check if supplied value is in the array) or a single value (check if the supplied value is a match).
Here's a little more info on how MongoDB handles array queries:
http://www.mongodb.org/display/DOCS/Advanced+Queries#AdvancedQueries-ValueinanArray