I want to format dateTime with in the cts:element-values itself. Can anyone help me around this?
I have a dateTime format string -
let $date-format := "[Y0001]-[M01]-[D01]T[h01]:[m01]:[s01].[f1]"
and I want to use it in a query like this -
cts:element-values(
xs:QName($field),
(),
($direction),
cts:and-query((cts:collection-query("urn:iddn:collections:searchable"), cts:query($cts-query)))
)
Provided $field is of type dateTime.
You can accomplish this by writing a User-Defined Function. UDFs are run as map/reduce, so they are very fast even with a large data set. I wrote an example UDF to create a day-of-the-week facet based on dateTime data. That example is based on MarkLogic 6, but should still work in MarkLogic 8.
The good thing is that UDFs are very fast. The tricky part is that you'll have to write it in C++. Full documentation in the User-Defined Functions section of the MarkLogic documentation.
Related
In my ML db, we have documents with distributor code like 'DIST:5012' (DIST:XXXX) XXXX is a four-digit number.
currently, in my TDE, the below code works well.
However instead of concat all the raw distributor codes, I want to simply concat the number part only. I used the fn:substring-after XQuery function. However, it won't work. It won't show that distributorCode column in the SQL View anymore. (Below code does not work.)
What is wrong? How to fix that?
Both fn:substring-after and fn:string-join is in TDE Dialect page.
https://docs.marklogic.com/9.0/guide/app-dev/TDE#id_99178
substring-after() expects a single string as input, not a sequence of strings.
To demonstrate, this will not work:
let $dist := ("DIST:5012", "DIST:5013")
return substring-after($dist, "DIST:")
This will:
for $dist in ("DIST:5012", "DIST:5013")
return substring-after($dist, "DIST:")
I need to double check what XPath expressions will work in a DTE, you might be able to change it to apply the substring-after() function in the last step:
fn:string-join( distributors/distributor/urn/substring-after(., 'DIST:'), ';')
My preceding module in Integromat gives me an expiration date in UNIX time, which is 1640930400.
When I use the FormatDate function to convert this, I'm getting 12/31/1969 when I was expecting to get 12/31/2021. I can't seem to figure out what I'm doing wrong here. Any help would be much appreciated.
Use this instead to first parse the date and then apply the desired formatting to get the results that you want,
{{formatDate(parseDate(1.date; "X"); "MM/DD/YYYY")}}
I want to import a huge csv-file with datetime-column in format YYYY.MM.DD hh:mm:ss.nnn (2015.09.28 00:00:02.721). It is the supported string literal format for datetime:
datetime (Transact-SQL).
I'm using DT_DBTIMESTAMP as the Integration Services data type:
Integration Services Data Types.
But the import does not work due to convertion error. I can only import in format YYYY-MM-DD hh:mm:ss.nnn. My OS, SQL 2014 and DB are all in german. How can I execute this task without search and replace in csv with regular expression?
I'm not sure I understand what you mean about Regular Expressions - but here is a solution for you:
You can import the file into a temporary holding table and then convert it to a datetime using the DataConversion task.
You can also apply the following SQL to your table - which is probably going to be quicker:
UPDATE tempTable SET NewDateCol = CAST(OldDateCol AS DATETIME)
where OldDateCol is the imported text field containing the "invalid" date format.
Because we know that this will work:
SELECT CAST('2012.05.06 11:25:33.123' AS DATETIME)
Pure SSIS dataflow solution - Derived Column transformation with the following formula for NewDateCol
(DT_DBTIMESTAMP2)(REPLACE(SUBSTRING(OldDateCol,1,10),".","-")+SUBSTRING(OldDateCol, 11, LEN(OldDateCol)))
Basically, this expression replaces . with - on the first 10 symbols and then appends the rest.
My workaround:
1. CREATE TABLE with datetime (it makes milliseconds rounding error - 2015.09.28 00:00:01.159 is 2015-09-28 00:00:01.160 etc., for whatever reason) or datetime2(3) (2015.09.28 00:00:01.159 is 2015-09-28 00:00:01.1590000 - correct but oversized)
2. SET DATEFORMAT ymd
3. BULK INSERT from csv with TABLOCK and CODEPAGE='RAW' to increase perfomance
I am currently writing a block of code on R which collects data via a SPARQL query. My problem is when I am trying to filter the query by date, R gives an error of "unexpected numeric constant".
There is no any mistake in the SPARQL code because when I run the exact code on the endpoint I receive data normally.
You will find the block of code where I have the problem. It does not matter the lines before and after, just the second line of the date filter.
...
OPTIONAL {?seller gr:legalName ?sellerLegalName} .
FILTER REGEX (STR(?date) >= "2015-01-01") .
FILTER NOT EXISTS {?spendingItem elod:hasCorrectedDecision ?correctedDecision} .
...
Please, I would kindly ask for your help! :)
For any further information that you want in order to solve the problem, feel free to contact with me.
Thank you all!!!
SOLVED!
I found that the date should be passed as timestamp!
Also, I found a useful site where you can convert any date in timestamp and vice versa.
I would like to thank you all for your responses and your useful help!
You should filter it as a date/time value rather than as a string - perhaps that will help:
FILTER (?date > "2015-01-01"^^xsd:date)
See this answer: SPARQL date range
I´m new to MDX and I have a simple question. I work with the TFS Cube it is named as Team System. My problem:
I have an IIF expression where I want to check additional my expression with an AND operator. There I want to compare two DateTime objects. The report should only show me the data from the actual date. Here my code:
IIF(ISEMPTY(SUM(YTD(
[Work Item].[PlannedWeek__HierarchyByWeek].CurrentMember),
[Measures].[EffectivelyValue]))
AND[Work Item].[PlannedWeek__HierarchyByWeek].CurrentMember < Now()
, [Measures].[EffectivelyValue]
, SUM(YTD(
[Work Item].[PlannedWeek__HierarchyByWeek].CurrentMember),
[Measures].[EffectivelyValue]) )
Planned Week is a self created field which has the DateTime datatype. The Now() function has also a DateTime datatype so the comparision should be right but it happens nothing.
Thanking you in anticipation
Eugen
Hierarchy members in MDX have a data type of 'member', and do not have a 'primitive' data type like datetime, string, or integer. Only member properties have 'primitive' data types. You could either define a property like datetime of your week attribute. Assuming you are SQL Server Analysis Services, this would be done via relationships.
Or you could use string operations to extract the date information from the UniqueName property which avoids having to change the cube. The UniqueName contains the data that you defined as the key in your cube design. Assuming your week hierarchy members have a key from which you can extract something like 20130820 for August 20, 3013 via string functions (I just will use Mid(, 30, 8) as an example below), you could do something like
CLng(Mid([Work Item].[PlannedWeek__HierarchyByWeek].CurrentMember.UniqueName, 30, 8))
<
CLng(Format(Now(), "yyyymmdd"))
You will have to check what exactly the CurrentMember.UniqueName shows in your cube to adapt the above code.
And finally, you could of course also use string methods to extract the relevant parts from the UniqueName and then the CDate function on that to compare to an unchanged Now(), i. e. do all operations on the left side of the <.
Hello thank you very much for you answer. I tried:
Mid([Work Item].[xxxx_PlannedWeek__HierarchyByWeek].CurrentMember.UniqueName,58,10)
shows e.g. 2013-07-21, 2013-07-28 (it shows the week endings)
so I tried this:
AND CLng(Mid([Work Item].[xxxx_PlannedWeek__HierarchyByWeek].CurrentMember.UniqueName,58,10))
<CLng(Format(Now(), "yyyy-mm-dd"))
But it happens nothing. If I execute it in a single way it shows everywhere "true". But I have datasets with dates which are e.g. > 2013-08-23. So there should be false values too.
EDIT: OK I solved the problem. The
Format(Now(), "yyyy-mm-dd")
must be
Format(Now(), "yyyy-MM-dd")