I am creating a report and would like to retrieve data based on a date field (string) in the format YYYYWW. I want to filter the data so I retrieve the current month, previous month, the month before that and older values - using 4 different datasets.
For example:
201547 = November 2015
How do I create a calculated member which converts a string into the month name and year? I would like to use this value to filter out the data from the cube. Is there any other way of doing this?
:]1
Try adding a calculated field and set it to this expression:
=MonthName(Month(DateAdd("ww",CInt(MID("201547",5,LEN("201547")))
,DateSerial(CInt(LEFT("201547",4)),1,1)))) & " " & LEFT("201547",4)
Then use that field to filter your report.
Replace the string for the field returning "201547"
=MonthName(Month(DateAdd("ww",
CInt(MID(Fields!YearWeek.Value,5,LEN(Fields!YearWeek.Value)))
,DateSerial(CInt(LEFT(Fields!YearWeek.Value,4)),1,1))))
& " " & LEFT(Fields!YearWeek.Value,4)
UPDATE: MDX script for querying a calulated member with the required format:
I've tested with the following MDX:
WITH MEMBER [Measures].[month&year] AS
'FORMAT(DateAdd("ww",STRTOVALUE(MID("201547",5,LEN("201547"))) ,CDATE(DateSerial(LEFT("201547",4),1,1))),"MMMM") + " " + LEFT("201547",4)'
SELECT { [month&year] } ON COLUMNS
FROM [Your Cube]
To use it in your query replace "201547" for [Dimension].[Attribute].MemberValue that corresponds to your cube structure:
WITH MEMBER [Measures].[month&year] as
'FORMAT(DateAdd("ww",STRTOVALUE(MID([Dimension].[Attribute].MemberValue,5,LEN([Dimension].[Attribute].MemberValue)))
,CDATE(DateSerial(LEFT([Dimension].[Attribute].MemberValue,4),1,1))),"MMMM") + " " + LEFT([Dimension].[Attribute].MemberValue,4)'
SELECT
{ [Dimension].[Attribute].[Attribute] }
{[month&year]} ON COLUMNS
FROM [Your Cube]
Once you have this field in your dataset you can easily filter the values using a parameter.
Let me know if this can help you.
Using a combination of the resources found out on the internet you can find the information you require.
First you need to find the first day of the first week of the year. This has already been determined by this answer as
DateAdd("d",
(-1) * (CInt(New DateTime(Year(Now), 1, 1).DayOfWeek) +
IIf(CInt(New DateTime(Year(Now), 1, 1).DayOfWeek) < DayOfWeek.Monday,
7,
0)
) + 1,
New DateTime(Year(Now), 1, 1))
This needs to be adapted for any year - not just the current one by finding the year from the string as follows
CInt(Left(Parameters!myDate.Value,4))
Now, as noted in this resource you can use a DateAdd to add the number of weeks of the year you have entered to the first date of the year
=(DateAdd(DateInterval.WeekOfYear,
CInt(Right(Parameters!myDate.Value,2))-1,
<<FIRST_DAY_OF_YEAR>>)
You then want to get the month name of the month of this date, using Month and MonthName. Smashing it all together would result in the following
=MonthName(Month(DateAdd(DateInterval.WeekOfYear,
CInt(Right(Parameters!myDate.Value,2))-1,
DateAdd("d",
(-1) * (CInt(New DateTime(CInt(Left(Parameters!myDate.Value,4)), 1, 1).DayOfWeek) +
IIf(CInt(New DateTime(CInt(Left(Parameters!myDate.Value,4)), 1, 1).DayOfWeek) < DayOfWeek.Monday,
7,
0)
) + 1,
New DateTime(CInt(Left(Parameters!myDate.Value,4)), 1, 1))
)))
This will get a Textual value of the month for the interval
Create a new placeholder in the same textbox and then repeat the above for finding the year
=Year(DateAdd(DateInterval.WeekOfYear, ...
(Remember to remove the last brace otherwise you’ll get a End of Statement expected error.)
This shoud give what you require, so
201501 = December 2014
201547 = November 2015
201553 = December 2015
201622 = May 2016
Let me know if this helps, or if you have further questions
Part 1: Establishing which day falls into which week
ISO-8601 declares the international standards for converting dates into week numbers, and vice-versa.
ISO standard weeks start on a Monday. A year can have either 52 or 53 weeks.
There are 4 ways of establishing when Week 1 starts:
- It is the first week with a majority (4 or more) of its days in January.
- Its first day is the Monday nearest to 1 January.
- It has 4 January in it. Hence the earliest possible dates are 29 December through 4 January, the latest 4 through 10 January.
- It has the year's first working day in it - i.e. excluding Saturdays, Sundays and the 1st of January.
Part 2: Excel calculations
This answer assumes that the dates are being stored as 6-character strings.
First determine the year and what day of the week it starts on.
1) Strip the first four characters from the string.`
=left(<cell containing the year+week string>, 4)
2) Prepend "1/1/" to it.
="1/1/"&left(<cell containing the year+week string>, 4)
3) Convert this string to an Excel date
=datevalue(<the last result>)
4) Determine the Weekday of the 1st day of that year.
=weekday(<the last result>)
5) Calculate the Excel date of the first Monday.
=if((<the last result>-4)<0,<the result of step 3>+8-<the last result>, <the result of step 3>+1-<the last result>)
6) Work out how many weeks on from that we are:
=right(<original 6 char year/week string>,2)
7) Convert 6) to a number
=value(<last result>)
8) Add that to the Monday of Week 1
=<result of step 5>+<result of step 7>*7
9) Convert this to a month number
=month(<last result>)
10) Convert this to a month name
=choose(<last result>,"Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec")
11) Add the year back on
=<lastresult>&" "&<result of 1>
This gives the month number which contains the Monday of the current numbered week.
If cell A1 contains '201547
then cell B1 should contain:
=CHOOSE(MONTH(IF((WEEKDAY(DATEVALUE("1/1/"&LEFT(A1,4)))-4)<0,DATEVALUE("1/1/"&LEFT(A1,4))+1-WEEKDAY(DATEVALUE("1/1/"&LEFT(A1,4))),DATEVALUE("1/1/"&LEFT(A1,4))+8-WEEKDAY(DATEVALUE("1/1/"&LEFT(A1,4))))+7*(VALUE(RIGHT(A1,2)))),"January","February","March","April","May","June","July","August","September","October","November","December")&" "&LEFT(A1,4)
If A1 contains 201547 then B1 returns "November 2015".
Further considerations:
You may want to make it more sophisicated, so that a month ending in a Monday does not have the week commencing with its final Monday counted as part of the month. More generally, that week only counts as part of a month is 3 or more of the days are in the month. Or 3 or more of its working days are part of the month.
As it is, I would sort with the calculated month and year as primary key and the YYYYMM code as secondary key.
Related
I have a movies table in the database with separate columns (integers) for month and year of release date.
I want to get all the movies with the release date from 09/2016 till 02/2017.
This query won't work because of other unwanted results:
Movie.where(month: [09, 10, 11, 12, 1, 2], year: [2016, 2017])
How do I query making pairs of month & year?
Storing dates in that way is a very poor design decision (if you don't also store an actual date value).
So, either do data migration and create a date column which will be trivial to query, or do a big-ass conditional query:
SELECT movies.*
FROM movies
WHERE month = 9 AND year = 2016 OR
month = 10 AND year = 2016 OR
...
month = 2 AND year = 2017
To build this query in activerecord, you can use this answer (rails-5 specific).
I'm using MySQL Workbench to run my query.
I want to run a couple of different date queries, I don't know if it's possible in SQL.
1) Run the report from the first date in the system to X date.
How can I find what the first orderID or date is as part of the calculations?
I know that if I have the date to begin with, I can use:
where T5.date_purchased BETWEEN '2005-01-01' AND '2015-12-31' OR:
where T5.date_purchased BETWEEN '2005-01-01' AND CURDATE() + INTERVAL 1 DAY
UPDATE
----- Query 2 has been answered (although open to any improvement) -----
2) Run the report from the first time a product shows up, to + 6 months (to see it's first 6 months of order) ie: Widget 1 (first order) + 6 months from order date. Something like:
where widget=widgetID AND date between widget1's first purchase and +6 months
Update: This doesn't work, however this is somewhat of what I was thinking:
where (T3.products_id = 39) and DATE_ADD((T1.products_date_added), INTERVAL 1 MONTH)
I would use my P3.products_date_added, however, I don't know how to use it as part of the above, correctly.
Are either of these possible, I know how to pull the records when I know the date, I just don't know if it can be done with 'date unknown' or if I have to run a "pre-report" first. Or is it a post processing filter in excel?
Thank you in advance.
Answering problem #2:
2) Run the report from the first time a product shows up, to + 6 months (to see it's first 6 months of order) ie: Widget 1 (first order) + 6 months from order date. Something like:
where widget=widgetID AND date between widget1's first purchase and +6 months
Answer:
-- Use for specific comparisons of products OR for the first X months of sales
where (T3.products_id = 39) and T5.date_purchased between T1.products_date_added and DATE_ADD((T1.products_date_added), INTERVAL 2 MONTH)
-- This results in PID: release: 10th July, +2 months; 10th Sep. 31 units.
OR
-- (T3.products_id = 11 or T3.products_id = 39) gives the results of the 2 product orders from release date to the first 2 months of each
-- (T3.products_id) gives all products, their first 2 months of release
-- (T3.products_id = 39) gives specific product release sales
-- Inspired by: http://stackoverflow.com/questions/28788691/mysql-range-between-dates-first-order-6-months?noredirect=1#comment45853546_28788691
If it is possible for your solution, try and run 2 separate queries:
- ONE for finding the date of first order
- then calculate +6 months from query 1 result
- AND the second one to get the purchases BETWEEN both dates
I was able to select last 150 days from database when having column 'year' as follow:
data1 = dbGetQuery(conn_data, statement=paste("SELECT *, STR_TO_DATE(CONCAT(yyyy,'-',mm,'-',dd),'%Y-%m-%d') as dt FROM stations_daily_data", "WHERE STR_TO_DATE(CONCAT(yyyy,'-',mm,'-',dd),'%Y-%m-%d') >= DATE_SUB(CURDATE(), INTERVAL 150 DAY)"))
But now all data were averaged to date and thus only have columns 'month' and 'day' (no column 'year'), and I was stuck in how to select last 150 days this time. Here is the simplified example of data frame with original one of 17 million rows:
df <- data.frame(ID=c(1:5,50001:50005),mm=c(rep(1,5),rep(12,5)),dd=c(1:5,27:31),value=c(21:30))
Feb 29 can be ignored since 150 days is a significant amount of time period.
I tried add column 'year' so that I could use the code above, but it would be wrong if say, current date is at the beginning of a year, also make changes to a big table in R would run out of R memory, I'm not familiar with database query, is it possible that I can do this by just using query instead of read the table into R and then make changes in the data frame in R, any suggestion would be appreciated!
EDIT:
The column 'year' is no longer needed since its all been averaged to date, which means now May 5th would be the average of 60 years of May 5th of each year. Next I would like to select last 150 days(averaged), the reason I tried to add column 'year' was simply try to make it easier to select.
Since I need to run the data every day, so if the day is after the month of June it would be easy just to use the current year, but if it's the month of February, then it would be current year-1, this could be done if the data is much smaller, now if I make change to the data frame, the R would pop out error of 'out of memory', that's why I was wondering if there is a way to select in database query or functions in R that wouldn't cost much memory, thanks!
You could write a function to calculate year based on a reference year plus an adjustment based on a cut off month. Then you could use the order function to order the data.frame based on calculated year, month, and day, without inserting the new calculated year field into the data.frame.
Won't have a great performance on 17 million row dataset though, since you are still ordering every row.
# some dummy data (not worrying about illegal dates like Feb 31)
set.seed(123)
da <- data.frame(mm=sample(1:12, 20, replace=T),
dd=sample(1:31, 20, replace=T))
# function to calculate year from reference year and cut off month
calc_year <- function(mm_vec, ref_year, cut_month) {
ref_year + ifelse(mm_vec >= cut_month, 0, -1)
}
# order the data.frame by year, month, and day
# (taking 2014 as ref. year & assuming months before June are from prior year
da[with(da, order(calc_year(mm_vec=mm, ref_year=2014, cut_month=6), mm, dd)), ]
# if you want just the first 5 rows
da[with(da, order(calc_year(mm_vec=mm, ref_year=2014, cut_month=6), mm, dd)), ][1:5,]
I have some records in a query with a column of dates. I want it to only show the records that occur in March. This link
https://support.office.com/en-US/Article/Examples-of-query-criteria-3197228c-8684-4552-ac03-aba746fb29d8#bm1 shows the different types of criterion.
The one below in the table describes how to show only what occurs in a particular month:
"Contain a date that falls in a specific month (irrespective of year), such as December
DatePart("m", [SalesDate]) = 12
Returns records where the transactions took place in December of any year."
I don't know what the SalesDate means in the criteria function, and there isn't any explanation on the page.
[SalesDate] implied a Date/Time field named SalesDate. If your Date/Time field is named something else, such as invoice_date, substitute that name in the DatePart expression:
DatePart("m", [invoice_date]) = 12
For March instead of December invoices, use this:
DatePart("m", [invoice_date]) = 3
You could also use Month() instead of DatePart to get the same result:
Month([invoice_date]) = 3
Using MySQL and PHP I am building a JSON array to populate a data table.
For the purposed of my question suppose the data table has the following columns:
Year: 2010,2011,2012,2013...<br/>
Month: 1,2,3,4,5...<br/>
Value: 100, 150, 200 etc...<br/>
The table structure cannot be altered and my solution needs come into the MySQL query
The data can be viewed either monthly, quarterly or yearly. Monthly and yearly is achieved easily through grouping by year and month.
Quarterly data can be grouped by calendar quarter (Jan-Mar, Apr-Jun, Jul-Sep, Oct-Dec) by this group statement:
GROUP BY year, round((month/3)+0.3,0)
So where Jan, Feb and March might all have 100 for their value the summed result is 300, same for other months.
Now my problem comes when I want to group values by a financial quarter, for example a quarter that starts in Feb, or any other quarters.
I have a statement that works for the quarter grouping using two variables that can be accessed via the SQL query, start_year (i.e. 2014) and start_month (i.e. 2)
GROUP BY year, (((round(((((month-(start_month-1))+((year-start_year)*12))-((year-start_year)*12))/3)+0.33,0)/4)+4)-floor(((round(((((month-(start_month, '%m')-1))+((year-start_year)*12))-((year-start_year*12))/3)+0.33,0)/4)+4)))*12
which basically will assign a 0,3,6,9 value to each calendar month for the purposes of grouping.
In the financial year starting February this works fine for quarters 1-3, however breaks for the final quarter as it includes Nov and Dec 2014 data and Jan from 2015.
As a result I get 5 rows of data instead of 4.
This is because of the preceding GROUP by year clause, an important addition as we might want to generate a table that views sequential quarters for multiple years.
So what I am looking for is a way of grouping the years together by offsetting the start month.
So when the year starts in Jan it will group Jan-Dec but if we change that to starting Feb it will group Feb-Jan.
Any ideas, suggestions most welcome!
Regards,
Carl
I solved a similar problem just now (a Moodle report aggregating assignment scores by year and quarter) with something like this:
select year(from_unixtime(s.timemarked)) as year, quarter(from_unixtime(s.timemarked)) % 4 + 1 as quarter, count(distinct data1) as "tickets graded" from mdlassignment_submissions s where grade >= 0 group by year, quarter order by year, quarter;
The relevant part for what you're doing is quarter(from_unixtime(s.timemarked)) % 4 + 1 as quarter
As another commenter pointed out, MySQL has a quarter() function, but it doesn't do financial quarters. However, since (as I understand it, at least, based on consulting the relevant wikipedia page) financial quarters are just offset by 1, the % 4 + 1 at the end should convert it.