Split commas delimited into 1D complex array

I have comma delimited string in this format
[0.866405665874481,0.127425819635391,0.983153462409973,-0.0711551383137703,0.955700755119324,-0.27884304523468,0.787182509899139,-0.501964211463928,0.512355387210846,-0.687117278575897,0.177220240235329,-0.777789652347565,-0.0876994803547859,-0.792799115180969,-0.324512422084808,-0.736825287342072,-0.517065346240997,-0.627932071685791,-0.666604995727539,-0.471743047237396,-0.767350137233734,-0.287838608026505,-0.819185018539429,-0.0853987112641335,-0.81872695684433,0.123056441545486,-0.761552691459656]
The data contains pairs of complex numbers. The first 2 numbers is one pair, next 2 numbers one pair, next pairs.....
How can I extract the pairs of number and make it into a 1D complex array?
This is the expected output:
(0.866405665874481,0.127425819635391),(0.983153462409973,-0.0711551383137703),(0.955700755119324,-0.27884304523468),(0.787182509899139,-0.501964211463928),(0.512355387210846,-0.687117278575897),(0.177220240235329,-0.777789652347565),(-0.0876994803547859,-0.792799115180969),(-0.324512422084808,-0.736825287342072),(-0.517065346240997,-0.627932071685791),(-0.666604995727539,-0.471743047237396),(-0.767350137233734,-0.287838608026505),(-0.819185018539429,-0.0853987112641335),(-0.81872695684433),(0.123056441545486,-0.761552691459656)
Any gurus pls advise
Thanks and great day
Solved!
Go to Solution.

I used the Spreadsheet String to Array, Decimate Array, and then Re/Im to Complex.
There are only two ways to tell somebody thanks: Kudos and Marked Solutions
Unofficial Forum Rules and Guidelines
Attachments:
Delimited String to Complex Number Array.png ‏13 KB

Similar Messages

Split Comma Delimited String Oracle

I want to Split Comma Delimited string in oracle9i into rowwise data.i do not want to use Functions or connect by level...
is there any other way from which we can split strings.???
if we use Connect by level method the problem im facing is we cannot use suqueries in Connect by clause...
Edited by: user11908943 on Sep 16, 2009 8:37 AM

michaels2 wrote:
I prefer using XMLTABLE which has superceded XMLSEQUENCE and I personally find it easier to read...Agree, but that would be hard in 9i since it is not implemented ;)Ah! missed that bit (gawd I wish people would upgrade to supported versions)
In still later versions (11g) I prefer
SQL> with t as
select 'a,b,c,d,e' str from dual
select trim(column_value) str
from t, xmltable(('"' || replace(str, ',', '","') || '"'))
STR
a
b
c
d
e
5 rows selected.
Natty! But completely illogical (I mean creating an XMLTABLE from a non-XML string). I'm guessing that's a development of XQuery?

How to split comma delimited string?

Hi,
I have a parameter called Id in my SP which will be of nvarchar data type and i'm going to get the multiple ids at a time seperated by commas in that parameter from the application. Now my requirement is to update a table with these comma seperated ids in seperate rows.
For example, if i have 2 parameters called Id1 and Id2. Id1 will contain only one value and will be of int data type and Id2 will be of nvarchar data type as i can get multiple ids delimited by a comma from the application. Suppose Id1 = 1 and Id2 = '1,2,3,4'. Then I have to update id2 in the tables seperately like wherever Id1 is '1' i need to update Id2 column for 4 rows with the value 1, 2, 3, 4 respectively in different rows.
Could you please tell me how can i do this in T-SQL? How can i split the data of parameter Id2 in 4 different rows?
Please let me know if you did not understand the question.
Thanks,
Deepti

deeptideshpande wrote:
Hi,
I got a similar function which is like this:
CREATE FUNCTION dbo.fnSplit(
@sInputList VARCHAR(8000) -- List of delimited items
, @sDelimiter VARCHAR(8000) = ',' -- delimiter that separates items
) RETURNS @List TABLE (item VARCHAR(8000))
BEGIN
DECLARE @sItem VARCHAR(8000)
WHILE CHARINDEX(@sDelimiter,@sInputList,0) <> 0
BEGIN
SELECT
@sItem=RTRIM(LTRIM(SUBSTRING(@sInputList,1,CHARINDEX(@sDelimiter,@sInputList,0)-1))),
@sInputList=RTRIM(LTRIM(SUBSTRING(@sInputList,CHARINDEX(@sDelimiter,@sInputList,0)+LEN(@sDelimiter),LEN(@sInputList))))
IF LEN(@sItem) > 0
INSERT INTO @List SELECT @sItem
END
IF LEN(@sInputList) > 0
INSERT INTO @List SELECT @sInputList -- Put the last item in
RETURN
END
GO
Now when i use this in my select statement, i get data in a tabular structure as shown below:
select
from dbo.fnSplit('1,22,333,444,,5555,666',
Results
Item
1
22
333
444
5555
666
Now my requirement is to insert this value in a table which has 2 columns ID1 and ID2.
At any point of time ID1 will be an int value(only one id in that parameter) and ID2 can have multiple ids delimited by comma. With the help of above split function i can split values of ID2 into 6 different ids but i need to insert the table
in the following way:
ID1     ID2
1        1
1        22
1        333
1        444
1        5555
1        666
How can i insert the data in the above mentioned fashion.
Thanks,
Deepti
Like this:
Code Snippet
select 1 as id,* from dbo.fnSplit('1,22,333,444,,5555,666', ',')
Great!!!! Wonderfull!!!... Thanks a lot.

How do I split a comma delimited string into an array.

I have a string that I am passing into a function that is Comma delimited. I want to split the string into an array. Is there a way to do this.
Thanks in advance.

trouble confirmed on 10gR1 and 10gR2
works with 'a,b,c' and also with ' "1" , "2" , "3" '
does not work with '1,2,3'
throwing ORA-6512

How to split a delimited string into an array...

I have a delimited string coming in to a stored procedure (SP) for vb.net that needs to be parsed/converted into multi-dimensional arrays. The reason I am using delimited string because I do not know how to pass arrays to oracle SP from vb.net using ODP or MPO (Microsoft's Provider for Oracle).
My delimited string looks something like this: "level1_1|level2_1:level3_1;level3_2|level2_2:level3_1;level3_2~Level1_2|level2_2:level3_1"
So, its going to be 3d array. I would greatly appreciate if anyone could help me solving this puzzle. Any examples would be a great help.
Thanks for everything in advance,
Vincent

Hi Andrew,
I appreciate your response. Currently I use "Microsoft provider for Oracle" to connect to Oracle Database because when I started with my application there wasn't ODP (Oracle Data Provider for .Net) available. So, I used Microsoft provider and still using it.
I know how to use arrays in .Net but the problem is I can not find the Oracle Type to pass the array to Oralce Stored Procedure using this provider.
It would be very helpful if someone shows me the syntax for both to receive from oracle and transport to Oracle. I haven't tried using ODP yet. May be now its the time to see how it works.
And of course, your link was very helpful but I am already doing that (which is very slow for multi-dimentional array). So, I thought there might be a built in function in oracle which does that. Because most of the programming languages does have built in function for this conversion. for example: Split() in vb, java, javascript etc.
Anyway, if anyone finds any clues as to how to pass milti-dimentional arrays from .Net (microsoft) to oracle stored procedure please, please let me know how to do that.
-Vincent

Help with formatting multiline output into comma delimited ordered output

I have 2 tables:
SQL> desc table_1;
Name Null? Type
===================================
ID NOT NULL NUMBER
DATA1 NOT NULL VARCHAR2(440)
DATA2 NOT NULL VARCHAR2(1024)
SQL> desc table_2;
Name Null? Type
===================================
ID NOT NULL NUMBER
ATTNAME NOT NULL VARCHAR2(255)
ATTVAL                          VARCHAR2 (4000)
DATA1 NOT NULL CHAR(1)
DATA2                          VARCHAR2 (2000)
DATA3                          VARCHAR2 (255)
I need to get ATTVAL from where e.g. ATTNAME=att_name1 to ATTNAME=att_name6 from every entry (with its unique ID), and format the output into comma delimited format in e.g. this order:
att_val1,att_val3,att_val6,att_val4,att_val5,att_val6
So e.g. for entry with ID "9812" from the query below, the output I need would be:
187,179,156,134,1436,1809
What I've got so far is as follows:
SQL> SELECT id,attname,attval FROM table_2 WHERE id in (SELECT id from table_1 WHERE data2='xxx')
AND attname in ('att_name1','att_name3','att_name6','att_name4','att_name5','att_name6');
ID               ATTNAME               ATTVAL
===============================
1970 att_name1 123
1970 att_name2 abc
1970 att_name3 1234
1970 att_name4 def
1970 att_name5 1134
1970 att_name6 ghj
9812 att_name4 134
9812 att_name5 1436
9812 att_name3 156
9812 att_name1 187
9812 att_name2 179
9812 att_name6 1809
77 att_name1 1980
77 att_name5 1867
77 att_name3 174
77 att_name4 1345
77 att_name2 1345
77 att_name6 1345
but I don't know how to format the output comma limited in certain order. (or if this is the best way of getting the data out)
Would someone have idea how this could be done?

846954 wrote:
Thanks Frank!
I got much further now :).
I've got Oracle 10g So I used the "SYS_CONNECT_BY_PATH" for my query.
Now I get the output in the format I want, however, it comes out in the order that the attributes are (which is pretty random).The values you posted are in order: 'attval1' < 'attval2' < ...
So I'm using this now(had to use "|" as separator because SYS_CONNECT_BY_PATH would not work with comma due to some of the attval having comma in them ):The values you posted don't contain and commas.
You're hiding what the problem is. It would really help if you posted actual data. It always helps if you post CREATE TABLE and INSERT statements for a little sample data, and the results you want from that data.
Assuming you really have something that has to be in a certain order, and that order is not based on anything in the values themselves, then DECODE might be a good way to do it. Storing the sort value in a table might be even better.
It looks like you were using an Oracle 9 exanple. In Oracle 10, using SYS_CONNECT_BY_PATH is simpler:
SELECT     id
,     LTRIM ( SYS_CONNECT_BY_PATH (attval, '|')
           , '|'
           )          AS attvals
FROM     (
          SELECT     id
          ,     attval
          ,     ROW_NUMBER () OVER ( PARTITION BY id
                              ORDER BY         DECODE (...)
                            )     AS curr
          WHERE     id     IN (
                         SELECT id
                         FROM     table_1
                         WHERE     data2     = 'xxx'
          AND     attname     IN ('attname1','attname2','attname3','attname4','attname5','attname6')
WHERE     CONNECT_BY_ISLEAF     = 1
START WITH     curr     = 1
CONNECT BY     curr     = PRIOR curr + 1
     AND     id     = PRIOR id
;You don't need two almost-identical ROW_NUMBERs; one will do just fine.

Is it possible to convert a table of values into a comma-delimited-list?

Hi,
I'd like to turn the following dataset:
Parent | Child
Charles | William
Charles | Harry
Anne | Peter
Anne | Zara
Andrew | Beatrice
Andrew | Eugenie
into this:
Parent | Children
Charles | Diana,Camilla
Anne | Peter,Zara
Andrew | Beatrice,Eugenie
In other words, I'd like to take a list of values pertaining to some key and produce them as a comma-delimited-list.
I know its is possible in T-SQL although the method is a bit of a nasty hack. Is it possible in PL-SQL?
Thaks in advance
Jamie

Hi,
With model clause (10g)
with t as(
select 'Charles' parent, 'William' child from dual union
select 'Charles', 'Harry' from dual union
select 'Anne', 'Peter' from dual union
select 'Anne', 'Zara' from dual union
select 'Andrew', 'Beatrice' from dual union
select 'Andrew', 'Eugenie' from dual
select parent,substr(res,2) res
from t
model
return updated rows
partition by ( parent)
dimension by ( row_number()over(partition by parent order by child) rn)
measures(child, cast( null as varchar2(4000)) as res)
rules
iterate(100000)
until(presentv(res[iteration_number+2],1,0)=0)
( res[0]=res[0]||','||child[iteration_number+1]);
PARENT RES
Anne    Peter,Zara
Andrew Beatrice,Eugenie
Charles Harry,William
3 rows selected.

SPLIT data-record AT delimiter into table-fieldname

Hi all,
I have the following SPLIT statement. But unable to split into respective field in an internal table. I'm using the standard delimeter.
Please help me how to create a file using the delimeter w_delimeter.
When I create the file with separated by '#' even it is not spliting.
w_delimeter TYPE c value cl_abap_char_utilities=>horizontal_tab.
SPLIT data-record AT w_delimeter INTO i_0208-pernr
                                                                     i_0208-begda
                                                                    i_0208-endda
                                                                    i_0208-wtart
                                                                    i_p0208-allpc.
Thanks,
Prasad

Hi Prasad,
it should work try this bellow piece of code.
REPORT ZTEST_A.
data: Begin of itab occurs 1,
        name(30),
        street(30),
        Apt(3),
        city(30),
        state(2),
        zipcode(5),
      end of itab,
      text1(250),
      text2(250).
Text1 = 'Jacson 3 xyz dr B4 abcd DE 12345'.
Text2 = 'Edward 3 Caaa dr B4 pqr DE 54623'.
split text1 AT ' ' into   itab-name itab-street itab-Apt
        itab-city itab-state itab-zipcode.
append itab.
split text2 AT ' ' into   itab-name itab-street itab-Apt
        itab-city itab-state itab-zipcode.
append itab.
thanks,
Venkat

Split String at delimiter into table.

Hi Experts,
I have a requirement wherein i need to split aaaa#bbbbbbbbbbbb#ccc#dddddddddddddddddddddd#e# into internal table as below:
Header 1
Header 2
1
aaaa
2
bbbbbbbbbbbb
3
ccc
4
dddddddddddddddddddddd
5
e
how can this be done.
Thanks & Regards,
Anil

Hi Ashok,
split <string> at '#' into table it. -Doesn't work
split <string> at SPACE into table it. - Works
But same statement with '#' is not working.
Feeling strange.
Thanks,
Anil

Crystal report - how to split a field into more fields

Hello,
I`m new to Crystal reports and I`ve got a trouble. I have field which contains an address - street, city, zip code. The example is:
STEHLIKOVA 977 165 00 PRAHA 620 - SUCHDOL 165 00.
What I need to achieve is to split this string into three separated fields. I`ve trouhg a couple of forums but haven`t been able to find a proper answer. The problem is that the addresses differ so I can`t use an absolute defining of a start position. Looking at the DB (HEXA code) the parts in the string are divided by two dots:
STEHLIKOVA 977..165 00 PRAHA 620 - SUCHDOL..165 00
I`ve been able to work out this solution:
stringVar array x := split({cparty.STREET_ADD},"..");
Local numberVar i;
Local stringVar outputString := "";
For i:=1 to Count(x) do
outputString := outputString + x[i] + Chr(10)
outputString;
It splits the string into three rows:
STEHLIKOVA 977
165 00 PRAHA 620 - SUCHDOL
165 00
And I don`t know how to find the end of each row so to be able to separate the strings and report them as three different fields.
Would be anyone so kind and help me out with this?
Thank you.
Petr

Hi
Actually using a for loop is not necessary here. All you need to do is to add several formula fields: one for street, one for city and one for zipcode. In @street field you add formula:
stringVar array x := split({cparty.STREET_ADD},Chr(13));
x[1];
Then you drag such formula field to details section of your report and watch preview to check if everything looks alright. Repeat for every formula field that you'd like see in your report.
In @city you add almost identical formula but you change index, i. e. instead of x[1] you need to use x[2]. Then for @zip x[3].
You may need to check if your address has all three parts - for example if you want to use formula in the second part of your address field you may need check first if there are at least two parts after split:
stringVar array := split({cparty.STREET_ADD},Chr(13));
numbervar c;
c := count(x);
if 2 <= c then
x[c];
Var 'c' is used to store the number of elements in array after split. Then I'd like to check if the part (second) actually exists. So I try to check if number of part that I want to refer to is not bigger than the number of elements in array after split (here stored in var 'c').
Actually the code presented above is not enough since you have no guarantee that you'll always get address structured in the very same way. For example in demo PL database I have zip code and city in the same row after split with Chr(13) as delimiter.
You'll need to experiment or ask someone to prepare correctly structured data coming from B1 to your report.
Kind regards,
Radek

SSIS importing comma delimited with double quote text qualifier - Works in VS - SQL Job ignores text qualifier and fails (truncation)

I've created an SSIS package to import a comma delimited file (csv) with double quotes for a text qualifier ("). Some of the fields contain the delimiter inside the qualified text. An example row is:
15,"Doe, John",IS2,Alabama
In SSIS I've specified the text qualifier as ". The sample output in the connection manager looks great. The package runs perfectly from VS and when manually executed on the SSIS server itself. The problem comes when I schedule the package to run via SQL
job. At this point the package ignores the text qualifier, and in doing so pushes half of a field into the next available column. But instead of having too many columns, it concatenates the last 2 columns ignoring the delimiter. For example (pipes are fields):
15|"Doe| John"|IS2,Alabama
So the failure happens when the last half of a field is too large to fit into the next available field. In the case above _John" is 6 characters where the IS2 field is char(3). This would cause a truncation failure, which is the error I receive from the
job history.
To further test this I created a failure flow in my data flow to capture the records failing to be pulled from the source csv. Out of ~5000 records, ~1200 are failing, and every one of the failures has a comma delimiter inside the quoted text with a 'split'
length greater than the next ordinal field. The ones without the comma were inserted as normal and records where the split fit into the next ordinal column where also imported, with the last field being concatenated w/delimiter. Example records when selected
from table:
25|"Allan Percone Trucking"|GI6|California --Imported as intended
36|"Renolds| J."|UI6,Colorado --Field position offset by 1 to right - Last field overloaded
To further ensure this is the problem, I changed the csv file and flat file connection manager to pipe delimited, and sure enough every record makes it in perfectly from the SQL job execution.
I've tried comma delimited on the following set ups. Each set up failed.
SSIS Server 2008 R2 RTM
DB Server 2008 RTM
DB Compat 80
SSIS Server 2008 R2 RTM
DB Server 2008 R2 RTM
DB Compat 80
SSIS Server 2008 R2 RTM
DB Server 2008 RTM
DB Compat 100
SSIS Server 2008 R2 RTM
DB Server 2008 R2 RTM
DB Compat 100
Since a lot of our data comes in via comma delimited flat files, I really need a fix for this. If not I'll have to rebuild all my files when I import them to use pipe delimiters instaed of commas. I'd like to avoid the extra processing overhead if possible.
Also, is this a known bug? If so can someone point me to the posting of said bug?
Edit: I can't ask the vendor to change the format to pipe delimited because it comes from a federal government site.

Just wanted to share my experience of this for anyone else since I wasted a morning on it today.
I encountered the same problem where I could run the package fine on my machine but when I deployed to a server and ran the package via dtexec, the " delimeter was being replaced with _x0022_ and columns all slurped up together and overflowing columns/truncating
data etc.
Since I didn't want to manually hack the DTS XML and can't install updates, the solution I used was to set an expression on the textdelimiter field of the flat file connection with the value "\"" (a double quote). That way, even if the one stored in XML
gets bodged somewhere along the way, it is overridden with the correct value at run time. The package works fine everywhere now.

Comma delimited file in application server.

How to extarct a comma delimited file from an internal table in to SAP application server.

Hi Arun,
Can you be a bit more clearer? Are you uploading the dat to the appl server or downloading?
1) Comma separated info from itab to file server.
loop at itab.
concatenate itab-field1
            itab-field2
             itab-field3
into itab_new-data separated by ','.
append itab_new.
clear itab_new.
endloop.
open dataset dsn for output in text mode.
if sy-subrc = 0.
loop at itab_new.
transfer itab_new to dsn.
endloop.
endif.
close dataset.
2) Comma separated info from file to itab.
open dataset dsn for input in text mode.
if sy-subrc = 0.
do.
read datset dsn into itab-data.
if sy-subrc <> 0.
exit.
else.
append itab.
clear itab.
endif.
enddo.
endif.
close dataset.
loop at itab.
split itab-data at ',' into itab_new-field1
            itab_new-field2
             itab_new-field3.
append itab_new.
clear itab_new.
endloop.
REgards,
Ravi

Passing data to different internal tables with different columns from a comma delimited file

Hi,
I have a program wherein we upload a comma delimited file and based on the region( we have drop down in the selection screen to pick the region). Based on the region, the data from the file is passed to internal table. For region A, we have 10 columns and for region B we have 9 columns.
There is a split statement (split at comma) used to break the data into different columns.
I need to add hard error messages if the no. of columns in the uploaded file are incorrect. For example, if the uploaded file is of type region A, then the uploaded file should be split into 10 columns. If the file contains lesser or more columns thenan error message should be added. Similar is the case with region B.
I do not want to remove the existing split statement(existing code). Is there a way I can exactly pass the data into the internal table accurately? I have gone through some posts where in they have made use of the method cl_alv_table_create=>create_dynamic_table by passing the field catalog. But I cannot use this as I have two different internal tables to be populated based on the region. Appreciate help on this.
Thanks,
Pavan

Hi Abhishek,
I have no issues with the rows. I have a file with format like a1,b1,c1,d1,e1, the file should be uploaded and split at comma. So far its fine. After this, if the file is related to region A say Asia, then it should have 5 fields( as an example). So, all the 5 values a1,b1..e1 will be passed to 5 fields of itab1.
I also have region B( say Europe) whose file will have only 4 fields. So, file is of the form a2,b2,c2,d2. Again data is split at comma and passed to itab2.
If some one loads file related to Asia and the file has only 4 fields then the data would be incorrect. Similar is the case when someone tries to load Europe file with 5 fields related data. To avoid this, I want to validate the data uploaded. For this, I want to count the no. of fields (seperated by comma). If no. of fields is 5 then the file is related to Asia or if no. of fields is 4 then it is Europe file.
Well, the no. of commas is nothing but no. of fields - 1. If the file is of the form a1,b1..e1 then I can say like if no. of commas = 4 then it is File Asia.But I am not sure how to write a code for this.Please advise.
Thanks,
Pavan

Wrapping A SqlBulkCopy Class To Export A Proper Comma Delimited Table

Hello!
If anyone can point me in the right direction it would be greatly appreciated!
In my search, it seems SQL Server cannot export tables to a proper comma delimited format. I have not found a out of the box solution for SQL Server to properly escape commas and double quotes within fields on a record.
Which brings me here. I'm searching for a way to leverage the SqlBulkCopy class (or a alternative) to quickly export a record's values in a array or list so I can feed it into my custom DelimitedStringBuilder class.
So far it seems the out of the box SqlBulkCopy class only supports import into other tables and not a text file.
One thought (I have no idea how I would implement it) would be to attach a DataReader class to the SqlBulkCopy class and then pass the values from the DataReader to my DelimitedStringBuilder class.
Has anyone been able to write something like this? Or is there a simpler solution?

Have you tried just using SQL and ExecuteNonQuery? Below is an example:
Dim TextConnection As New System.Data.OleDb.OleDbConnection("Provider=Microsoft.Jet.OLEDB.4.0;" & _
"Data Source=" & "C:\Documents and Settings\...\My Documents\My Database\Text" & ";" & _
"Extended Properties=""Text;HDR=NO;""")
TextConnection.Open()
'New table
Dim TextCommand As New System.Data.OleDb.OleDbCommand("SELECT * INTO [Orders#csv] FROM [Orders] IN '' [ODBC;Driver={SQL Server Native Client 11.0};Server=.\SQLExpress;Database=Northwind;Trusted_Connection=yes];", TextConnection)
TextCommand.ExecuteNonQuery()
TextConnection.Close()
Paul ~~~~ Microsoft MVP (Visual Basic)

Reading comma delimited files

Hi,
What are the FMs that I can use to read comma delimited files. Also, will it be able to run in the background?
Thanks,
RT

Hi Rob
   As far as i know, we can not upload data from
persentation server in background. For that the file
needs to be placed in application server and use Open dataset command.
Below is just an example to help you have a feel of the
same.
eg:
type: begin of t_data,
        vbeln like vbak-vblen,
        posnr like vbap-posnr,
        matnr like vbap-matnr,
        menge like vbap-menge,
      end of t_data.
data: it_data type standard table of t_data.
data: wa_data type t_data.
data: l_content(100) type c.
open dataset p_file for output in text mode.
if sy-subrc ne 0.
*** error reading file.
else.
    do.
      read dataset p_file into l_content.
      if sy-subrc ne 0.
         close dataset p_file.
         exit.
      else.
         split l_content at ',' into wa_data-vbeln
           wa_data-posnr wa_data-matnr wa_data-menge.
         append wa_data to it_data.
      endif
    enddo.
endif.
Hope this helps you.
Kind Regards
Eswar

Split commas delimited into 1D complex array

Similar Messages

Maybe you are looking for