The schema of the DB input component does not automatically change to match the query you use; you must change it manually.  When you do, it will ask to propagate to the following components.

I also note an error symbol against your tHashInput.  This possibly indicates you have not set up a schema for it.

Ok, it was more simple as I thought...perfect!
Thanks a lot alevy

Please, can you paste in the topic a simple example made in TOS about first steps above?
To count users, have I to use a tMap, or have i to change table schema query?

Thanks all, sorry but I'm moving first experiences with TOS and not all my doubts are explaned in the Help or I'm not able to find them in the forum.

Go to File > Edit Project properties > Designer > Palette Settings and move the Technical folder from Hide to Show.

alevy wrote:

So if I understand correctly, you want to compare the users in the source to the count of each user in the destination, summarise the count comparisons to a log file and write 1-1 matches from source to destination.

t[DB]Input (count by user) --> tHashOutput
|
OnSubjobOK
|
Source input   --main-->                                      --> log output
|                                     tMap (left-join on user)
| tHashInput --lookup-->                                      --> destination output (filter 1-1)
|
OnSubjobOK
|
| Source input --lookup-->
|                                      tMap (left-join on user) --> append log output (filter 0-N)
tHashInput       --main-->
|
OnSubjobOK
|
tSendMail

Yes Alevy, you understood very well! Thanks a lot for your answer!

I'll try to implement job and if I'll have some doubts I will ask againg. Thanks!

Sorry Jans...but yours it's not an answer to my question, please open another topic.

Thanks

Good morning to all,
I have a question about an operation I wanna do with Talend; the situation is the following:

I use Talend to take some data from origin (usually sql db, txt, csv, ...) and I put them always into SQL db. (this is a simple concept...)
I have, in origin and destination the same "User" record, I mean that I have a users table in the origin and I import them into the destination; in the origin file, the user has a field where there is a daily sum of credits (divided by type), instead into the destination, I can have more records for every type of credits;

e.g.
ORIGIN:            USER1 - type 1 - 10 credits (3+3+4)
DESTINATION:   USER1 - type 1 (ID = 1) - 3 credits
                        USER1 - type 1 (ID = 2) - 3 credits
                        USER1 - type 1 (ID = 3) - 4 credits

Now I want, before import data, check:
   1. count and trace users that have 1 (origin) to 1 (destination) match - in this case I can import directly data
   2. count and trace users that have 1 to N (> 1) match - I don't import data
   3. count and trace users that have 1 to 0 match - I don't import data
   4. count and trace users that have 0 to N (=> 1) match - I don't import data

Then I have 2 ways to follow:
   1. I want count recors per each case and put these in a .txt log file, save log file in a folder and also send it by e-mail
       e.g. of .txt structure:
       "There are XX 1-1 match records;
        There are YY 1-N match records (users' list);
        There are ZZ 1-0 match records (users' list);
        There are WW 0-N match records (users' list)"

   2. Import only data where I have 1-1 match records (but in this case i think that could be simple doing in the tMap a count before insert: in destination field I put for example IF COUNT(...) = 1 "USER1 - 10 credits...)

Thanks a lot for help!

P.S. Sorry if my english is very bad...if you need more specs please ask me. If you need, I could also attach a screenshot of my "standard" job.