insert problem with rails3

tran =Transaction.select(“transactions.id,transactions.user_id,transactions.branch_id,transactions.customer_id,transactions.membership_type_id,transactions.bill_amount,transactions.bill_date”)

treport = TransactionReport.new

puts “from report”

tran.each do |u|

bs = TransactionReport.new(

:user_id => u.user_id,

:branch_id => u.branch_id,

:transaction_id => u.id,

:customer_id => u.customer_id,

:membership_type_id => u.membership_type_id,

:bill_amount => u.bill_amount,

:bill_date => u.bill_date

)

bs.save

Here the object creation will happen in loop so this will create new object for TransactionReport with each insertion. How to avoid this in rails3…?

Thank you

vishnu

Exactly what is it that you are trying to avoid? If you don't want it to create new objects in the loop then just remove the code.

Colin

Thanks for the reply.

Exactly what going on here is now,after the execution of one insertion some amount of data will insert, around 172 rows, but after i truncate the tables, again when i execute that operation, the row count starting from 173… i think that happens of new object creation for each transaction… Also i want to know… how to do the one time object creation for all insertion in that code…?

Thank you

vishnu

Please don't top post, it makes it difficult to follow the thread. Insert your reply at appropriate points in previous message. Thanks.

Thanks for the reply.

Exactly what going on here is now,after the execution of one insertion some amount of data will insert, around 172 rows

Is that what you expect or is that part of the problem?

, but after i truncate the tables

What do you mean by truncate the tables? Do you mean you have removed all the rows?

, again when i execute that operation, the row count starting from 173

If there are still records in the database then you have not removed them.

... i think that happens of new object creation for each transaction... Also i want to know... how to do the one time object creation for all insertion in that code...?

Sorry, I still don't understand exactly what you want. Please try and write very carefully a bit at a time exactly what happens and what you are want to happen.

Colin

Thanks for the reply.

Exactly what going on here is now,after the execution of one insertion some amount of data will insert, around 172 rows, but after i truncate the tables, again when i execute that operation, the row count starting from 173… i think that happens of new object creation for each transaction…

It has to do with the sequence that is set on the primary key. Why does that matter? The id is just a number that is sequential (that is “unique” and “monotonic” upwards, it is not even continuous … upon failed inserts, a “sequence number” will be consumed and

“lost forever”, but … why care).

If you really wanted something like “line numbers” in your report that have meaning in the real business context, than you should make them yourself as separate column

(and not abuse the database sequence for that)

Also i want to know… how to do the one time object creation for all insertion in that code…?

Please explain again what is the problem…

Also, a few remarks about naming:

  • “Transaction” is a dangerous word to use in the context of databases (I presume you intend it for FinancialTransaction but in database terms, it has a special meaning)

  • tran is a poor name for “transactions”. It is a “list” (Enumerable), so use a name that represent that (a plural form, “trans” or “input_transactions” or …).

  • The class TransactionReport really has the wrong name … What you are making inside the loop is a TransactionReportLine or TransactionReportEntry and that whole list of TransactionReportLines together, make for a TransactionReport. Such a

    TransactionReport will have separate info, such as:

    • data create (created_at)
    • by who was the report ran
    • for which context (all FinancialTransaction, or only the one’s in one currency, one branch etc).
  • and then if you have TransactionReport and TransactionReportLine, you could even do

class TransactionReport has_many :transaction_report_lines, :autosave => true end

tr = TransactionReport.new(:creator => “Shawn”, :period => “Jan 2012”)

input_transactions.each do |input_transaction|

this will NOT save to the db

tr.transaction_report_lines.build(input_transaction.attributes.slice(:user_id, …)) # not tested end

tr.save! # this will save all at once, or nothing, if that was your intention

Code not tested, but indicative of a different style.

HTH,

Peter

Colin simply i am saying here the truncate means, remove all the rows. so after the truncate, it should start from id 1. But now the default id start from the next to old one( that means 173,174.........).

Yes, that is the way it works. If you want anything different I think you will have to put a column in the database (sequence for example) and manage it yourself. Though I think it may be possible to override the numbering system but I would not recommend it.

Did you not understand my request about top posting?

Colin