Skip to content
  • Categories
  • Recent
  • Tags
  • Popular
  • World
  • Users
  • Groups
Skins
  • Light
  • Cerulean
  • Cosmo
  • Flatly
  • Journal
  • Litera
  • Lumen
  • Lux
  • Materia
  • Minty
  • Morph
  • Pulse
  • Sandstone
  • Simplex
  • Sketchy
  • Spacelab
  • United
  • Yeti
  • Zephyr
  • Dark
  • Cyborg
  • Darkly
  • Quartz
  • Slate
  • Solar
  • Superhero
  • Vapor

  • Default (No Skin)
  • No Skin
Collapse
Code Project
  1. Home
  2. General Programming
  3. Design and Architecture
  4. Design a system to Transform source Files to target file.

Design a system to Transform source Files to target file.

Scheduled Pinned Locked Moved Design and Architecture
designquestionperformancecareer
3 Posts 3 Posters 0 Views 1 Watching
  • Oldest to Newest
  • Newest to Oldest
  • Most Votes
Reply
  • Reply as topic
Log in to reply
This topic has been deleted. Only users with topic management privileges can see it.
  • L Offline
    L Offline
    lucky akky
    wrote on last edited by
    #1

    I need to design a system which will process hundreds of source files (different format) and convert to one target files. There should be two interfaces 1. Command line 2. web user interface Command line interface is to run the transform job though batch and web user interface is to define the format of source file and mapping details of source file to target file i.e. one time job for each source file. All the source files are fixed width or delimited files. What is the correct approach? Should I create one stage table for each source file to the stage data? Should I create one stage table on runtime when user will define of layout (Most are fixed width mainframe files) of source file? Should I create only one generic table, containing around 100 columns all varchar type? I am looking for the best approach to design the system. Performance is very critical for this app. There are hundreds of files and we need to transform all the files daily within certain time. Thanks in Advance Akshay

    Lucky akky keep smiling

    M P 2 Replies Last reply
    0
    • L lucky akky

      I need to design a system which will process hundreds of source files (different format) and convert to one target files. There should be two interfaces 1. Command line 2. web user interface Command line interface is to run the transform job though batch and web user interface is to define the format of source file and mapping details of source file to target file i.e. one time job for each source file. All the source files are fixed width or delimited files. What is the correct approach? Should I create one stage table for each source file to the stage data? Should I create one stage table on runtime when user will define of layout (Most are fixed width mainframe files) of source file? Should I create only one generic table, containing around 100 columns all varchar type? I am looking for the best approach to design the system. Performance is very critical for this app. There are hundreds of files and we need to transform all the files daily within certain time. Thanks in Advance Akshay

      Lucky akky keep smiling

      M Offline
      M Offline
      Mycroft Holmes
      wrote on last edited by
      #2

      I've built many of these and not once have I used a command line interface. I use a service to do timed and repeated processing. Generally I use a separate staging file/database for each source file/set. I find you can usually group the files into sets which have the same data structure. I also use stored procs to do the processing from the staging tables to the final data table, this may be frowned upon as it is less flexible than a full ETL tool but I find it suits my style. I ended up with a winforms app that allows the user to configure a file for loading, defining the title and data rows, the delimiter or column widths and create a staging table with varchar fields so I can either BCP or bulk copy into the staging table. I then assign 1 of about 8 procedures to process to the final data table. BCP in 2005/8 is more fragile than 2000 so I use bulk copy a lot, slower but more robust.

      1 Reply Last reply
      0
      • L lucky akky

        I need to design a system which will process hundreds of source files (different format) and convert to one target files. There should be two interfaces 1. Command line 2. web user interface Command line interface is to run the transform job though batch and web user interface is to define the format of source file and mapping details of source file to target file i.e. one time job for each source file. All the source files are fixed width or delimited files. What is the correct approach? Should I create one stage table for each source file to the stage data? Should I create one stage table on runtime when user will define of layout (Most are fixed width mainframe files) of source file? Should I create only one generic table, containing around 100 columns all varchar type? I am looking for the best approach to design the system. Performance is very critical for this app. There are hundreds of files and we need to transform all the files daily within certain time. Thanks in Advance Akshay

        Lucky akky keep smiling

        P Offline
        P Offline
        puri keemti
        wrote on last edited by
        #3

        Create run time stage table depending upon the format of targetting file.....

        1 Reply Last reply
        0
        Reply
        • Reply as topic
        Log in to reply
        • Oldest to Newest
        • Newest to Oldest
        • Most Votes


        • Login

        • Don't have an account? Register

        • Login or register to search.
        • First post
          Last post
        0
        • Categories
        • Recent
        • Tags
        • Popular
        • World
        • Users
        • Groups