|
JCapper Message Board JCapper 101
|
| Home |
Register
Log In |
| By | Building JCapper Databases - simplified | |||
| jeff 5/26/2011 5:20:31 PM |
Data Folder Structure Data folder structure is something each user gets to decide on. I personally have a folder for each year with a folder for each quarter under that. My data folder structure for 2011 looks something like this: C:\2011 C:\2011\Q1_2011 C:\2011\Q2_2011 C:\2011\Q3_2011 C:\2011\Q4_2011 All files for the months of Jan, Feb, and Mar 2011 go in my c:\2011\Q1_2011 folder. Files for the months of Apr, May, and June 2011 go in my c:\2011\Q2_2011 folder, etc. At the end of each quarter, I make the switch to the next quarterly folder in the DFM. For the first 2-3 days of each new quarter special care is taken to look at the current folder and manually copy any files from the prior quarter I may have accidentally downloaded to the wrong folder. Other than that, the above data folder structure requires no special maintenance on my part. Other users are using data folder structures that are quite different from mine. Some use a simple yearly folder (with all data and results files for that year being stored in that folder.) Others are using a track/circuit folder. Still others use a monthly folder structure. Over the years the number of trouble reports related to data folder structure has been minimal. The most frequent (and annoying) problem happens when users accidentally mix data and results files from two different years on the same target folder. The most common rookie mistake for those who can't seem to build databases at all is that they don't really understand what zip files are and and can't get past the basic requirement of getting data and results files unzipped and sitting on a target folder first before attempting to run a build database routine. The best advice I can give related to data folder structure is to pick a format that makes intuitive sense and have at it. (You can always change it later.) Database Build Modes If seamless building of databases is the goal, then understanding Build Modes during build database routines is in order. There are five Build Modes explained in the Building JCapper Databases Help Doc. Link here: http://www.jcapper.com/helpDocs/JCapper_BuildDatabaseHowTo.htm Chris, you mentioned that your build database routines are taking too long. Did you know that you can build your databases as Appends (Modes 2 and 5) instead of Building from Scratch (Modes 1, 3, and 4?) User Sys Defs Setting: Append to Database During Database Build On the Main Module click System Settings. On the System Settings Interface click System Definitions. In the upper left screen area of the User Sys Defs Screen there is a checkbox setting labeled Append to Database During Database Build. When this box is checked and the setting is persisted on the User Sys Defs Screen, the system default when building databases becomes building as an append instead of building from scratch. Hint: When you build as an append, the interface leaves existing pl_profile.txt file and starterhistory table content intact and simply appends data for those race cards added to the target folder since the last build database routine run on that folder. Because append builds (generally) involve just a handful of race card files, append builds tend to take a lot less time to run than builds from scratch. Checkboxes on Database Builder Screen There are a number of checkboxes on the Database Builder Screen. These checkboxes are designed to give the user explicit control over build database routines. An advanced user might have valid reasons for clicking some boxes but not others. I’ll leave that discussion for another time/thread. Build Modes and Checkboxes A simpler way to gain control over the type of build database routine about to be run is to select a Build Mode from the Build Modes drop down. When a build mode is selected, the interface will auto check the appropriate boxes for that build mode. IMPORTANT HINT: In program versions older than about 3/30/2011, in order for everything to work smoothly – you the user need to make sure you have the right append setting (see above) persisted on the User Sys Defs Screen for the type of build database routine you are about to run. In other words, before running an append build (modes 2 and 5) the append box needs to be checked and the setting persisted on the User Sys Defs Screen. Before running a build from scratch (modes 1, 3, and 4) the append setting needs to be unchecked and the setting persisted on the User Sys Defs Screen. In the current program version, selecting a build mode (which auto checks the boxes for you) completely overrides the append setting from the User Sys Defs Screen. Chris, based on what you’ve written, my suggestion would be to use a quarterly folder structure (or keep your current folder structure intact) and start building your databases as appends. Hint: You WILL want to start off each new year with a a brand new folder structure. Data and results file names do not have a year element. One of the cardinal rules you never want to break is to mix data and results files from different years on the same target folder. One More Hint: The FIRST build on any folder should be done from scratch (modes 1, 3, and 4.) Subsequent builds on a target folder can be done as appends (modes 2 or 5.) -jp . ~Edited by: jeff on: 5/26/2011 at: 5:20:31 PM~
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |