|
JCapper Message Board JCapper 101
|
| Home |
Register
Log In |
| By | Bridge Folder - a strategy for quick index files spanning multiple folders | |
| jeff 2/4/2009 3:24:08 PM | Jeff Quick Index Files Spanning Multiple Databases, this the only problem I am having. I have a folder 2008 with one folder in it Q4-08, all tracks in it. Next folder 2009 with folder in it Q1-09 all tracks. Reading your instruct multiple times, I get Complete _History.txt to run fine. When I down load the next day, this will not show up in the Complete _History.txt . So I erase it, redo Quick Index Files Spanning Multiple Databases. 2009 is the new index, 2008 is add existing index. Your program is great, this the only problem I am having. Arnold Hi Arnold, After you accumulate enough data the need to constantly update a complete history file isn't as pressing as it is at the beginning. In the setup you describe right now - with just over 1 quarter of data - the need to keep the complete history file updated is a lot more pressing. You might try a strategy I call a Bridge Folder. I'll mark the key bullet points with a * character. Understand that while the instructions for using this strategy are going to be several paragraphs long - the idea behind it is really simple. You'll almost never have to erase and create the complete history file from scratch again. And once you get into the swing of things the steps become almost second nature. Here's what you do: *1. Create a new folder under c:\2009. Give the folder a name that describes what it will be used for. Something like "Bridge" would work. The full path to this folder would be c:\2009\Bridge. *2. Each day when you download new data and results files save and unzip them onto your c:\2009\Bridge folder. On your System Definitions Screen, keep the Build Database as Append box unchecked so that when you run a build database it always runs as a rebuild from scratch. *3. At the end of each day - or every 2-3 days - the frequency is really up to you - run a Build Database from Scratch on your c:\2009\Bridge folder. *4. Immediately after each build database, move all data and results files from the Bridge folder onto the Q1 (or current quarter) folder. Windows Explorer is perfect for this. Right click the Start button and select Explore to launch Windows Explorer and then navigate to your c:\2009\Bridge folder - highlight all of your data and results files - right click on the highlighted group - and select CUT. Then navigate to your c:\2009\Q1 folder and PASTE the highlighted group of data and results files onto your c:\2009\Q1 folder. This will cause all of the highlighted files to be moved from the Bridge folder to your Q1 folder. *5. The purpose behind doing this is to get these files off of the Bridge folder so that they are not included the next time you run a build database on that folder. *6. Each time you run a build database on the Bridge folder you will have a new set of files on that folder. The resulting pl_playlist.txt file from each new build database that you run on the Bridge folder will contain data only for those files saved and unzipped onto that folder since the last time you wiped it clean in step 4 above. *7. The pl_profile.txt file sitting on the Bridge folder makes a perfect source that you can use to add to existing playlist, quick index, or complete history files. Here's how to use the Data Window's Quick Index File function: *8. In the Data Window, click MENU, select Exports, and then select Quick Index File Append to Existing. Use the dialog box to navigate to the folder containing the target playlist or quick index file that you wish to add or append to. Click the target file and hit the Open button. The selected target file name along with text indicating that you are about to append to it will be displayed along the top of the Data Window on the title bar. The next Data Window query that you run will drive the append to existing file function that you just toggled "on." *9. Point the Data Window at your Bridge folder and run an ALL button Data Window query on the pl_profile.txt file sitting there. The Data Window will append every horse returned by the query to the end of the target file selected in step 8. This will accomplish exactly the same thing as if you had initially saved the data and results files onto the Q1 folder and run a build database on that folder. The playlist file on the Q1 folder is now updated and contains the latest and most current data from the data and results files that you have. Repeat step 8 using your complete history file as the target instead of the Q1 playlist file. Then repeat step 9. Your complete history file now contains starters from the most current data and results files that you have. When I first started using this strategy I was a little unsure of what I was doing at first. Then after I got into the swing of things and started repeating the steps on a regular basis - I probably run a database build on the Bridge folder 3-5 times a week and have been doing so for years... It's pretty much second nature at this point. * One more key point - and the reason I LIKE using a Bridge folder: Whenever I run a build database it's only for a few day's worth of files and it's always run on the Bridge folder. That means that the build database routine only takes a few minutes to run - same as an append. * Final key point - IMPORTANT - make regular backups of your complete history file and the pl_profile.txt file on your current quarterly folder. Because this is a manual operation (maybe someday I'll get around to writing a routine so that the database builder runs a quick index file append automatically to a file list supplied by the user) it IS possible (even likely) that a user will screw up at some point and run the Quick Index file function as a create new instead of an append - in which case the target file for the append will be overwritten and the user would have to rebuild it all over again. Having file backups is the easiest way to insure against this. That's it! -jp . ~Edited by: jeff on: 2/4/2009 at: 3:24:08 PM~
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |