Re: Closing Databases

This WebDNA talk-list message is from

1998


It keeps the original formatting.
numero = 16600
interpreted = N
texte = New to 2.1 to hopefullly address this istaken from the online ref....[COMMITDATABASE FileName] Putting [COMMITDATABASE FileName] in your template will cause the specified database file (or full path) to be written but not closed (so it will remain in RAM). This is only needed for special cases where you want to be absolutely certain that a database has been written to disk.CloseDatabase continues to both committ and release the ram that the database is taking up...We are always open to constructive criticism about new tags etc... Keep the comments coming, but we hope that the above is at least helpful... >>>You only need to worry if the database in WebCatalog has been modified -- >>>using something like [Append] [Replace] [Delete], etc. If the database is > >>>'read-only', then simply uploading a new file and then doing a >>>[CloseDatabase] or [FlushDatabases] will re-read the new file. >> >>Sorry, I still do not understand why I would have to close of flush the >>database afterwards. Closing/Flusing a database writes what's in memory >>back to disk. If I first upload a new file and then close or flash the >>database, to my understanding the file in memory is overwriting the new >>file with old data?! What's wrong with my way of thinking. >> >>John > >This is a problem that has been plagueing PCS since I first joined this >list, and could be fixed in one easy step, that for some reason, PCS doesn't >see or won't implement. The problem is based on the confusion of the term >closedatabase. The way PCS has the closedatabase tag define, it is dual >purpose (which is a good thing from a minimalist point of view), to both >write changed databases and to flush unchanged databases (no need to write >to disk.) > >But the users don't get it! (and most users don't want it this way!) Why not >have two tags, [closedatabase] which saves any database to disk again, and >[reloaddatabase] which throws away the ram copy (changes or not) and reloads >the database of the same name. Or perhaps [closedatabase commit=T] or >[closedatabase commit=F]. Personally, I'd like a reload tag so I could >upload a new database, issue the command, and know that the ram file is >flused, and the new database is opened, with practically no downtime. > >I'll now await PCS's reply about why they don't want to do this. (it will >have nothing to do with it not being possible...) > > Brian B. Burton > BOFH >--------------------------------------------------------------- > MMT Solutions - Specializing in Online Shopping Solutions > 973-808-8644 http://www.safecommerce.com Jay Van Vark ******************************* Pacific Coast Software | eCommerce (WebCatalog) 11770 Bernardo Plaza Court | Site Hosting & Development Suite 453 | http://www.smithmicro.com San Diego, CA 92128 ******************************* 619.675.1106** PGP key available upon request. ** Associated Messages, from the most recent to the oldest:

    
  1. Re: Closing databases (Glenn Busbin 2003)
  2. Re: Closing databases (Frank Nordberg 2003)
  3. Re: Closing databases (Glenn Busbin 2003)
  4. Re: Closing databases (Frank Nordberg 2003)
  5. Re: Closing databases (Glenn Busbin 2003)
  6. Closing databases (Frank Nordberg 2003)
  7. Re: Closing Databases (Jay Van Vark 1998)
  8. Re: Closing Databases (Brian B. Burton 1998)
  9. Re: Closing Databases (Bob Minor 1998)
  10. Re: Closing Databases (Sjoerd Jan ter Welle 1998)
  11. Re: Closing Databases (PCS Technical Support 1998)
  12. Re: Closing Databases (Sjoerd Jan ter Welle 1998)
  13. Re: Closing Databases (Grant Hulbert 1998)
  14. Re: Closing Databases (Olin Lagon 1998)
  15. Re: Closing Databases (Kenneth Grome 1998)
  16. Re: Closing Databases (Sjoerd Jan ter Welle 1998)
  17. Re: Closing Databases (Kenneth Grome 1998)
  18. Closing Databases (take 2) (Brian B. Burton 1998)
New to 2.1 to hopefullly address this istaken from the online ref....[COMMITDATABASE FileName] Putting [COMMITDATABASE FileName] in your template will cause the specified database file (or full path) to be written but not closed (so it will remain in RAM). This is only needed for special cases where you want to be absolutely certain that a database has been written to disk.CloseDatabase continues to both committ and release the ram that the database is taking up...We are always open to constructive criticism about new tags etc... Keep the comments coming, but we hope that the above is at least helpful... >>>You only need to worry if the database in WebCatalog has been modified -- >>>using something like [append] [replace] [delete], etc. If the database is > >>>'read-only', then simply uploading a new file and then doing a >>>[closedatabase] or [flushdatabases] will re-read the new file. >> >>Sorry, I still do not understand why I would have to close of flush the >>database afterwards. Closing/Flusing a database writes what's in memory >>back to disk. If I first upload a new file and then close or flash the >>database, to my understanding the file in memory is overwriting the new >>file with old data?! What's wrong with my way of thinking. >> >>John > >This is a problem that has been plagueing PCS since I first joined this >list, and could be fixed in one easy step, that for some reason, PCS doesn't >see or won't implement. The problem is based on the confusion of the term >closedatabase. The way PCS has the closedatabase tag define, it is dual >purpose (which is a good thing from a minimalist point of view), to both >write changed databases and to flush unchanged databases (no need to write >to disk.) > >But the users don't get it! (and most users don't want it this way!) Why not >have two tags, [closedatabase] which saves any database to disk again, and >[reloaddatabase] which throws away the ram copy (changes or not) and reloads >the database of the same name. Or perhaps [closedatabase commit=T] or >[closedatabase commit=F]. Personally, I'd like a reload tag so I could >upload a new database, issue the command, and know that the ram file is >flused, and the new database is opened, with practically no downtime. > >I'll now await PCS's reply about why they don't want to do this. (it will >have nothing to do with it not being possible...) > > Brian B. Burton > BOFH >--------------------------------------------------------------- > MMT Solutions - Specializing in Online Shopping Solutions > 973-808-8644 http://www.safecommerce.com Jay Van Vark ******************************* Pacific Coast Software | eCommerce (WebCatalog) 11770 Bernardo Plaza Court | Site Hosting & Development Suite 453 | http://www.smithmicro.com San Diego, CA 92128 ******************************* 619.675.1106** PGP key available upon request. ** Jay Van Vark

DOWNLOAD WEBDNA NOW!

Top Articles:

Talk List

The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...

Related Readings:

RE: IIS4b2 and WebCatalog b19 (1997) Expected behavior? (1998) WebDNA Grep help wanted (2004) Grep is Broke (2004) Grep -- remove extra returns first (2003) Next (1997) Migrating to NT (1997) WebCat2final1 crashes (1997) Re:Review comparison by PC Magazine: Open for On-line Business (1997) WebCat2: Items xx to xx shown, etc. (1997) Using Plug-In while running 1.6.1 (1997) lookups (2000) Off Topic: Help Wanted (1997) Weird problems with [SHOWIF]s (1997) Forms Search Questions (1997) Modifying order output (1997) Re:E-Mailer (WebCatb15acgiMac) (1997) Calculating multiple shipping... (1997) [WebDNA] Hosting (2015) Extended [ConvertChars] (1997)