Hello Nello, I have been considering the code required to do this and it will be an interesting exercise to compare the two search types; using an integer index as you suggest and the case statement which I suspect is applying an array type of search as well. BTW each event of the search is for one filename; there is not a lot of files in one go. The parameters are to parse a file OK to decompress then action appropriately, then step to the next compressed file. Thanks all. Al+
Tuesday, October 15, 2002, 2:11:46 PM, you wrote: NS> if by "filetype" you mean a 3 byte extension (like .pas or .txt) NS> and you can constrain the character set (say to case-indifferent NS> alpha, digits and a few special characters like '$' and ' ' and NS> '_') then you have about 40^3 (about 65k) possible file types. NS> you could comfortably hold an array of 65k byte codes for NS> "include/exclude/whatever" and map each 3byte filetype directly NS> to an integer index into this array. NS> BTW you can use a similar trick to quickly generate the NS> index into the table from the 3 byte filetype: construct NS> 3 tables of 256 integers (1 for each possible character in each NS> of the 3 positions 1,2,3 of the filetype) - then sum those integers NS> to create the index into the 65k array. NS> if you have a lot of filetypes this will be faster than doing NS> searching lookups into a table of 65000 codes. NS> -ns NS> http://www.roserox.co.th NS> ----- Original Message ----- NS> From: "Alistair George" <[EMAIL PROTECTED]> NS> To: "Multiple recipients of list delphi" <[EMAIL PROTECTED]> NS> Sent: Tuesday, October 15, 2002 07:02 NS> Subject: [DUG]: To CASE or not? >> Hi all. >> I am writing a backup routine which includes or excludes bunches of >> files/folders. >> For example, during the backup process the compressor must be told which NS> file >> types to exclude. >> My concern is that if this bunch of types is numerous that the >> processing/parsing could be quite resource intensive. >> >> Any suggestions to achieve this would be appreciated. My own feeling is to NS> run >> with a dynamic CASE statement, but even this would seem to take a pretty NS> big >> chunk of processing time given each file is parsed? >> Thanks, >> Alistair George+ >> >> -------------------------------------------------------------------------- NS> - >> New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] >> Website: http://www.delphi.org.nz >> To UnSub, send email to: [EMAIL PROTECTED] >> with body of "unsubscribe delphi" >> Web Archive at: http://www.mail-archive.com/delphi%40delphi.org.nz/ NS> --------------------------------------------------------------------------- NS> New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] NS> Website: http://www.delphi.org.nz NS> To UnSub, send email to: [EMAIL PROTECTED] NS> with body of "unsubscribe delphi" NS> Web Archive at: http://www.mail-archive.com/delphi%40delphi.org.nz/ -- Regards, Alistair+ --------------------------------------------------------------------------- New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] Website: http://www.delphi.org.nz To UnSub, send email to: [EMAIL PROTECTED] with body of "unsubscribe delphi" Web Archive at: http://www.mail-archive.com/delphi%40delphi.org.nz/
