home *** CD-ROM | disk | FTP | other *** search
- **** VERY IMPORTANT **** March 29/92 - TLR
-
- ** CBU101 is a format changing program. If you have an old Comixbase program
- with the pre-Comixbase 2.50 format, then you need only run this program
- once.
-
- ** CBR164 is an update to the data file restructuring program. This is
- program used to clean and remove data which is no longer used.
- It also sorts the title file in alphabetical order.
-
- IE: If you delete a file, the information will still be in the database,
- just unaccessable. That data is taking up space, and so CBR164 will
- remove it from the database.
-
- ****************************************************************************
- First make sure you have made a backup of your database. I try to ensure that
- this program is bug-free on release, however some bugs may be very subtle,
- and will only arise under the right circumstances. For instance, a bug may
- cause small corruptions in the database structure which may render the
- database useless.
- ****************************************************************************
-
- Howdy, if you have never used Comixbase before then ignore the rest of
- this file and the program called CBU101, otherwise read on.
-
- (For pre-Comixbase 2.50 user's only):
-
- To use Comixbase 2.50 and up, you need to run the program called CBU101.
- This program will change the format of the database very slightly, but in
- a necessary way in order to get some improved speed.
- After running CBU101, it is advisable that you run the program called CBR164.
- This is an update to CBR162 (Comixbase Restructure). It will clean up
- the data files.
-
- **** Do NOT use CBR164 before you have used the CBU101 program. ****
-
- In order to make Comixbase a little faster I have made a slight change
- to the text data formats (comix.d1 thru comix.d5). I have noticed that
- on average a good percentage of my comics in the database had blank lines
- in the various data fields (especially the artist fields).
-
- Since the database actually stores the text lines as a byte representing
- the length of the line, followed by the text itself, for lines which
- contain absolutely no data (IE: no blank spaces), it would store a single
- byte which was equal to zero.
-
- What that means is everytime you change to the next or previous comic,
- if there were blanks lines in that new comic it would have to access the
- text data files and then read that single byte in. Since this occurs often
- I decided that if there was a common place where the zero byte would always
- be then I could simply detect it with an if/then type statement and cut
- out the need to access the file.
-
- So I came up with this: the first byte of every text data file will
- be the 0 byte or null string. This speeds access a fair bit (not much from
- ram though), but off of a hard drive or (egad!) a floppy it's noticible.
-
- Example: With my collection moving from the end (the Z's area) back up to
- the A's area (Action Comics) using the left arrow key to skip
- titles at a time (on the hard drive), the old format took 53 seconds,
- the new format took 45 seconds. So it is a bit faster.
-
- (I will add buffering in the near future. This will make running off floppies
- feasible.)
-