[EMAIL PROTECTED] wrote:
So is using the NULL type of delimiting faster or using
Char var1 = "test";
Char var2 = 25";
typedef struct
{
Char name[64];
Char level[64];
} QuestType;
QuestType Appt:
StrCopy (Appt.name, var1);
StrCopy ( Appt.level, var1);
blah, blah, blah.........
this would make reading back the record easier, but seems it would
have limitations and some wasted memory space. But you would allways
know the offset IE names data 0-63 levels data 64-127
If you are going to have a lot of records, then it is probably best to
reduce the record size by using null terminated strings. Hard to say if
you will suffer a speed penalty, without really knowing what the
database will look like, and what king of access (how often) you need to
find a record in the db. Remember that you might save some computation
time by using fixed record field lengths, but you might also LOSE time
when you need to read/write/clear those larger blocks of memory for each
record.
Often, people will put a length variable preceding the string of
characters, so that you can quickly access the start of the next
string. So, your record structure might look something like this concept:
UInt16 str1len
char[#] str1
char null
UInt16 str2len
char[#] str2
char null
When you lock the record handle to access the data, you could
immediately read the str1len value, and use that as an offset to jump
right to the start of the next string, or its corresponding length variable.
Bob
Robert Moynihan <[EMAIL PROTECTED]> wrote ..
[EMAIL PROTECTED] wrote:
Logan Shaw <[EMAIL PROTECTED]> wrote ..
druid wrote:
Yea, but that brings up another question
If the structure of a record is actualy a series on
teminated strings, when you retrieve records, how do you
know the length of any field in that record
You just have to figure out a way to do this yourself.
You mean that a record is not normally just a single string
with delimiters
That's right. A record is a chunk of memory, and you can format it any
way that you like. You CAN choose to delimit it with NULL characters,
as you have done below by specifically including the trailing NULLs in
the DmWrite commands.
Like this
size = StrLen(name)+1+ StrLen(level)+1;
index = dmMaxRecordIndex;
h = DmNewRecord(dbPtr, &index, size);
ptr = MemHandleLock(h);
Offset = 0;
DmWrite(ptr, Offset, (const void *)name, StrLen(name)+1);
Offset += StrLen(name)+1;
DmWrite(ptr, Offset, (const void *)level, StrLen(level)+1);
err = DmCloseDatabase(dbPtr);
maybe im doing this all wrong
This stuff seems to look OK, as long as you have previously defined all
those variable types correctly.
How do you know the length of any field? Using the method above, the
only real way is to iterate through the record, looking at each string
in turn.
Bob
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please
see http://www.palmos.com/dev/support/forums/
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
