D3 Callx - DICT trigger

#11

there is a trigger on the *bp* file.

On Jan 12, 7:08 am, "Scott Ballinger" <scott.ballin... (AT) gmail (DOT) com>
wrote:

Quote:

Why not put the trigger on the BP file?

/Scott

#12

does it save off a copy of the original?
does it save off a copy of the DIFF?

No .. that is why we are using triggers to track changes on tables.

thanks

On Jan 12, 1:05 pm, "Scott Ballinger" <scott.ballin... (AT) gmail (DOT) com>
wrote:

Quote:

Not to beat a dead horse here, but note that the system does maintain a
bunch of information regarding compiled items in the dict:

d3.doc compile.time.date.stamp.ap
token Compile stamps
syntax
category BASIC
type Definition
terse Obtaining the internal information about compiles in D3.
desc contains descriptive internal information about programs
compiled for FlashBASIC.

When a program is compiled, information is placed in the
dictionary level of the file in which the source program
resides. The structure of the information is as follows:

Attr Description
0 item-id
1 CC
2 starting frame number of object code
3 Pick/BASIC object frame count
4 date, internal, when last compiled
5 time in seconds, internal, when last compiled
6 port number of who compiled program last
7 user who compiled program last

9 release version from where program was last compiled
10 options used when last compiled, alphabetically
11 bytes in Pick/BASIC object code
12 FlashBASIC object frame count

e.g., I use attribute 12 to test if a program is currently flash
compiled, then automatically add the "(o" option to the compile command
to insure that it stays flashed (particularly handy for subroutines).
Also, <4> <5> <6> and <7> are handy for assigning blame when something
blows up.

/Scott Ballinger

#13

is it possible to edit program foobar in table mine. THen

copy dict mine foobar (o
(dict bp

what happens in this case .. not at the terminal to try but is this
possible? will test monday

dtsig

On Jan 12, 1:12 pm, "Mark Brown" <mbr... (AT) drexelmgt (DOT) com> wrote:

Quote:

Object code always reflects the current state of source code. If the source
code didn't change, who cares if the program is re-compiled. If the source
did change, you have a trigger to say it did. Maybe add a "Compile (y/N): "
to the trigger and disable compile-catalog, compile-run and :ccompile and
control it internally.

If you are changing INCLUDEs from another file and then recompiling, maybe
you just need to widen the scope of your source file triggers.

Mark

"dtsig" <d... (AT) hotmail (DOT) com> wrote in messagenews:1168612056.957521.317660 (AT) q2g2000cwa (DOT) googlegroups.com...

Mark Brown wrote:
That might be because the "object" code is NOT in the dictionary, the
pointer to the object code is.

Mark Brown

"dtsig" <d... (AT) hotmail (DOT) com> wrote in message
news:1168555695.208770.288860 (AT) i39g2000hsf (DOT) googlegroups.com...
I have placed a trigger on the dict and data portions of a file as we
want to know of any changes dict or data.

It appears to work correct with the one exception that it does not get
triggered when object is changed.

Has anyone seen this?

thanks

That would do it. Is there a way to 'trigger' on any change to object?

thanks

#14

Yes it is possible to deliberately circumvent most simple security.

As my pa used to say, "A lock only keeps honest people honest. It won't
stop a theif."

And, as someone else wise said, "There's no defense against stupidity."

I'll fall back on one of my other posts, update and retrieval locks. Add
something to the update lock of the dict of the bp file and make sure you're
the only one who can update it.

I actually worked for a bank one time that had this kind of scheme for
updates. You took the source, did whatever you wanted, then gave them the
name of the object code and they moved it into the proper DICT once they
were sure you weren't paying yourself extra.

Mark Brown

"dtsig" <dtsig (AT) hotmail (DOT) com> wrote

Quote:

is it possible to edit program foobar in table mine. THen

copy dict mine foobar (o
(dict bp

what happens in this case .. not at the terminal to try but is this
possible? will test monday

dtsig

On Jan 12, 1:12 pm, "Mark Brown" <mbr... (AT) drexelmgt (DOT) com> wrote:
Object code always reflects the current state of source code. If the
source
code didn't change, who cares if the program is re-compiled. If the
source
did change, you have a trigger to say it did. Maybe add a "Compile
(y/N): "
to the trigger and disable compile-catalog, compile-run and :ccompile and
control it internally.

If you are changing INCLUDEs from another file and then recompiling,
maybe
you just need to widen the scope of your source file triggers.

Mark

"dtsig" <d... (AT) hotmail (DOT) com> wrote in
messagenews:1168612056.957521.317660 (AT) q2g2000cwa (DOT) googlegroups.com...

Mark Brown wrote:
That might be because the "object" code is NOT in the dictionary, the
pointer to the object code is.

Mark Brown

"dtsig" <d... (AT) hotmail (DOT) com> wrote in message
news:1168555695.208770.288860 (AT) i39g2000hsf (DOT) googlegroups.com...
I have placed a trigger on the dict and data portions of a file as we
want to know of any changes dict or data.

It appears to work correct with the one exception that it does not
get
triggered when object is changed.

Has anyone seen this?

thanks

That would do it. Is there a way to 'trigger' on any change to object?

thanks

#15

So .. the answer is .. it cant be done with a trigger ..

thanks

Mark Brown wrote:

Quote:

Yes it is possible to deliberately circumvent most simple security.

As my pa used to say, "A lock only keeps honest people honest. It won't
stop a theif."

And, as someone else wise said, "There's no defense against stupidity."

I'll fall back on one of my other posts, update and retrieval locks. Add
something to the update lock of the dict of the bp file and make sure you're
the only one who can update it.

I actually worked for a bank one time that had this kind of scheme for
updates. You took the source, did whatever you wanted, then gave them the
name of the object code and they moved it into the proper DICT once they
were sure you weren't paying yourself extra.

Mark Brown

"dtsig" <dtsig (AT) hotmail (DOT) com> wrote in message
news:1168749525.608782.53090 (AT) v45g2000cwv (DOT) googlegroups.com...
is it possible to edit program foobar in table mine. THen

copy dict mine foobar (o
(dict bp

what happens in this case .. not at the terminal to try but is this
possible? will test monday

dtsig

On Jan 12, 1:12 pm, "Mark Brown" <mbr... (AT) drexelmgt (DOT) com> wrote:
Object code always reflects the current state of source code. If the
source
code didn't change, who cares if the program is re-compiled. If the
source
did change, you have a trigger to say it did. Maybe add a "Compile
(y/N): "
to the trigger and disable compile-catalog, compile-run and :ccompile and
control it internally.

If you are changing INCLUDEs from another file and then recompiling,
maybe
you just need to widen the scope of your source file triggers.

Mark

"dtsig" <d... (AT) hotmail (DOT) com> wrote in
messagenews:1168612056.957521.317660 (AT) q2g2000cwa (DOT) googlegroups.com...

Mark Brown wrote:
That might be because the "object" code is NOT in the dictionary, the
pointer to the object code is.

Mark Brown

"dtsig" <d... (AT) hotmail (DOT) com> wrote in message
news:1168555695.208770.288860 (AT) i39g2000hsf (DOT) googlegroups.com...
I have placed a trigger on the dict and data portions of a file as we
want to know of any changes dict or data.

It appears to work correct with the one exception that it does not
get
triggered when object is changed.

Has anyone seen this?

thanks

That would do it. Is there a way to 'trigger' on any change to object?

thanks

#16

Dude, we need to talk. I don't mind my code getting posted but I want
it made clear that I didn't write most of that code.

T

#17

Sorry T. I guess I just assumed you wrote it because it works so well. Did
you notice I added the part where it catalogs the program? Did you also
notice that there was not one single response to the posting of the code? I
guess no one noticed except you, including the OP.

"Tony Gravagno" <g6q3x9lu53001 (AT) sneakemail (DOT) com.invalid> wrote

Quote:

Dude, we need to talk. I don't mind my code getting posted but I want
it made clear that I didn't write most of that code.

T

#18

Thanks for the stroke but it doesn't earn you discounts.

All I was saying is that the code as-is needs to be optimized. I
didn't want people to think I'd leave code in that condition. (It's
hard to walk with so many eggshells around here. Sorry.)

Lack of response is normal. That's why I sometimes tag my text with
stuff that will tick people off, just to see if they're paying
attention. LOL

T

"Jeffrey Kaufman" wrote:

Quote:

Sorry T. I guess I just assumed you wrote it because it works so well. Did
you notice I added the part where it catalogs the program? Did you also
notice that there was not one single response to the posting of the code? I
guess no one noticed except you, including the OP.

"Tony Gravagno" wrote
Dude, we need to talk. I don't mind my code getting posted but I want
it made clear that I didn't write most of that code.

T

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

D3 Callx - DICT trigger

comp.databases.pick comp.databases.pick

Re: D3 Callx - DICT trigger - 01-13-2007 , 10:33 PM

Re: D3 Callx - DICT trigger - 01-13-2007 , 10:35 PM

Re: D3 Callx - DICT trigger - 01-13-2007 , 10:38 PM

Re: D3 Callx - DICT trigger - 01-14-2007 , 02:24 AM

Re: D3 Callx - DICT trigger - 01-14-2007 , 11:47 PM

Re: D3 Callx - DICT trigger - 01-15-2007 , 10:42 PM

Re: D3 Callx - DICT trigger - 01-15-2007 , 11:14 PM

Re: D3 Callx - DICT trigger - 01-16-2007 , 03:05 AM