1. RE: Anyone want to code a useful utility and get some EuBucks?
SR.Williamson wrote:
> Then it should parse that part of the file, and ask you for each
> sequence whether it should be interpreted as a string or a number
One of the problems with parsing the sequences is that
Rob is truncating long sequences.
Bernie
2. RE: Anyone want to code a useful utility and get some EuBucks?
I didn't realize that long sequences are truncated until Bernie
mentioned it. I normally don't use long sequences.
By strange coincidence, on this project, I do use them, and yep, they
are truncated. I had planned to use ex.err as a debugging tool, but it's
pretty useless now.
Anyone up for coding a debug include file that will write a list of
variables and sequences to an eds database?
tone.skoda at gmx.net wrote:
> I was thinking about doing something similar: to write it to database
> file
> and to view it with edsgui because now it's hard to look at long
> sequences
> in ex.err. But someone said they are truncated in ex.err so ... I don't
> know.
>
> ----- Original Message -----
> From: "SR.Williamson" <writeneu at hotmail.com>
> To: "EUforum" <EUforum at topica.com>
> Sent: Wednesday, June 12, 2002 5:41 PM
> Subject: Anyone want to code a useful utility and get some EuBucks?
>
>
> >
> > This one would be guaranteed to get about 6 months worth of my EuBucks,
> > provided you made it a win or dos file.
> >
> > A utility file that will parse ex.err, strip out EVERYTHING except the
> > variables list in the main Eu file, the list that is right at the end.
> >
> > Then it should parse that part of the file, and ask you for each
> > sequence whether it should be interpreted as a string or a number
> > sequence. I'll give you another 2 months worth of EuBucks if it saves
> > the answers in some sort of preference file so you don't have to enter
> > the answer for sequences you've already decided each time you go through
> > the ex.err file for a specific program.
> >
> > The output file should be the list of variables corrected into something
> > more meaningful.
> >
> > IOW, you'll get something that looks like this.
> >
> > contents of myprog.err
> > number = 4
> > myAccounts = 17
> > stuffILike = {cool stuff}
> > kinds of cool things I have = {books, music, rocks, dead bugs}
> > number of cool things = {books, {sci fi, romance, mags}, {12, 1, 37}}
> >
> > Hopefully that gives you an idea. I would dearly love to program this
> > myself, but I'm afraid it's beyond my meager abilities.
> >
> > Anyone up for it?
> >
> >
> >
3. RE: Anyone want to code a useful utility and get some EuBucks?
On 13 Jun 2002, at 11:59, SR.Williamson wrote:
>
> I didn't realize that long sequences are truncated until Bernie
> mentioned it. I normally don't use long sequences.
>
> By strange coincidence, on this project, I do use them, and yep, they
> are truncated. I had planned to use ex.err as a debugging tool, but it's
> pretty useless now.
>
> Anyone up for coding a debug include file that will write a list of
> variables and sequences to an eds database?
You can't write a general include to list all the vars used up until then, for
two
reasons.
1) you can't point a crash routine at it, so it will not have the correct state
of
the vars at crash time. Many people have asked for this ability.
2) you can't know the names and state of vars unless explicitly coded into
the include at run time. I have asked for this repeatedly, it would be a step up
from mirc.
Kat
4. RE: Anyone want to code a useful utility and get some EuBucks?
Here's what I had in mind.
A little include that sets up the db/file/etc to hide the details for
newbies, and me.
A function/procedure in the include that allows you to call it and send
it a variable name. Whenever you have a variable that needs to be
written, you call, for example (pseudocode)
/start
debugger(myVaribleName, myvariable, variableLocation)
/end
and it writes the state to a file.
Heck, I think most of the work is already done even. I submitted
something called trackvar to Rob ages ago, which was a
lazy/sloppy/amateurish (like all my programs) attempt to get a Win32
watch window in Eu. Probably could easily be modified to write the
results. Which I probably should do. I have no idea how to use eds
though, so I guess that's not the way I'll do it.
Kat wrote:
> On 13 Jun 2002, at 11:59, SR.Williamson wrote:
>
> >
> > I didn't realize that long sequences are truncated until Bernie
> > mentioned it. I normally don't use long sequences.
> >
> > By strange coincidence, on this project, I do use them, and yep, they
> > are truncated. I had planned to use ex.err as a debugging tool, but
it's
> >
> > pretty useless now.
> >
> > Anyone up for coding a debug include file that will write a list of
> > variables and sequences to an eds database?
>
> You can't write a general include to list all the vars used up until
> then, for two
> reasons.
>
> 1) you can't point a crash routine at it, so it will not have the
> correct state of
> the vars at crash time. Many people have asked for this ability.
> 2) you can't know the names and state of vars unless explicitly coded
> into
> the include at run time. I have asked for this repeatedly, it would be a
> step up
> from mirc.
>
> Kat
>
>
>
5. RE: Anyone want to code a useful utility and get some EuBucks?
Guess I'm an idiot. I just downloaded it. Nobody uses it (not even me,
that's why I had to download it), because it's got at least 2 bugs in
it.
Plus, it already writes the variables to a file, unless I screwed that
up also.
SR.Williamson wrote:
> Here's what I had in mind.
>
> A little include that sets up the db/file/etc to hide the details for
> newbies, and me.
>
> A function/procedure in the include that allows you to call it and send
> it a variable name. Whenever you have a variable that needs to be
> written, you call, for example (pseudocode)
> /start
> debugger(myVaribleName, myvariable, variableLocation)
> /end
>
> and it writes the state to a file.
>
> Heck, I think most of the work is already done even. I submitted
> something called trackvar to Rob ages ago, which was a
> lazy/sloppy/amateurish (like all my programs) attempt to get a Win32
> watch window in Eu. Probably could easily be modified to write the
> results. Which I probably should do. I have no idea how to use eds
> though, so I guess that's not the way I'll do it.
>
> Kat wrote:
> > On 13 Jun 2002, at 11:59, SR.Williamson wrote:
> >
> > >
> > > I didn't realize that long sequences are truncated until Bernie
> > > mentioned it. I normally don't use long sequences.
> > >
> > > By strange coincidence, on this project, I do use them, and yep, they
> > > are truncated. I had planned to use ex.err as a debugging tool, but
> it's
> > >
> > > pretty useless now.
> > >
> > > Anyone up for coding a debug include file that will write a list of
> > > variables and sequences to an eds database?
> >
> > You can't write a general include to list all the vars used up until
> > then, for two
> > reasons.
> >
> > 1) you can't point a crash routine at it, so it will not have the
> > correct state of
> > the vars at crash time. Many people have asked for this ability.
> > 2) you can't know the names and state of vars unless explicitly coded
> > into
> > the include at run time. I have asked for this repeatedly, it would be a
> > step up
> > from mirc.
> >
> > Kat
> >
> >
> >
6. RE: Anyone want to code a useful utility and get some EuBucks?
> -----Original Message-----
> From: Bernie Ryan [mailto:xotron at localnet.com]
> SR.Williamson wrote:
> > Then it should parse that part of the file, and ask you for each
> > sequence whether it should be interpreted as a string or a number
>
> One of the problems with parsing the sequences is that
> Rob is truncating long sequences.
<looking at source code>Rob, it looks like you've limited the number of
lines of output in ex.err. What about a manual switch to set this
limit.</looking at source code>
Something like a call to:
crash_lines( integer lines )
Then we could set this to however long we wanted (at least, longer than
anyone should probably make a single sequence:). I don't have the time to
muck with the source right now, but the key seems to be in the call to
Print() from DumpPrivates() and DumpGlobals().
Matt Lewis
