"text_new" bug ?

Questions about MultiCharts and user contributed studies.
User avatar
ym
Posts: 53
Joined: 26 Feb 2010

"text_new" bug ?

Postby ym » 31 Jul 2010

I think I have a random (non systematic) bug in using the 'text_new' keyword with a time-offset attribute (X axis).

It's sometime drawn in the wrong place as shown in the attached pic:
tss.jpg
tss.jpg (389.55 KiB) Viewed 1078 times

Here's the code :

Code: Select all

   vars: txt(0), OffsetY(5);
   Value2 = GetAppInfo(aiHighestDispValue ) ;
   Value3 = GetAppInfo(aiLowestDispValue ) ;
   value1=TimeToMinutes(time);  //converts currentbar time to minutes
   value1=value1-1;  //substracts 1 minute
   value1=MinutesToTime(value1);  //converts back to time
   txt=text_new(date , value1, high  + minmove/pricescale*15*OffsetY, "");    
   text_setstring(txt, "next bar is:" + numtostr(time,0));  //displays next bar's time !
   text_setstyle(1,2,2);
   text_setcolor(txt, yellow);

Anyone who knows what's wrong ?

User avatar
TJ
Posts: 6552
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 966 times
Been thanked: 1892 times

Re: "text_new" bug ?

Postby TJ » 31 Jul 2010

Wish I could help...
but don't feel like debugging generic variables today.
Maybe someone can give it a try.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 51 times

Re: "text_new" bug ?

Postby Tresor » 31 Jul 2010

ym, I noticed you are using MBT as a data feed provider for forex. Can one plot ''volume built on tick count'' with MBT data in MC?

Thanks

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 01 Aug 2010

This bug does NOT appear in TS as shown in the here-attached pic ...
Anyone's got an idea ?
Thanks
Attachments
tss004.jpg
TS
tss004.jpg (138.14 KiB) Viewed 1074 times

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 01 Aug 2010

Tresor (build volume on tick count) ,
Yes, that's how my instruments are set in MC.

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 01 Aug 2010

I have never had a problem with this command set.

MinutesToTime(TimeToMinutes(Time) -1)

Try putting the above directly in the Text_new command.

Either that or put wtime below in that Text_New command after setting it as below.

Wtime = MinutesToTime(TimeToMinutes(Time) -1);

Doing it this way reduces chances of your own bugs too.

Also if your code is executing on every tick rather than only on barstatus = 2 be aware that once lastbaronchart is hit, it is deleting the text on each new tick and redrawing it until finally on barstatus=2 is leaves it there. For this reason for a while now I almost always have been creating my trend-lines/text/arrows (the ones I will be reusing at least) when currentbar=1 and I hide them and just move them into view when I need them as well as hide them when I do not need them. It also saves problems with the Text_delete process which can end up deleting the wrong Text because of this constant redrawing problem if more than one study on the same chart is doing it (timing issues come into play).
Last edited by bowlesj3 on 01 Aug 2010, edited 2 times in total.

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 01 Aug 2010

Bowles,
The bug is still there (see here-attached pic).
The formula is not involved (your is concise and elegant, mine was longer and graceless ;-) but they both work)
The proof is that it works perfectly in TS.
I think it's a bug with MC which does not correctly set the drawing position one bar out of hundreds (it seldom happens, but it sure does !).
Any other clue ?
Attachments
tss006.jpg
the bug is still there whatever the formula ...
tss006.jpg (481.72 KiB) Viewed 1061 times

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 01 Aug 2010

Question, I noticed some other fields that are outside the scope of the EL code you show. Is that the only Text_new statement or Text move related statement in the script?

If it is the only one then try getting the date and time from the Text itself (Text_GetDate Text_GetTime) immediately after the Text_new command is issued and comparing the result to the date and time of the bar (actually I guess you have to subtract 1 again because you are shifting that text back a bar). If they are different then do a print to file statement along with an alert. In theory this should not get tripped. Maybe print the bartstatus value in a print statement to see if there is something strange going on. There is a bug that exists in MC and TS too where the study does not get executed for every tick. It is random. It seems to occur much less in MC 6.?. Maybe it is causing it somehow. It should not because the barstatus=2 tick always seems to execute the study without fail and that is the one that locks the Text in place (rather than the constant delete and redraw process). I would suggest reading over the threads on this problem but I can not find the posts because the new forum search is not working properly.

That leads to another question. Is this problem occuring during lastbaronchart or is it a history bar?

I noticed that text_setstyle(1,2,2) does not have txt for the ID. Maybe that is somehow getting the code confused. Try changing it to text_setstyle(txt,2,2).

Try printing the txt ID of the text in a print to file statement. That may give info or a pattern to the problem. You seem to be creating a text on each new bar. Maybe test to see if the current txt field ID is always one greater than the prior txt field ID and if not then issue an alert.

You can see what I am doing. It is a standard debugging technique. You try and expose everything that should be okay to see if it is not (you test your assumptions). You keep doing that and doing that (over and over) until you expose something out of line.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 51 times

Re: "text_new" bug ?

Postby Tresor » 02 Aug 2010

ym wrote:Tresor (build volume on tick count) ,
Yes, that's how my instruments are set in MC.


Thanks ym.

Can you bild volume histogram with MBT data - something like on the picture?
Attachments
Volume Histogram MBT data.png
Volume Histogram MBT data.png (31.38 KiB) Viewed 1055 times

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 02 Aug 2010

tresor,
yes, it looks like 'you can do it' !
(see pic)
Attachments
tresor.jpg
MBT feed with Trade volumes
tresor.jpg (507.41 KiB) Viewed 1079 times

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 02 Aug 2010

Bowles,

Answers to your questions :
1- Nope, the other fields are 'remnants' but not used in the code I sent. The code I sent can be used 'as is'.
2- The problem occurs on history bars (as shown in the .jpg)

Actions taken :
1- I have changed the text_setstyle(1,2,2) to text_setstyle(txt,2,2)
2- I have carefully monitored what's going on in the output log
Both previous actions to no avail ...

Here's the modified piece of code :

Code: Select all

   vars: txt(0), OffsetY(1);
   if currentbar=1 then cleardebug;
   value1=MinutesToTime(TimeToMinutes(Time) -1);
   txt=text_new(date , value1, high  + minmove/pricescale*15*Offsety, "");    
   text_setstring(txt, "next bar:" + numtostr(time,0));   //displays next bar's time
   text_setstyle(txt,2,2);
   text_setcolor(txt, cyan);
   if date=1100730 and time <0059 then  print ("time is: " +numtostr(time,0)+  " - I display: "+numtostr(time,0)+  " above previous bar which is: "+ numtostr(value1,0));

I also have sent a documented email to TSS.

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 02 Aug 2010

You might want to try printing to your log the results of Text_GetTime (immediately after the Text_New is issued). See if it matches the time you are submitting to the Text_new. My bet is they are different.

I would say barstatus is going to be 2 but that would be an assumption. Maybe it is worth seeing.
Last edited by bowlesj3 on 02 Aug 2010, edited 1 time in total.

Tresor
Posts: 1104
Joined: 29 Mar 2008
Has thanked: 12 times
Been thanked: 51 times

Re: "text_new" bug ?

Postby Tresor » 02 Aug 2010

ym wrote:tresor,
yes, it looks like 'you can do it' !
(see pic)


Thanks :mrgreen:

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 02 Aug 2010

Bowles,

See here-attached pic, the bug appears more clearly now with an added 'Text_GetTime' ...

Code: Select all

   vars: txt(0), OffsetY(1);
   if currentbar=1 then cleardebug;
   value1=MinutesToTime(TimeToMinutes(Time) -1);
   txt=text_new(date , value1, high  + minmove/pricescale*15*Offsety, "");    
   if date=1100802 and time <1912 then 
   print ("                  " + "Text_GetTime is: " + numtostr(Text_GetTime(txt),0) +
      " - Time is: " +numtostr(time,0)+  " - so, I display: "  + numtostr(time,0)+ 
      " above previous bar which is: "+ numtostr(value1,0));
   text_setstring(txt, "next bar:" + numtostr(time,0));   //displays next bar's time
   text_setstyle(txt,2,2);
   text_setcolor(txt, cyan);

What may we conclude now ?
Attachments
bowles.jpg
'Text_New' bug ...
bowles.jpg (781.48 KiB) Viewed 1132 times

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 02 Aug 2010

I can't get the attachment.

I think value1 and text_Gettime are the two that need comparison.

What happens if you create a special field such as Back1BarTime? You never know.

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 03 Aug 2010

Try this

If time <> 0000 then
begin
if time <> MinutesToTime(TimeToMinutes(Time[1]) +1) then
print "time not in sequence";
end;

I suggest this because I seem to remember actually seeing the time command once report the incorrect time for its bar. I only saw it the one time.

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 03 Aug 2010

bowles,
can you read the attachment now ?
if not, I'll PM it to you.
Attachments
bowles.png
bowles.png (397.42 KiB) Viewed 1059 times

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 03 Aug 2010

bowles,
I think that we REALLY have a bug here ... what do you think ?

Code: Select all

   vars: txt(0), OffsetY(1);
   if currentbar=1 then cleardebug;
   value1=MinutesToTime(TimeToMinutes(Time) -1);
   txt=text_new(date , value1, high  + minmove/pricescale*15*Offsety, "");    

   If time <> 0000 then begin
      if time <> MinutesToTime(TimeToMinutes(Time[1]) +1) then print ("**** Warning : time not in sequence (look below) !!! ****");
   end;
   print ("Text_GetTime is: " + numtostr(Text_GetTime(txt),0) +
         " - Time is: " +numtostr(time,0)+  " - so, I display: "  + numtostr(time,0)+ 
         " above previous bar which is: "+ numtostr(value1,0));
   end;
   text_setstring(txt, "next bar:" + numtostr(time,0));   //displays next bar's time
   text_setstyle(txt,2,2);
   text_setcolor(txt, cyan);

output log :
Text_GetTime is: 609 - Time is: 610 - so, I display: 610 above previous bar which is: 609
Text_GetTime is: 610 - Time is: 611 - so, I display: 611 above previous bar which is: 610
Text_GetTime is: 611 - Time is: 612 - so, I display: 612 above previous bar which is: 611
Text_GetTime is: 612 - Time is: 613 - so, I display: 613 above previous bar which is: 612
Text_GetTime is: 613 - Time is: 614 - so, I display: 614 above previous bar which is: 613
**** Warning : time not in sequence (look below) !!! ****
Text_GetTime is: 615 - Time is: 616 - so, I display: 616 above previous bar which is: 615
Text_GetTime is: 616 - Time is: 617 - so, I display: 617 above previous bar which is: 616
Text_GetTime is: 617 - Time is: 618 - so, I display: 618 above previous bar which is: 617
Text_GetTime is: 618 - Time is: 619 - so, I display: 619 above previous bar which is: 618
Text_GetTime is: 619 - Time is: 620 - so, I display: 620 above previous bar which is: 619
Text_GetTime is: 620 - Time is: 621 - so, I display: 621 above previous bar which is: 620
Text_GetTime is: 621 - Time is: 622 - so, I display: 622 above previous bar which is: 621
Text_GetTime is: 622 - Time is: 623 - so, I display: 623 above previous bar which is: 622
Text_GetTime is: 623 - Time is: 624 - so, I display: 624 above previous bar which is: 623
Text_GetTime is: 624 - Time is: 625 - so, I display: 625 above previous bar which is: 624
Text_GetTime is: 625 - Time is: 626 - so, I display: 626 above previous bar which is: 625
Text_GetTime is: 626 - Time is: 627 - so, I display: 627 above previous bar which is: 626
Text_GetTime is: 627 - Time is: 628 - so, I display: 628 above previous bar which is: 627
Text_GetTime is: 628 - Time is: 629 - so, I display: 629 above previous bar which is: 628
Text_GetTime is: 629 - Time is: 630 - so, I display: 630 above previous bar which is: 629
**** Warning : time not in sequence (look below) !!! ****
Text_GetTime is: 631 - Time is: 632 - so, I display: 632 above previous bar which is: 631
Text_GetTime is: 632 - Time is: 633 - so, I display: 633 above previous bar which is: 632
Text_GetTime is: 633 - Time is: 634 - so, I display: 634 above previous bar which is: 633
Text_GetTime is: 634 - Time is: 635 - so, I display: 635 above previous bar which is: 634
Text_GetTime is: 635 - Time is: 636 - so, I display: 636 above previous bar which is: 635
Text_GetTime is: 636 - Time is: 637 - so, I display: 637 above previous bar which is: 636

User avatar
TJ
Posts: 6552
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 966 times
Been thanked: 1892 times

Re: "text_new" bug ?

Postby TJ » 03 Aug 2010

suggestion:
set your maxbarsback and try again.

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 03 Aug 2010

maxbarsback set to 200 ...
same old song !!! ...

**** Warning : time not in sequence (look below) !!! ****
Text_GetTime is: 2314 - Time is: 2315 - so, I display: 2315 above previous bar which is: 2314
Text_GetTime is: 2315 - Time is: 2316 - so, I display: 2316 above previous bar which is: 2315
Text_GetTime is: 2316 - Time is: 2317 - so, I display: 2317 above previous bar which is: 2316
Text_GetTime is: 2317 - Time is: 2318 - so, I display: 2318 above previous bar which is: 2317
Text_GetTime is: 2318 - Time is: 2319 - so, I display: 2319 above previous bar which is: 2318
Text_GetTime is: 2319 - Time is: 2320 - so, I display: 2320 above previous bar which is: 2319
Text_GetTime is: 2320 - Time is: 2321 - so, I display: 2321 above previous bar which is: 2320
Text_GetTime is: 2321 - Time is: 2322 - so, I display: 2322 above previous bar which is: 2321
Text_GetTime is: 2322 - Time is: 2323 - so, I display: 2323 above previous bar which is: 2322
Text_GetTime is: 2323 - Time is: 2324 - so, I display: 2324 above previous bar which is: 2323
Text_GetTime is: 2324 - Time is: 2325 - so, I display: 2325 above previous bar which is: 2324
Text_GetTime is: 2325 - Time is: 2326 - so, I display: 2326 above previous bar which is: 2325
Text_GetTime is: 2326 - Time is: 2327 - so, I display: 2327 above previous bar which is: 2326
Text_GetTime is: 2327 - Time is: 2328 - so, I display: 2328 above previous bar which is: 2327
Text_GetTime is: 2328 - Time is: 2329 - so, I display: 2329 above previous bar which is: 2328
Text_GetTime is: 2329 - Time is: 2330 - so, I display: 2330 above previous bar which is: 2329
Text_GetTime is: 2330 - Time is: 2331 - so, I display: 2331 above previous bar which is: 2330
Text_GetTime is: 2331 - Time is: 2332 - so, I display: 2332 above previous bar which is: 2331
Text_GetTime is: 2332 - Time is: 2333 - so, I display: 2333 above previous bar which is: 2332
Text_GetTime is: 2333 - Time is: 2334 - so, I display: 2334 above previous bar which is: 2333
Text_GetTime is: 2334 - Time is: 2335 - so, I display: 2335 above previous bar which is: 2334
Text_GetTime is: 2335 - Time is: 2336 - so, I display: 2336 above previous bar which is: 2335
Text_GetTime is: 2336 - Time is: 2337 - so, I display: 2337 above previous bar which is: 2336
Text_GetTime is: 2337 - Time is: 2338 - so, I display: 2338 above previous bar which is: 2337
Text_GetTime is: 2338 - Time is: 2339 - so, I display: 2339 above previous bar which is: 2338
**** Warning : time not in sequence (look below) !!! ****
Text_GetTime is: 2340 - Time is: 2341 - so, I display: 2341 above previous bar which is: 2340
Text_GetTime is: 2341 - Time is: 2342 - so, I display: 2342 above previous bar which is: 2341
Text_GetTime is: 2342 - Time is: 2343 - so, I display: 2343 above previous bar which is: 2342
Text_GetTime is: 2343 - Time is: 2344 - so, I display: 2344 above previous bar which is: 2343
Text_GetTime is: 2344 - Time is: 2345 - so, I display: 2345 above previous bar which is: 2344
Text_GetTime is: 2345 - Time is: 2346 - so, I display: 2346 above previous bar which is: 2345
Text_GetTime is: 2346 - Time is: 2347 - so, I display: 2347 above previous bar which is: 2346
Text_GetTime is: 2347 - Time is: 2348 - so, I display: 2348 above previous bar which is: 2347
Text_GetTime is: 2348 - Time is: 2349 - so, I display: 2349 above previous bar which is: 2348
Text_GetTime is: 2349 - Time is: 2350 - so, I display: 2350 above previous bar which is: 2349
Text_GetTime is: 2350 - Time is: 2351 - so, I display: 2351 above previous bar which is: 2350
Text_GetTime is: 2351 - Time is: 2352 - so, I display: 2352 above previous bar which is: 2351
Text_GetTime is: 2352 - Time is: 2353 - so, I display: 2353 above previous bar which is: 2352
Text_GetTime is: 2353 - Time is: 2354 - so, I display: 2354 above previous bar which is: 2353
**** Warning : time not in sequence (look below) !!! ****
Text_GetTime is: 2355 - Time is: 2356 - so, I display: 2356 above previous bar which is: 2355
Attachments
tss.jpg
MaxBarsBack
tss.jpg (16.81 KiB) Viewed 1047 times

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 03 Aug 2010

Interesting. Lets format that seq error for more easy reading with this simple code. Maybe this is the problem.

Code: Select all

If time <> 0000 then
     begin
     if time <> MinutesToTime(TimeToMinutes(Time[1]) + 1) then
               begin
               print (
                       "SeqErr: ", " ",
                        "   time=", " " ,
                            Time, " " ,
                        "   Time[1]=", " " ,
                            Time[1], " " ,
                        "   BarStatus=", " " ,
                            BarStatus, " "
                       );
               print (
                       " ", " "
                       );
               end;
     end;

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 04 Aug 2010

Bowles,

Will you get some interesting info from the following ?

code :

Code: Select all

   [intrabarOrderGeneration=false]
   vars: txt(0), OffsetY(1);
   if currentbar=1 then cleardebug;
   value1=MinutesToTime(TimeToMinutes(Time) -1);
   txt=text_new(date , value1, high  + minmove/pricescale*15*Offsety, "");    
   //if date=1100628 and time <2312 then  begin
      If time <> 0000 then begin
         if time <> MinutesToTime(TimeToMinutes(Time[1]) +1) then begin
            //print ("**** Warning : time not in sequence (look below) !!! ****");
            print (NewLine, "*** Warning SeqErr -->   Time:", numtostr(Time,0),
            "   Time[1]:" ,  numtostr(Time[1],0),
            "   BarStatus:" , numtostr(BarStatus(1),0), " ****", NewLine);
         end;
      end;
      print ("Text_GetTime is: " + numtostr(Text_GetTime(txt),0) +
         " - Time is: " +numtostr(time,0)+  " - so, I display: "  + numtostr(time,0)+ 
         " above previous bar which is: "+ numtostr(value1,0));
   //end;
   text_setstring(txt, "next bar:" + numtostr(time,0));   //displays next bar's time
   text_setstyle(txt,2,2);
   text_setcolor(txt, cyan);

output log:
Text_GetTime is: 2336 - Time is: 2337 - so, I display: 2337 above previous bar which is: 2336
Text_GetTime is: 2337 - Time is: 2338 - so, I display: 2338 above previous bar which is: 2337
Text_GetTime is: 2338 - Time is: 2339 - so, I display: 2339 above previous bar which is: 2338

*** Warning SeqErr --> Time:2341 Time[1]:2339 BarStatus:2 ****

Text_GetTime is: 2340 - Time is: 2341 - so, I display: 2341 above previous bar which is: 2340
Text_GetTime is: 2341 - Time is: 2342 - so, I display: 2342 above previous bar which is: 2341
Text_GetTime is: 2342 - Time is: 2343 - so, I display: 2343 above previous bar which is: 2342
Text_GetTime is: 2343 - Time is: 2344 - so, I display: 2344 above previous bar which is: 2343
Text_GetTime is: 2344 - Time is: 2345 - so, I display: 2345 above previous bar which is: 2344
Text_GetTime is: 2345 - Time is: 2346 - so, I display: 2346 above previous bar which is: 2345
Text_GetTime is: 2346 - Time is: 2347 - so, I display: 2347 above previous bar which is: 2346
Text_GetTime is: 2347 - Time is: 2348 - so, I display: 2348 above previous bar which is: 2347
Text_GetTime is: 2348 - Time is: 2349 - so, I display: 2349 above previous bar which is: 2348
Text_GetTime is: 2349 - Time is: 2350 - so, I display: 2350 above previous bar which is: 2349
Text_GetTime is: 2350 - Time is: 2351 - so, I display: 2351 above previous bar which is: 2350
Text_GetTime is: 2351 - Time is: 2352 - so, I display: 2352 above previous bar which is: 2351
Text_GetTime is: 2352 - Time is: 2353 - so, I display: 2353 above previous bar which is: 2352
Text_GetTime is: 2353 - Time is: 2354 - so, I display: 2354 above previous bar which is: 2353

*** Warning SeqErr --> Time:2356 Time[1]:2354 BarStatus:2 ****

Text_GetTime is: 2355 - Time is: 2356 - so, I display: 2356 above previous bar which is: 2355
Text_GetTime is: 2356 - Time is: 2357 - so, I display: 2357 above previous bar which is: 2356

User avatar
Anastassia
Posts: 179
Joined: 18 Jan 2010
Been thanked: 4 times

Re: "text_new" bug ?

Postby Anastassia » 04 Aug 2010

Hi Yann,

I think I have a random (non systematic) bug in using the 'text_new' keyword with a time-offset attribute (X axis).

It's sometime drawn in the wrong place as shown in the attached pic..


Engineers have double checked and confirm it is a bug. It will be fixed in the following versions of MultiCharts.

Thank you

bowlesj3
Posts: 2017
Joined: 21 Jul 2007
Has thanked: 171 times
Been thanked: 402 times

Re: "text_new" bug ?

Postby bowlesj3 » 04 Aug 2010

Thanks all. Now to run that seqerr code on my setup and see what happens.

Interesting how you like to format your print statements YM. I format them so I can very quickly adapt them for new debugging use by simply copying and pasting an existing field down then putting the new field name in (speed to me is like gold). For example in my new version below I just copy and paste the two lines used for "time" then change the name to "date". I have a dummy version ready with both print and fileappend for all studies that give me a problem. fileappend has advantages: 1/ retained during raiseruntimeerror 2/ can be cleared while still in use, 3/ can be written to form any study to centralize info. However after too many writes it aborts the script. Also fileappend forces you to use numtostr.

I like your colon use YM. Maybe I will adjust my setup (did that below).

Code: Select all

If time <> 0000 then
     begin
     if time <> MinutesToTime(TimeToMinutes(Time[1]) + 1) then
               begin
               print ( File("C:\DeleteThis.txt"),
                       "SeqErr: ", " ",
                        "   Date", ":",
                            Date, " ",
                         "   time", ":",
                            Time, " ",
                        "   Time[1]", ":" ,
                            Time[1], " ",
                        "   BarStatus", ":",
                            BarStatus, " "
                       );
               print ( File("C:\DeleteThis.txt"),
                   " "
                       );
               end;
     end;


My result is okay because it is the e-mini and the gaps are during times when there are no trades coming through (always the same time during the day and very large gaps).

SeqErr: Date:1100726.00 time:1631.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100726.00 time:1801.00 Time[1]:1730.00 BarStatus: 2.00

SeqErr: Date:1100727.00 time:1631.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100727.00 time:1801.00 Time[1]:1730.00 BarStatus: 2.00

SeqErr: Date:1100728.00 time:1631.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100728.00 time:1801.00 Time[1]:1730.00 BarStatus: 2.00

SeqErr: Date:1100729.00 time:1631.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100729.00 time:1801.00 Time[1]:1730.00 BarStatus: 2.00

SeqErr: Date:1100801.00 time:1801.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100802.00 time:1631.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100802.00 time:1801.00 Time[1]:1730.00 BarStatus: 2.00

SeqErr: Date:1100803.00 time:1631.00 Time[1]:1615.00 BarStatus: 2.00

SeqErr: Date:1100803.00 time:1801.00 Time[1]:1730.00 BarStatus: 2.00


I don't bother with the numtostr to get rid of the .00 because no one is going to see most of it anyway. The fact that fileappend forces you to use numtostr for me is just a pain that slows me down. On the other hand the format allows one to copy it in and change the names fast so it is not that bad I guess. On the other hand numtostr slows processing a bit if you want to leave them in for a few weeks while maybe anticipating problems in "fresh off the press" code.

Cool, learn something every day. Thanks again all.
Last edited by bowlesj3 on 04 Aug 2010, edited 15 times in total.

User avatar
ym
Posts: 53
Joined: 26 Feb 2010

Re: "text_new" bug ?

Postby ym » 04 Aug 2010

Anastassia,

Thanks again to you and all guys in Rostov-on-Don for being so concerned and reactive to customers bug reports.

That's one point that makes me feel satisfied in using your platform.

Cheerio.


Return to “MultiCharts”