Power Editor will not compile imported XCAP_POLYFIT.ELD

Questions about MultiCharts and user contributed studies.
denizen2
Posts: 125
Joined: 17 Jul 2005

Power Editor will not compile imported XCAP_POLYFIT.ELD

Postby denizen2 » 17 Feb 2007

After updating my (beta) version of MC I have not been able to (re-)compile some files, e.g., XCAP_POLYFIT.ELD. [attached].

This file, or a couple of related ones, has been mentioned as taking a VERY long time to compile. The reason appears to be that maybe several hundreds of constant-parameters have to be pre-computed at the beginning of the compile process.

In the previous beta (Dec 2006), I was able to compile and use this function/indicator, but since updating (Feb 2), and being required to re-compile everything, the PL Editor-compiler just hangs while compiling this file. More specifically, the CPU time is almost zero, but the compiler just reports it is 'busy' and user must wait..... for hours.... yet nothing ever happens. I've tried this twice with the same result, and the second time was after a reboot of the system.

I would say that something has changed between the two versions that has 'broken' the PL compiler for this type of imported file.

I have also noticed that I can not launch the PL Editor from the start menu like I used to, .... now it can only be launched from the MC toolbar. This should be fixed, right?
Attachments
20061206115434XCAP_SAVITZKYGOLAY.ELD
PL Editor-compiler does not compile this anymore?
(298.04 KiB) Downloaded 121 times

ax
Posts: 34
Joined: 21 Nov 2005
Has thanked: 15 times
Been thanked: 2 times

Postby ax » 18 Feb 2007

I have no problem with this one
Attachments
XCAP_PolyFit.xml
(28.84 KiB) Downloaded 105 times

denizen2
Posts: 125
Joined: 17 Jul 2005

Postby denizen2 » 18 Feb 2007

I have no problem with this one


Thanks "ax",
Are you saying you can compile this file in something less than 1hour? And you are using MC version 2.0.763.930?

Do you have any problem with lauching the PL Editor from the Start-Menu->PLEditor, and/or the Window XP QuickStart toolbar? In my case the only place for launching is the MC's toolbar.

Question for TSSupport: Does the installation of the new version( ...763.930) replace and install a new version of the PL Editor? (.... AND the compiler?), or just the MultiChart software itself? Do I likely have some 'dead' shortcuts on my system that no longer point to the correct executables? How do I confirm this without doing yet another re-install?

ax
Posts: 34
Joined: 21 Nov 2005
Has thanked: 15 times
Been thanked: 2 times

Postby ax » 18 Feb 2007

The version I posted compiles in less than a minute

Yes I am using the latest Beta

I Launch PL Editor from a shortcut placed on my desktop during istallation and have never had a problem

I can post a xml file of XCap_PolyCycle and PolyPredict if you wish

Good Luck
ax

denizen2
Posts: 125
Joined: 17 Jul 2005

Postby denizen2 » 18 Feb 2007

The version I posted compiles in less than a minute

Yes I am using the latest Beta

I Launch PL Editor from a shortcut placed on my desktop during istallation and have never had a problem

I can post a xml file of XCap_PolyCycle and PolyPredict if you wish

Good Luck
ax


Thanx 'ax' for the feedback....
After some further 'investigation' I find that I have the problem of launching the PL-editor only from the Start-Menu-MultiCharts->PL-Editor and the QuickLaunch toolbar, but NO problem launching from shortcuts on desktop and 'Pin'-ed version on the StartMenu. It looks like I just discovered how 'shortcuts' can get 'orphaned' and the Windows system does not 'complain' about it (in all cases), rather it just fails to do anything. So we should take this lesson as a 'caveat' about re-installing new versions: must clean up ALL references to 'old' version. 8)

Regarding the xml files you offered: yes, please do post them. I will try to compile them and see if they compile on my system too. :)

ax
Posts: 34
Joined: 21 Nov 2005
Has thanked: 15 times
Been thanked: 2 times

Postby ax » 18 Feb 2007

Here ya go.........enjoy !!


moderator's comments:
this attachment is not the same file as in the first post.
the included functions might overwrite your originals
Attachments
XcapPolyCycle-Predict.xml
(98 KiB) Downloaded 120 times

denizen2
Posts: 125
Joined: 17 Jul 2005

Postby denizen2 » 18 Feb 2007

Here ya go.........enjoy !!


Your file compiles ok, and very quickly, but it is not the same file that I posted :o . BTW, your xml file had several functions and indicators that were not related to the one in question. I imported everything in your file, without thinking :cry: . The result of my importing the 'other' stuff in that file caused almost all of my charts to 'disable' most of the indicators on those charts. Clearly, I should have been more careful. :roll:

So I proceeded first to check if the missing indicators were still listed as 'valid' in the PL Editor. Opps... they were not even listed anymore :!: . What happened to cause this I have no idea. So then I went back to a 'backup' that was a couple of weeks old for the exported versions of 'every thing' being used then. An hour later, after importing and compiling it all again, I am back in business with something that looks like where I was before I started fooling around with that other challenge of a file that wouldn't compile.

I then tried one more time to compile the file that I posted and I notice this time that the "cc1plus.exe" file was being executed, as expected, but after a few minutes that process stopped, and then there was no more processing of any kind by the PL-Editor.... so nothing has changed on that score. I just never noticed before the momentary execution of the C++ compiler, but I presume it happened before too. In the both cases... the compile is never finished, and there is no errors reported... just 'spinning' nowwhere I guess?

So the saga continues :evil: : After terminating the PL-Editor-compiler process (that never returned any errors or status or control back to user), I tried to launch the MC software. The MC-launching process showed this sequence: authorizing, checking for updates, and then "Loading MCActiveX"... but that process would never complete :x . So I next check the process list and see that tsserver is running. Maybe that is ok, or not, I have no idea, but I am trying to start with nothing running, right? So I then 'kill' it, and tried again to launch MC.... this time there is 'success' :)

Please.... I hope TSSupport people are reading this topic. :wink: Clearly, there is something wrong with how all of these different MC program related 'services' are 'coordinated' and 'managed': (a) If the compiler can't finish, then it 'should' report some problem with it and not just continue into the black hole of now-where. (b) The user should NOT be responsible for somehow 'killing' processes that are not in the state expected by the next launching of the MC program, or any of the others. These issues are just 'signs' or 'red flags' that some more attention is needed down there in the 'bowels' of this otherwise great software, right?

I suspect that all of the 'issues' related to supporting the IB data vendor have distracted a lot from these other issues, but I hope that sometime in the near future there will be some extensive effort in just cleaning up the basic windows-processing loops and services that lead to the kind of 'hiccups' described in this topic. :roll:


Return to “MultiCharts”