MantisBT - MEGA
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0001918MEGATree Explorerpublic2020-04-14 03:102023-04-14 19:12
Reporterguest 
Assigned To 
PrioritynormalSeverityminorReproducibilityhave not tried
StatusnewResolutionopen 
PlatformPCOSWindowsOS Version10
Product Version 
Target VersionFixed in Version 
First NameAgnieszka
Last NameJaroszewska
Emailhermisia95@gmail.com
Confirm Emailhermisia95@gmail.com
Summary0001918: error when trying to compress a subtree
DescriptionReport bugs or other issues regarding the MEGA software - when trying to compress a subtree error comes up that says access violation
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Files
Issue History
Date ModifiedUsernameFieldChange
1969-12-31 17:33agaskyNew Issue
1969-12-31 17:33user19Note Added: 0001652
1969-12-31 17:33user19Statusnew => assigned
1969-12-31 17:33user19Assigned To => Nikita Vikhrev
1969-12-31 17:33Nikita VikhrevNote Added: 0001661
1969-12-31 17:33agaskyNote Added: 0001666
1969-12-31 17:33Nikita VikhrevNote Added: 0001673
1969-12-31 17:33agaskyNote Added: 0001676
1969-12-31 17:33user19Statusassigned => resolved
1969-12-31 17:33user19Resolutionopen => fixed
2020-04-14 03:10guestNew Issue
2022-03-08 06:55guestNote Added: 0004833
2023-04-14 19:12guestNote Added: 0006935

Notes
(0001652)
user19   
1969-12-31 17:33   
That's still there? I thought we removed that. It's an error I submitted to Dr.Tamura a while ago but he hasn't gotten back to me yet. You might want to assign this to Dr.Tamura. He might be able to fix it if he has time, it's too complicated for me to fix without breaking other stuff.
(0001661)
Nikita Vikhrev   
1969-12-31 17:33   
I think that the following modification may work.

function TTreeViewForm.DoMolClockTest:double;
var
  value: double;
  df: integer;
begin
  result := 1;
  value := Tree.Value[Tree.TreeIndex] -Tree.Value2[Tree.TreeIndex];
  df := 2*Tree.NoOfOTUs-3 -MLAnalyzer.NoOfClocks -1;
  if (value > 0.000001) and (df > 0) then
    result := GammaQ(df/2, value);
end;
(0001666)
agasky   
1969-12-31 17:33   
You are correct, the change seems to fix the problem. How did you determine to change the cutoff to 0.000001?
(0001673)
Nikita Vikhrev   
1969-12-31 17:33   
Usually, I use 0.00000000001 as a safe minimum value in the double precision. However, I have heard sometime somewhere that the precision is half in transcendental functions like exponential, logarithm, etc. So, I just halved it.

If it works, we had better to fix the GammaQ function itself as follows.

function GammaQ(a, x: Double): Double;
var
  Gln: Double;
begin
  if (x < 0.0) or (a <= 0.0) then
    RaiseErrorMessage(HC_Unexpected_Error, 'Error in GammaQ function');

  if x < 0.000001 then
    result := 1.0
  else if x < (a+1.0) then
    Result := 1.0- GammaSer(a,x, GLn)
  else
    Result := GammaCfn(a,x, GLn);
end;
(0001676)
agasky   
1969-12-31 17:33   
Fix implemented
(0004833)
guest   
2022-03-08 06:55   
https://www.bestwatchaaa.com/Bedat-Co-replica.html [^] Replica https://www.bestwatchss.com/Bvlgari-Replica-Watches.html [^] Jacob https://www.bestwatchaaa.com/Bell-Ross-replica.html [^] https://www.bestwatchaaa.com/Tissot-replica.html [^] https://www.replicawatchsshop.cc/ [^] Co. https://www.bestwatchaaa.com/Michele-replica.html [^] https://www.bestwatchaaa.com/Jacob-Co.-replica.html [^] https://www.swissrepicass.com/ [^] are https://www.bestwatchss.com/Zenith-Replica-Watches.html [^] hottest https://www.allswisswatchs.com/ [^] https://www.watchrm.com/ [^] https://www.bestwatchaaa.com/Piaget-replica.html [^] design https://www.bestwatchss.com/ [^] https://www.bestwatchss.com/Breguet-Replica-Watches.html [^] online website. These https://www.swisswatchesale.com/ [^] https://www.bestwatchaaa.com/Movado-replica.html [^] https://www.newwatchesale.com/ [^] https://www.bestwatchss.com/Franck-Muller-Replica-Watches.html [^] https://www.bestwatchss.com/Chronoswiss-Replica-Watches.html [^] https://www.allshopwatch.com/ [^] are the wholesale https://www.bestwatchaaa.com/Patek-Philippe-replica.html [^] https://www.perfectwatchen.com/ [^] the https://www.allwatchesen.com/ [^] price in https://www.bestwatchss.com/Tudor-Replica-Watches.html [^] world. https://www.bestwatchaaa.com/ [^] https://www.bestwatchss.com/Longines-Replica-Watches.html [^] https://www.goodswisswatch.com/ [^] the quality is similar to https://www.fakeswatchs.com/ [^] real ones.
(0006935)
guest   
2023-04-14 19:12   
https://www.estoolcarbide.com/product/corodrill-800-14d-16d-18d-20d-22d-24d-26d-support-pads-drill-heads-manufacturer-tungsten-carbide-drilling-inserts-for-deep-hole-machining/ [^] [Corodrill 800] https://ryanberger.exblog.jp/ [^] [surface milling inserts] https://edwinboyd.exblog.jp/ [^] [tungsten carbide inserts] https://anguselva.exblog.jp/ [^] [tungsten rods from space] https://brandonmil.exblog.jp/ [^] [carbide wear strips] http://good-time.blog.jp/ [^] [http://good-time.blog.jp/ [^]] https://bobeileen.exblog.jp/ [^] [cast iron inserts] https://williamisi.exblog.jp/ [^] [carbide round rod] http://yyds.blog.jp/ [^] [http://yyds.blog.jp/ [^]]