1. EM Classic tool (eccomagic.com)  helps when corruption caught too late for EM-ADT to repair.   (if you would have used period test with EM-ADT on copy of file,  would have caught corruption before became embedded in file structure).
 
 
A> use EM ADT to extract all rules. (may not be possible for you in very corrupted state)
 
B> can use special EM Classic copy to copy folder structure,
 
C> then paste into new file folder view
 
D> manually need to adjust folder types, special properties, and insert (cut and
paste) folder rules.
 
E> Manually reconstruct views.
 
F> then use EM classic tool to move view by view bulk items.  retains item RTF formatting, etc.
 
 
 
2. Let's see if we can figure what is causing the error.  So,  make some copies of the file.  Then try:
 
0> Close all your views.  (does not delete any data, and is easy place to start).
 
A> Delete all the folders you can.   Now can you run ADT ?   (if yes,  delete less folders, try again.. until isolate the crashing folder).  
 
B> Delete all your items.  Now can run ADT ?  (if yes delete less items, try again... until isolate the problematic items).
 
C> ie., see if possible to isolate what in file is stopping ADT from being able to perform basic functions.
 
 

Just ran another test:
 
1. I deleted all the views, data and folders (that could be deleted from a copy my original file).  
2. This file was previously posted as TestLUAError3_corrupt_file.eco i the Yahoo forum, and I believe that you downloaded it.
3. When I run the EM-ADT analyis on this file, I still got the application errors in Ecco.
 
4. I then tried to recover this "deleted" file (File/Database/Recover)
5. Interestingly, the recovered file did not have the Application errors when running the EM-ADT analysis.
(In contrast, the original file could not be recovered; It still showed Application Errors on F5 after recovery)
 
I am not sure how to use this information though.... Apparently one or more of the items or folders in the original file made it impossible to repair/recover the file. If I somehow could figure out what it is, I could remove it, recover the file and then ad back the item(s) or folder(s)..
It would be a lot of work to try to delete various folders & items and keep retrying, as I have hundreds of folders and roughly 10-100,000 items..

Quote from rb on 01/09/10 at 04:21:28:

 
I will look at your file if helpful to you.
 
also,  not seem possible to have 100,000 items as 65k max is possible.
 
 
what do your EM Memory reporter numbers look like ??
 
 
in slang,  what is the highest count you have for any folder ??
 
 

well you've got 34,426 items.
 
you have 2 items per text folder value (or so).
 
 
 
WHAT FOLDER TYPE HAS THE 8000+ items ???
 

This is a response to a post in the Yahoo forum.
http://tech.groups.yahoo.com/group/ecco_pro/message/6644
 
> You have an old file that is not corrupted ?  Yes as I understand it. depending on level of corruption in newer file,  it may be able to sync the two files,  updating old with changes made to new.
 
How do I synch two independent Ecco files?
 
> it may be able also to remove the corruption if we can isolate it.
TAKE A COPY OF YOUR NON-CORRUPT FILE.  PLAY WITH THE SLANG EXTENSION DRAGGING & DROPPING SUB ITEMS,  CONTEXT FREE,  APPLY YOUR RULES, ETC. CAN YOU GET YOUR OLD FILE TO CORRUPT THIS WAY ??
IF YOU CAN FIGURE OUT WHAT CAUSED CORRUPTION,  MAY BE SIMPLE TO FIX FOR YOU.
 
I have not been able to  
 
> other than that, delete groups of folders at a time and retest with EMADT until you find folder/folders which are causing the errors on F5 file analysis. that is not really so difficult to do.
 
After I have deleted all data and folders that could be deleted, I still get the error on F5. (However, as explained above, the deleted file can be recovered using Ecco).  Can you figure out what is wrong with the real file based on the corruption in the deleted file?
 
 
> your deleted version is clearly folded into itself.  What are the memory allocations vs. file size with that version ?
>  
That is the same file as you downloaded:  
Memory allocation is 649,832 while size on disk is 2,263,650. That is 1,613,818 "extra" bytes and similar to the difference in the non-deleted file (1,499,898 bytes). Maybe the data that can not be deleted is what causes the corruption. If so, how could I delete this data in the “real” file?
 
 
> you can try to use slang regex replace to replace all [a-zA-Z0-9] to X.
>  
 Sorry for my poor RegEx. What would the complete expression be for this?
>  
 
> you just use standard export and import.  is very simple.  does not transfer some data such as recurring dates, alarms, etc.  just specify all folders
 
 How? The export dialog only exports everything in a Notepad. There is no way to simply export all the items in a file.
 
We really need a utility for easy and complete export/import of data between Ecco files….

I am sorry, but I still do not understand how to sync the two files.  
 
I know how to keep files synchronized if I am working with a synchronized copy that generates a change file.
However, this is not the case here.
 
Can you please explain step-by-step, or point me to some resource?  
 
p.s. I will post this in the Yahoo forum as well, as it would be of great interest to learn how to do this.
 

It is worth a shot... Can I try it?
 
In this case, I would use the original uncorrupted file as a "base" and import the changes from the new file.
 
However, as an alternative, couldn't I also then just use a new "empty" file as the base instead.....?

1. if you delete items and not recover,  does EM-ADT run ?  (with shift F5 any different ?)
 
2. create a clean file.  make some items.  copy multi times as sub. copy multii times etc.  you can quickly fill in a new file.  Test with EMADT.  Work good ?

 
1. Well there is something wrong with the ecco file other than a corruption issue that the Tool is not picking up on.
 
 
2. "*" indicates the pointer to which the free space information applies.
 
 
 
One possibility is a defect in the default.ect template.   Which updated template was used to create ?
 
 
If you do same thing with raw (shift + new)  file...  do we get same result ? [ie., to see if problem is in default.ect]
 
 
 
 

 
try the shift new.  It will tell us much.
 
 
There could be a problem with the default.ect  
 
   if you don't have problems with a clean shift+new  file...  we have located the source of the issue.
 
 
try Shift + F5  for test.   does that work for you ?
 
Remember,   EM-ADT  destructive test is testing for corruption.   Corruption will cause data loss.  There can be other non-corruption issues with file but will not result in data loss.