|
I am creating a DOORS database script that will be used to get all differences between 2 versions of a particular DOORS table (new vs old). I want to get all differences between the 2 tables (additions, deletions and changes). For anyone who doesn't know about DOORS, when an object is removed from a DB, the program marks the item's attribute as deleted, but you can still see it when you loop through all items. What I want to do is create a function that will display any items that have been changed since the older version.
Here is the algorithm I'm using:
for each newItem in newVers
if (newItem.deleted) {
check if deleted in oldVers
if (!oldVers.deleted) {
}
} else {
if (newItem.id exists in oldVers) {
if (newVers != oldVers) {
}
} else {
}
}
}
Does this look like the way I should do this, or is there a better way? Just to reiterate, I am only looking for what, in the newer version has changed since the older one. I am not worried about changes in the opposite direction.
Any ideas would be appreciated.
Chris
modified 26-Sep-16 9:02am.
|
|
|
|
|
Your approach is accurate, but could be slow. I'm guessing that most of the items are unchanged, so
if (newItem.id exists in oldVers) could be an expensive operation, particularly if id is not indexed.
If your database supports fast sequential access on that field, then there are algorithms that "walk" the old and new, much the way text editors do file differencing.
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
I have yet another issue with this function that I didn't think of originally... There are a bunch of attributes that the user can choose from to check for changes and I want to list those attributes, if they've changed, in a new column. However, since the text and heading are attributes that we always check first, I want to skip over those attributes since they will always be in the list. What is the best way to ignore those 2 attributes in the list? My first thought was to include an if-statement within the loop to look for those items. Is this the best option, or is there another way that I'm not thinking of?
Chris
|
|
|
|
|
This is part of my last post here, but I made it much smaller in question.
I'm trying to write a function that will compress items measured by inches.
So say; I have 10 of these
Length = 72
Width = 4
Height = 3
And I package 6 of them into this, leaving the next 4 to be in another package.
Length = 72
Width = 24
Height = 3
I want to further compress them by decreasing the width and increasing the height such as this. I took 2 of them and stacked them taller increasing the height.
Length = 72
Width = 12
Height = 6
I have this, which works in 1 pass, but I'm trying to figure out a way to make multiple passes until there are no more passes to make. But I can't seem to figure out a method in which to do this.
If (item.Width > item.Height) Then
If (item.Qty > 2) Then
item.Width -= p.Width * 3
item.Height += p.Height
End If
End If
|
|
|
|
|
Debug the programme.
I do believe that below line is causing that "it works only in 1 pass" 
If (item.Width > item.Height) Then
|
|
|
|
|
Well... i'm pretty sure, you have to rethink your concept of programme...
Imagine, you have a box (container):
Height (3)
_ _ _ _ _ _
/| |
/ | |
/ |_ _ _ _ _ _|
/ / /
/ / /
| / /
| / / Length (72)
|/_ _ _ _ _ _/
Width (24)
In that container 6 smaller boxes(containers {72x4x3}) are there. You want to repack these boxes into container with changed width and height:
Height (6)
_ _ _ _
/| |
/ | |
/ | |
/ | |
/ | |
| |_ _ _ _|
| / /
| / /
| / /
| / / Length (72)
|/_ _ _ _/
Width (12)
Assuming, that you want to find out, if there's possibility to repack these innercontainers, your algorithm have to check, if:
- any dimension of innercontainer does not exceed the size of outercontainer,
- occupied area inside a box (sum of innercontainers areas) is smaller than outercontainer.
Till above condition is met, you can add another box (innercontainer)!
Note, that:
Area (of box) = Width * Length * Height<br />
Occupied area = sum of innercontainers areas.
And finall note: do not change the size of outer-box, create new one and repack all boxes
Try!
|
|
|
|
|
I thought about that.
It would be easier to say will this fit in the space rather than try to create the space.
Not sure If I mentioned this, but the point of the program is to just get a rate from UPS or Fedex by creating virtual packages of dimensional weight, gravity weight, and to not create large package conditions by making them smaller.
Maciej Los wrote: Area (of box) = Width * Length * Height
That's dimensional weight, yes I could just sum up the dWeight till I hit a target.
So If I were to sum up say
3 x 3 x 3 = 27
and I have 10 of them for 270
then 3 x 3 x 30 should equal 270 and equals 270. hmm ...
My question was to adjust the size of 3 x 3 x 30 to 15 x 6 x 3 then try 1 more time.
I should just hire someone to write this for me.
|
|
|
|
|
I should just hire someone to write this for me.
Jim, never give up! There's a ton of members ready to help you (at least me).
Seems, you forgot to use loop!
If (item.Qty>2) Then
Do While (item.Width > item.Height)
item.Width -= p.Width * 3
item.Height += p.Height;
Loop
End If
Am i right?
modified 28-Sep-16 2:03am.
|
|
|
|
|
I have no clue how to paste in my Linked In link, tried many times.
I actually have that sample code above working better now.
If your curious about the program ....
This is the code packaged in a console app in VS2015
There is a class called sample data, that you can remove the remarks to load extra data
In the main class, you can unmark data to load.
It outputs packages in the console. In the real program, those packages will be submitted to UPS for a rate quote, and be packaged in the packages XML section and looped around for a single multi-package quote.
The rules are long items will be shipped seperate
Isolated items are items that will ship in the Original box, slap a sticker on them
Common items are the rest of the stuff that will be tossed in a box
If items trigger the Large Package Indicator, then start a new package to save the customer money.
Ar last, see if we can further combine packages to keep the count down.
Dropbox - RTST Rev 2016092701.zip[^]
I'm not trying to get someone to do my work for me, I'm just having trouble wrapping my head around this.
I use to just add up all the mass until a package is created, then cube the mass into a square. But that failed for long items from 6FT to 8Ft long.
And it failed for isolated items as well.
Thin items are tough, I was doing thin separate, but have now combined them into common.
[edit]
This is a clean copy of the program, in which I started again today. I put the thin items back in and will try to streamline the packaging, see Rev5 class comments.
Dropbox - RTSR_Console 3.zip[^]
modified 28-Sep-16 2:03am.
|
|
|
|
|
|
Thank you for looking at it. I'm working on it today with some new ideas.
|
|
|
|
|
After writing it about 12 times, I came up with this. Works pretty well, but could be better.
Theirs a program called combine_virtual_packages_common that could use better refining.
I was thinking, perhaps I should treat a batch of items like water. Take a box that is 24x17x14 and calculate the item line as water level, and fill the box, say 1 or 2 inches at a time.
Dropbox - RTSR Console 2016093001.zip[^]
|
|
|
|
|
Sorry, Jim, i've been bit busy...
I've downloaded the latest version of RTSR Console. As it turned out i can't install .net framework 4.6.1 on my laptop (win10). After that, i tried to find combine_virtual_packages_common function, but i've failed. I've found only combine_virtual_Common.
Can you accept my invitation on LinkedIn? Then we'll be able to continue discussion in private.
Cheers
Maciej
|
|
|
|
|
The later is the correct name
combine_virtual _common
Sorry about the .net issue, and yes I'm running Win 10 and VS2015
Yes I can accept the invite, just need to receive it first.
|
|
|
|
|
Strange,
In my LinkedIn profile i see: "Invitation pending". I'll try to send you a message via InMail (on LinkedIn).
Maciej
|
|
|
|
|
It was pending in my queue, just accepted
|
|
|
|
|
Hi
I have a textfile and need to build a algorithm join lines
The lines have 14 characters that can be
1,2,3,4,5,6,or 7
Example
44444444444444
This numbers represent bit representation,
1= 001
2=010
3=011
4=100
5=101
6=110
7=111
For to join line I must to observe if two (or more) line have 1 or 2 characters differente then to XOR between characters
Example
Line 01=> 11122444444444
line 02==>12142444444444
First character each line is 1 zero different
Second character first line is 1 and second is 2 then 1 different
Same to 4th character (2 and 4), The others are equal.
due to having only two differences I DO a XOR bits
Second character
1 xor 2 ==> 3
001 xor 010 ==> 011
4th character
2 xor 4 ==> 6
010 or 100 ==> 110
I need build best algorithm to join lines with rules
1)I can have more than two lines being used
2) Each line can have a maximum 1 or 2 , character 7 (parameterized) after join
3) each line used in another must, can not be used more
Somebody know how can I to build a recursive algorithm ?
Tia
|
|
|
|
|
By specifying that you need the "best" algorithm, you have severely limited any response.
Oh, and that's "Octal" you're fiddling with.
|
|
|
|
|
Thank you,
better perhaps put examples
81 LINES
11111111111111
11121111111111
11141111111111
11211111111111
11221111111111
11241111111111
11411111111111
11421111111111
11441111111111
12111111111111
12121111111111
12141111111111
12211111111111
12221111111111
12241111111111
12411111111111
12421111111111
12441111111111
14111111111111
14121111111111
14141111111111
14211111111111
14221111111111
14241111111111
14411111111111
14421111111111
14441111111111
21111111111111
21121111111111
21141111111111
21211111111111
21221111111111
21241111111111
21411111111111
21421111111111
21441111111111
22111111111111
22121111111111
22141111111111
22211111111111
22221111111111
22241111111111
22411111111111
22421111111111
22441111111111
24111111111111
24121111111111
24141111111111
24211111111111
24221111111111
24241111111111
24411111111111
24421111111111
24441111111111
41111111111111
41121111111111
41141111111111
41211111111111
41221111111111
41241111111111
41411111111111
41421111111111
41441111111111
42111111111111
42121111111111
42141111111111
42211111111111
42221111111111
42241111111111
42411111111111
42421111111111
42441111111111
44111111111111
44121111111111
44141111111111
44211111111111
44221111111111
44241111111111
44411111111111
44421111111111
44441111111111
after joining
77771111111111
Example 3 first lines
11111111111111
11121111111111
11141111111111
Therea only difference in 3th character
001 xor 010 ==> 011
and there only a different in 3th charcter in 3th line then
011 xor 100 ==> 111 => 7
|
|
|
|
|
There's no recursion required; just a "for" loop.
XOR the first and second "lines"; then XOR that result with the 3rd; and so forth.
Maybe not the "best", but "easy" to understand and implement; and you're only dealing with 81 "lines".
|
|
|
|
|
Thank You
My Post was a trivial example , but can have somee restrictions
Example:
3,5 and 6 can show only 2 times
or
7 with 3
or
7 with 5
or
7 with 6
show only one time in file
example
75444444444444 is possible
75344444444444 no is possible because 7 is together 5 and 3
35644444444444 no possible 3,5 and 6
35444444444444 is possible
etc
My problem is way to let
smaller number of possible lines
|
|
|
|
|
If you need to "filter" your "list of lines" use LINQ.
You can remove items; count items; etc.
e.g.
... = MyList.Where( line => line.contains() || line.StartsWith() || etc. )
|
|
|
|
|
Hello to All!
I am performing a small academic research regarding distributed storage systems and some practical problems that are connected with this. One problem is fail-over and primary-backup replication algorithms. Some DSS use consensus algorithms to elect leader for replication like Raft and Paxos (Multi-Paxos). So my question is:
How do these algorithms address replication issues in storage systems? What are advantages and disadvantages to use Paxos or Raft? E.g.:
1. Raft has following advantages:
2. Raft has following limits:
3. Paxos has following advantages:
4. Paxos has following limits:
Can anyone point me on what basic pros and cons are for these two algorithms in using for storage systems?
What is the practical usage of both of them in DSS?
Thanks in advance,
Nick.
|
|
|
|
|
|
Thank you Richard, I know these resources. But mostly information there is about internals of algorithms and how they work. By now I need to see a brief comparison of limits of each for the purpose of building a fault tolerant system using consensus.
So maybe more simple question for people who already familiar with both
protocols: what do like and dislike in each other? I need information not like "it's more complex" or "it's more popular" but better their limits in particular purpose (storage fault tolerance). Why one prefer Paxos or other likes Raft?
Regards,
Nick.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
