Discussions
Categories
Groups
Community Home
Categories
INTERNAL ENABLEMENT
POPULAR
PUBLIC CLOUD
PRIVATE CLOUD
Quick Links
MY LINKS
HELPFUL TIPS
Back to website
Home
Intelligence (Analytics)
Parallel development and merging rptlibraries and rptdesigns.
Mithun Bs
We have an issue while merging the changes done in rptdesign and
rptlibraries.
If the file has been changed in two branches and finally during merge,
it is almost impossible for us to do the merge manually.
For example, many column ids get generated in both files, if it doesn't
match with IDs generated in both files we will have conflicts.
We are using Perforce to maintain these files.
It shows line by line diff.
Is there any efficient methods for merging of the files.
Also sometimes, the xml code gets re-shuffled(through BIRT designer)
even if there is small change. Even this will impact us during resolve
of conflicts.
Find more posts tagged with
Comments
mwilliams
Hi Mithun Bs,
If you use an element out of a library in a report, it should change if you change the element in the library. The library element doesn't change if you change the report element. I guess maybe I'm not sure exactly what you're asking.
Mithun Bs
<blockquote class='ipsBlockquote' data-author="'mwilliams'" data-cid="70983" data-time="1291327529" data-date="02 December 2010 - 03:05 PM"><p>
Hi Mithun Bs,<br />
<br />
If you use an element out of a library in a report, it should change if you change the element in the library. The library element doesn't change if you change the report element. I guess maybe I'm not sure exactly what you're asking.<br /></p></blockquote>
<br />
<br />
Hi mwilliams,<br />
<br />
The issue is merging an rptlibrary.<br />
For example consider, there is a dataset in a rptlibrary.<br />
Two people want to change something in that dataset parallely.(The same rptlibrary will be in different branches) (we use perforce to store and manage the rptlibrary)<br />
If the first person adds a new column (in middle of many others) in the SQL query in dataset like,<br />
<br />
"select a,b,d,e,f from table" has been changed to<br />
"select a,b,c,d,e,f from table". <br />
<br />
In this case the xml source of rptlibrary will have lot of changes with respect to column IDs etc.<br />
<br />
Parallely in other branch, one more person adds similarly a new column.<br />
<br />
Finally we will merge both the changes so that we will have functionality introduced by both of them.<br />
While merging perforce will do line based diff and merge by default(not xml diff). But since in this rptlibrary we have many changes it will ask manual merge which is cumbersome.( and will lead to invalid XML most of times)
mwilliams
I think I see now. You're running into problems when two people make changes to the library at the same time. Your storage program doesn't have a feature that only allows one person to have write access to a file at one time to avoid this issue altogether?
Mithun Bs
<blockquote class='ipsBlockquote' data-author="'mwilliams'" data-cid="71110" data-time="1291738471" data-date="07 December 2010 - 09:14 AM"><p>
I think I see now. You're running into problems when two people make changes to the library at the same time. Your storage program doesn't have a feature that only allows one person to have write access to a file at one time to avoid this issue altogether?<br /></p></blockquote>
<br />
No, we don't want to restrict it to be done only by one person.<br />
Because when there are two/three releases at a time, We create two branches from main altogether.<br />
Finally both of them gets merged into main(at least during this time, there will be having changes in same dataset in both branches and while merging to main we will have resolve conflicts).