There are 2 branches of a file. I have to merge from one branch to the other. The automatic merges fails as there are conflicts. The conflicts are due to the date command output stored in file. Can the findmerge tool ignore some con开发者_开发问答flicts using some filter options? I want findmerge to ignore Date: .*
lines and auto merge the rest of file.
As there are many such files, manual merge of all such file having difference of "Date: .*" takes too much time. How can I automate such a merge ?
Date is different in all 3 files, so there is conflict:
file1.txt@@/main/branch1/LATEST
Date: 03/03/2010 11:00PM
Some information1
file1.txt@@/main/branch2/LATEST
Date: 11/11/2009 10:30AM
Some information1
New information2
New information3
Base file: file1.txt@@/main/main/20
Date: 07/07/2005 05:30AM
Some information1
Thanks Deepak
Keyword expansions in ClearCase have been debated before: without the right type manager, it isn't supported.
(Not mentioning the fact is doesn't bring much value in a VCS)
The crux of the issue is that the findmerge algorithm has a case where the actual file content is compared. Unfortunately, findmerge does not use the type manager's compare function, but something hard coded and will think the files are different even though the only difference is in the keywords
You still have in theory a way to develop a type manager, combined with a trigger, as discussed here. This isn't trivial by any mean, so the best solution is to:
- either avoid modification in both branches (the merge will then be trivial for that section)
- or avoid keyword expansion entirely (for instance, a meta data like a date should be associated with the revision date itself, and not as a textual metadata within the data itself)
精彩评论