I am working with ClearCase and I am very new to the tool. I find it confusing and cumbersome compared to other source control, like git with svn, for example. I was wondering if someone could walk through a typical workflow. So right now we have a main integration stream, then a child stream on that, then from there child streams for each project.
Use Case 1:
I check in a file and want everyone to get changes. I have to first add to ClearCase or check in, then I have to deliver to each stream above me (twice), then I would create a baseline and do a rebase to cascade the changes back down. Is this correct?
Use Case 2:
If I forget what changes I made, is there a mechanism to see all the files I modified? and check them all in at once.
Use Case 3:
If I want to modify a file, do I always have to check it out?
Use Case 4:
If I move a file, must I remove it from clear case and then re-add it? Or what about if I just want to rename a package or file?
Use Case 5:
Right now I have a dynamic view on the main integration stream, then child views in between. If I am just working on one stream, is it necessary to have views on all the other streams?
I am just trying to use clear case effectively, but it seems to accomplish simple tasks, there is a lot of overhead. Can anyone give any tips or advice on how to simplify the experience?
Thanks
You can see the main differences between Git and ClearCase here, to get you started.
Use Case 1:
Yes, which is why it is advised to use one stream if possible, instead of the native "one stream per developer" model (that I denounced here)
Use Case 2:
You can use some commands based on lsco | ci, as in this answer.
I detail that approach in "Recursive checkin using Clearcase"
ct lsco -r -cvi -fmt "ci -nc \"%n\"\n" | ct
ct lsco -r -cvi -fmt "unco -rm %n\n" | ct
Use Case 3:
Yes, you need to checkout a file, before modification and checkin, unless you modify it directly in a snapshot view, as an hijacked file; or in a dynamic view as an eclipsed file.
See for instance "Eclipse not allowing to modify files without checkout in ClearCase" as an illustration of that issue.
Use Case 4:
cleartool mv will do that for you (as I mentioned here), but you need to checkout the parent folder first.
The directory where the element to be moved or renamed resides must be checked out. The destination directory must also be checked out; this directory may be the same as the source directory.
Use Case 5:
If I am just working on one stream, is it necessary to have views on all the other streams?
No, those views don't have to be started or even created.
Related
In our continuous integration we use ClearCase UCM. We occasionally need to access the sources from the recommended baseline (which not necessarily equals the newest baseline). Note: All baselines are full.
I figured out how to access the sources belonging to the recommend baseline manually, by entering cleartool edcs in the command line within the dynamic view and adding the line element * MyRecommendedBaseline below # Select checked out versions, saving and closing the text-file.
Unfortunately I don't know how to do this from a script. One way I could think about is read the contents of the config_spec into a stream, add the line, save it to a new text file and use "cleartool setcs newcs.txt".
But apart from being cumbersome I'm not exactly sure if this is possible. Anyone knows a simple way to do this?
You can script listing the recommended baselines of a given stream: see "How can I list the recommended base line in ClearCase"
On Windows:
cleartool descr -fmt "%[rec_bls]CXp" stream:streamName#\aPVob
On Unix:
cleartool descr -fmt "%[rec_bls]CXp" stream:streamName#/vobs/aPVob
From there, you can generate a new file with simple rules:
element * MyRecommendedBaseline1
element * MyRecommendedBaseline2
...
And you can setcs that file to a dedicate base-CC dynamic view (not an existing UCM view).
As Brian Cowan points out in the comments, this only works because those baselines, as the OP mentions, are full baselines (not incremental or not-labeled, like deliverbl are).
See also "What is the difference between Full baseline and Incremental baseline in Clearcase UCM?".
I want to merge all files from a particular branch to main using a script or a command. Is there a way to do it without checking out each file in target or should i do each file manually.
ClearCase is file-based, not repository-based: any merge will be done file by file, and actually first folder by folder (you merge the folders first, then the files).
The easiest way to initiate a merge based on a branch is to use the cleartool findmerge, which can use a version selector, like -fve/rsion .../branch1/LATEST.
See also "To prepare to merge"
The usual approach though is to use a view or a tag to select the elements you want to merge, as I described in "How merge sub branch to main branch using clearcase command line under linux?" (using -ftag).
Note that this works also in an UCM environment, based on activities (even though the deliver and rebase commands remain the recommended merge methods).
In my current working environment, I make use of eclipsed files extensively for testing purposes. When it comes time to formalize things, I have a script which generates a diff by comparing the file with a backup saved by my eclipse script.
I'd like to be able to work without the backup of the original file; is there any way I can retrieve the current version of the file that I can pass to diff? Even though the file is view-private, I see I can enter foo##/ and see a list of versions, but I'm not sure how to find which version is the latest. Everything I've tried using cleartool ls or describe with the file name tells me that it's not a VOB object (which is true, although cleartool ls does show it as eclipsed, so it must know, somehow, that there is an element there)
Thanks
Eclipsed file means dynamic view.
The simplest solution would to make a second dynamic view based on the same config spec.
Considering how cheap and quick those views are, this isn't an issue.
On that second dynamic view, you can do a
cleartool descr -fmt "%Xn" /path/to/element
In order to get the extended pathname of the file (see fmt_ccase for more on the %Xn syntax).
I currently have a jacked-up delivery from a child stream to a parent in ClearCase. If I try to undo the delivery it tells me I can't because the "integration activity has checkins" or "checked in versions".
If I try to resume the delivery it says it encountered an error attempting to checkout or merge an element, but doesn't specifically tell me which one.
So I'm looking for a way to either:
Manually stop the delivery (undo all checkouts in the parent stream?)
Find out what element is causing the delivery problem (is it same as the one causing the undo problem)
or Find out what element is causing the undo problem and find a way to undo the checkin (I don't know how to do this. I tried to delete a version in the version tree, but I don't have permission).
For 3/ "Find out what element is causing the undo problem", this is easy (but not recommended): you need to remove all the checked-in versions done during the complete phase of the deliver.
And that is by far the most dangerous solution, especially if any type of activity (other checkins, baselines, ...) has been done on the destination Stream (the stream to which you are delivering file, ie the Stream with the view you are using to deliver to)
You can see those checked-in file by describing the deliver activity (which always starts with deliverbl.xxx)
cleartool descr -l activity:deliverbl.xxx#\myPVob
1/ and 2/ are linked.
A good solution to easily detect the issue is to resume the deliver graphically: open the ClearCase project Explorer (clearprojexp), right-click on the source Stream and select deliver (Baseline or Activities, to default or alternate target: it doesn't matter).
ClearCase will detect that a deliver is in progress and will propose to resume.
All you need to do is check all the files with a red circle and white cross (not the files with a yellow warning sign, those are not blocking the deliver).
Once you have one of those files, right-click on it, and select "display element merge": you will have a more precise error message that you can copy-paste.
If those files are in lost+found directory, all you need to do is to edit the config spec of the view used for the deliver, and add a non-selection rule to avoid selecting anything from lost+found:
cd /path/to/your/view
cleartool edcs
#add at the start of the config spec
element /myVob/lost+found/... -none
Then resume again your deliver, and you will see that those 'lost+found' files become ignored (with a warning non-blocking status attached to them).
If those files aren't in lost+found and are failing the deliver because of "Not a vob object <directory name>", the first check to do is to go to the parent directory of said files in a shell session and type cleartool ls: you will see their status.
In this case, the OP Ian reports them as hijacked, so it was simply about undoing their hijacked status.
He reports also having to delete (rmname) some binary files, although my answer to the question "Clearcase UCM is trying to merge pdf files" is pointing to an alternative solution (copy merge).
My recommendation: in that particular state (deliver with checkins already there), try hard to complete the deliver, not to cancel it.
In Clearcase, I want to copy (fork, split) a file while preserving its history. Something like svn cp old.txt new.txt. How do I do it?
It isn't possible do fork a file in ClearCase.
If you refactor your code and split a file in two, one of them will appear as a new file and you will loose the information about who coded it. The annotate command will say the author of the lines are who splited it.
UCM or not, you cannot duplicate easily the full history of a file.
The best way to isolate an history is still to create a branch in order to make new versions to that file without impacting the same file in the original branch.
Thinking 'svn cp' should be available in ClearCase might come from the fact that, in SVN, branches are directories, and a tool like cc2svn will actually replicate ClearCase branches using 'svn cp'.
But since, with ClearCase, branches are first-class citizen, it is best to reason in term of branch than in term of copy/fork.
From the main page of cc2svn:
There is a difference in creating the branches in ClearCase and SVN:
SVN copies all files from parent branch to the target like: svn cp branches/main branches/dev_branch
ClearCase creates the actual branch for file upon checkout operation only.
Pretty simply done
Check out parent folder
Move element you wish to duplicate to appropriate location (not within the checked out parent folder)
Undo Checkout of parent folder
All the files get returned to the original folder with history and also the duplicate ones remain in the new location with the history too. Now each file can be checked out and changed individually