Hello Atlassian community,
I am hoping that there is something I am missing, because I really enjoy the use of Fisheye when it falls in line with my process. That being said, I am having a lot of trouble combining our process (SVN usage) with my Fisheye installation. Our general structure is
RootUrl/
RootUrl/Application1/tags/1.0
RootUrl/Application1/tags/1.1
RootUrl/Application1/trunk
RootUrl/Application1/branches
...
RootUrl/Application100/tags
RootUrl/Application100/trunk
RootUrl/Application100/branches
While some projects are larger, we have about about 100 applications, most of which are distinct and have less than the order of 50 files. We have about 22,000 SVN revisions. So an average project has around 200 revisions. Most of the time, there is between 1 and 2 developers commiting changes to trunks within their application. Once complete, tags are created from (usually Tortoise SVN gui) branch/tag, taking the code from trunk and copying it into a tag. To my understanding, this is the correct way to use SVN. To put a possible method of reviewing code, I will provide an example. It allows me to perform manual code reviews by selecting tag 1.0 and selecting tag 1.1 and seeing the changes from a directory, then asking for a diff on the modified files.
I do not immediately see this functionality in Fisheye. This is not necessarily a bad thing, because I believe (from using http://sandbox.fisheye.atlassian.com/browse/rails) that it seems like the functionality I desire can happen. It just isn't happening for me, yet.
What it seems like Fisheye wants me to do is go through the revisions (or changesets) committed to trunk between tags, and then compare them. However, depending on how/when people have been creating their tags/trunk commits, etc. there could be pages and pages of changesets to mark; the ability to select a range of changesets under a given application would help me, but I do not see this option.
Next, it seems like Fisheye is not picking up my tags, branch and truck properly, as the search seems to fail when it should not. What I mean by this is when browsing repositories, it seems to only pick up some tags (~1% of them). To my knowledge nothing was done differently in these cases, but these are old revisions/tags from before I begun working with this company.
Lastly, Fisheye crashes (404 error) when trying to search on username. The only thing I can think of is that in our SVN, we use DOMAIN/username when specifying users. However, this search functionality has not worked properly yet. This is not a particularly big issue, because if our SVN-Fisheye combination would work properly, I would not need this functionality anyway.
Any advice with respect to our use of SVN or Fisheye would help. We are using Fisheye version 2.6.1 evaluation. Ideally if we can get this to work, we would like to integrate it with our Jira and Confluence systems, but until I can successfully nail the process down, it will have to wait or we will have to look elsewhere. It seems like it is a great product, so I hope there is something I can adjust to use it successfully.
Thanks