Forums : Ohloh General Discussion
Dear Open Hub Users,
We’re excited to announce that we will be moving the Open Hub Forum to
https://community.blackduck.com/s/black-duck-open-hub.
Beginning immediately, users can head over,
register,
get technical help and discuss issue pertinent to the Open Hub. Registered users can also subscribe to Open Hub announcements here.
On May 1, 2020, we will be freezing https://www.openhub.net/forums and users will not be able to create new discussions. If you have any questions and concerns, please email us at
[email protected]
Rules for analysing SVN repos
How does Ohloh analyse SVN repositories? Does it care about the trunk/branches/tags structure or does it create just combined statistics?
My issue is that I need to add the documentation of my code as a bunch of HTML files into the repository yet I don't want it to be scanned by Ohloh. Does this mean that I have to redirect Ohloh from path/to/my/repo to path/to/my/repo/trunk or will is scan trunk/ and branches/ only anyway? Do I have to create a dummy account for updating the html pages in the repo?
As far as I know, for SVN Ohloh ignores any possible semantics of the structure, leaving that to you, which means it'll scan whatever path you give it entirely.
I remember reading on some related page to only register the project's trunk to save them some resources. So, once you do that, I think you can safely place anything you wish off the trunk and it won't be counted.
Why don't you want your docs counted? IMHO the docs are an important part of a software project, and lines of docs are sometimes more important for consumers than actual lines of code.
Ohloh doesn't pay any attention to your directory structure. Although trunk/branches/tags is pretty common, a large number of projects are (dis-)organized differently.
Ohloh's Subversion importer is pretty simple, and it doesn't understand branching at all. We follow only a single line of development, and that line of development cannot change its directory or URL.
There's no hard rule about what Ohloh will or won't allow you to import -- you can give us an URL to any directory, and we will run a report on it. However, giving Ohloh the tags directory or a very full branches directory is cruel. While a tag can be created with a single command in Subversion, it creates an entire virtual clone of the project at that point, which is a serious work burden for our brute-force importer.
[edit -- When you initially give an URL to Ohloh, we do an initial simple check to see if you are using the trunk/branches/tags structure, and will try to avoid including the branches and tags if we can. This is just to protect us from a casual mistake that sends one of our servers off the deep end for a week.]
Also, if you want to have an accurate count of the lines of code, you should only give Ohloh a single copy of your code. If you include branches and tags in the URL you give to Ohloh, you will have redundant copies of the code in your report.
Someday we hope to change all this. In the future, we'd like to process the entire repository to get the full activity history, so that we can show all commits on all branches. Then you can simply tell us which directory represents the current trunk for purposes of getting a total current line count. However, this kind of processing is pretty far down our development timeline at this point.
Robin
When you merge the branch back into the trunk, that will be seen by ohloh, but the individual commits on the branch will not be seen by ohloh.
There are a whole range of changes which people seem to be screaming for louder than this, so don't count on it.
But then I'm not an ohloh insider, so they might be about to release it.
