I am writing my first real gem. It depends on three gems, and those gems have roughly a dozen dependencies. I'd like this all to be self-contained so my gem will always use the dependencies from inside the gem, and will not require the other gems to be installed at the system level. Doable?
I played around with bundle install --path vendor --standalone. It did indeed store all the gems' source code into vendor, but I could then not get my 'requires' to work correctly and the code could not be found.
I suspect either that it's not possible to do this with a gem (only and app) or I was missing some piece of configuration to point my gem's code at the vendored dependencies. I played around with $LOADPATH but could not make it work.
Anyone know if this can be done and, if so, how to make it work? Pointers to info welcome.
I think you need to add those gems as a dependency to your .gemspec file. That way when your gem gets installed, its dependencies will also be installed.
More information can be found here
Related
I currently have a large gem installed that I only use part of. So, a few questions:
Is is possible to only install the part I need?
If so, does that make more sense then installing the whole thing and only requireing the part I need?
It is only possible to install the gem in its entirety, i.e. everything declared in the gemspec.
If so, does that make more sense then installing the whole thing and
only requireing the part I need?
Only requiring what you need is exactly what you're looking for. If the gem is declared in your Gemfile like this:
gem "your_gem", require: false
Then, it won't be automatically required in your app and therefore, won't be loaded into memory, which seems to be the effect you want. Then, just require only the parts you need.
Maybe someone knows if this sort of RubyGems plugin exists already, before I try to sink a lot of time into writing one myself.
I'm not using RVM* or Bundler (edit: see thread in comments), so I have a big ball of system gems installed. I want to sort out which ones are gems I really want to use, which ones are dependencies that have to be there for the gems I really want to use, and which ones are just junk that can be cleaned up.
If anyone is familiar with Gentoo, I'm thinking of something similar to Gentoo's package management, but for RubyGems instead--a way to tag a small list of gems I really want to keep, then run a command that can go through all of my gems and clean up the ones that aren't dependencies of those gems. A "#world" set for RubyGems, in other words.
*RVM isn't an option for me anyway, because I have to use Microsoft Windows.
If you install graphviz, then you can runbundle viz this will produce a dot notation mapping of the dependencies and you can use that to trace dependencies down to those gem's you actually need to have installed (but only those that you have installed via bundler).
bundle viz --format=dot
I am trying to build a gem for a project that has a dependency on an unnamed external gem :)
During development I found a small bug in the external project and I added a one line fix that resolves it. I submitted a pull request on github, but I have no response from the maintainer for some time now.
I want to make my project available as a gem but it wont work without this fix. What can I do?
What would be the best way to fix this.
One option I thought about was to create a gem of the forked project and publish it under a convoluted name, and make a dependency on that. But I don't like the idea of polluting the servers with such a stupid gem.
So I was wondering if it is possible to bundle the external gem into my application, and make it install together with my gem. What would be the cleanest and easiest way to do this?
Have you considered overriding the function with your own code? I was having a similar problem with some software at work a few weeks ago and I just redefined the function.
Since it was just one line you found, it seems like this would be the easiest solution, but I am a little bit new to Ruby so maybe there is a problem with this plan that I have not considered.
You could publish it under a different name and once the upstream maintainer accepted your fix, you can yank your version.
It's quite simple, in fact. In your Gemfile add the dependency as:
gem "nokogiri", :git => "git://github.com/tenderlove/nokogiri.git"
To do this you would also need to be using bundler to manage your gem, you can get more info on this here.
The other option is to add the code you changed to a vendor directory in your gem and distribute it with your code, this way you can just add the main directory of this other gem to your load path and you will be able to require it without any issues.
To add something to the load path you simply do:
$LOAD_PATH.unshift( File.join(File.dirname(__FILE__), '..', 'vendor', 'some_gem', 'lib') )
And you will be able to directly require files at some_gem.
I want to install a gem via gem install, but I need it to resolve with dependencies of the current project.
Basically I want the functionality that bundler gives me when I specify gem 'xyz' in a Gemfile, but I don't want to add that specific gem into the Gemfile.
I tried doing bundle exec gem install ... but it doesn't seem to work.
edit: The reason why I don't want to add it to the Gemfile is that it might be something like metric_fu, metrical, saikuro, rails_best_practices, etc. Simply gems that are kind of utility use and might only cluttler the project.
I might only want to use them temporarily, or install them, try out, if it doesn't work out the way I want do rvm gemset empty and bundle install again to clean up.
The point of Bundler is, in part, to prevent you from doing things like that (to prevent you from injecting gems from outside when your project doesn't declare them).
Looking for a way of doing that is looking for a bug in Bundler. If you did manage to find some way of skirting Bundler's enforcement mechanisms, you should probably not use it; instead, you may consider filing it as a bug with Bundler's issue tracker.
Now we come to the real questions: what can you do? and what should you do?
You should use either RVM gemsets or Bundler to isolate your application and its gem dependencies. You don't need both. I would recommend Bundler for this purpose over RVM gemsets.
You should add to your Gemfile any gems that you want to use and that integrate with your application (i.e., that either load your application or that are loaded as part of your application). This is not a requirement for any gems that refrain from integrating with your application.
You should refrain from committing a changed Gemfile or Gemfile.lock to version control until you are satisfied that your application continues to operate acceptably (tests pass, new gem does something useful, etc.).
Or you should stop using Bundler, because you want to do things it is explicitly designed to prevent you from doing (not recommended).
At the risk of sounding dumb, why not add it to the gemfile? You can always add it to its own group if you don't want to have to install it everywhere.
A slightly different approach is, if you're using version control, such as Git, to create a new branch and install the gems. If it doesn't work out, uninstall the gem (I'm not sure this will be done by bundle update on the old branch) and trash the branch. If it does, work, merge your stuff into the old the branch.
Though I do believe the other answers and comments have some very good points.
Would it be possible to override the default "require" by adding automatic download-and-install code for any missing includes (provided that the missing include is published as a ruby gem).
Which would not work in situations where Ruby is not interfaced with a shell. But still I think it would be an interesting idea.
Is there such a mechanism in existence today?
Edit:Removed portion about password check. I just checked and gem install doesn't seem to require me to type my password.
You would be able to hijack require method so as gems are installed when an attempt is made to require them, but still you won't have access to newly installed gem in current process, because gem index has to be reloaded.
I understand the intentions but I think exercise might not be worth it.
When installing a fresh gem the gem will be installed in the GEM_HOME. If that is not writable then it will try in the user's home .gem directory (on *NIX at least).
You could certainly script this. In a way Rail's rake gems:build is just this. Just not on demand.
But, I would recommend against this. You could run into build, versioning, dependency and network issues. And probably security issues as well.
PS: Francis Hwang did something related a while ago, although only as a require, not a require gems.
http://fhwang.net/2005/11/01/urirequire-I-got-yer-Web-2-0-right-here
A better option would be to use bundler and distribute the required gems with the application.
It is also quite simple to write a script to bootstrap the installation of gems if you didn't
want to distribute them with your code (using the bundle install/check commands)