summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Extract override version for use installing in devindirect/override-version-for-devAndre Arko2017-08-073-14/+12
|
* Auto merge of #5922 - bundler:colby/fix-new-init-env-var, r=indirectThe Bundler Bot2017-08-041-1/+1
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | fix documented init gems rb config env var ### What was the end-user problem that led to this PR? I documented the env var for the init new gem file feature flag incorrectly ### What was your diagnosis of the problem? β€Ί BUNDLE_NEW_GEMFILE_NAME=1 dbundler init Writing new Gemfile to /Users/c/Desktop/Projects/testapp/Gemfile does not generate the gems.rb as expected ### What is your fix for the problem, implemented in this PR? Use the correct env name to generate gems.rb
| * fix documented init gems rb config env varcolby/fix-new-init-env-varColby Swandale2017-08-041-1/+1
| |
* | Auto merge of #5921 - arbonap:pa-triage-strategies, r=colby-swandaleThe Bundler Bot2017-08-042-0/+9
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add Bug Triage Strategies 🐞 Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - I asked about how to go about triaging bugs in bundler. To what extent should contributor's env try to match the user's env? ### What was your diagnosis of the problem? My diagnosis was... - To include more strategies of how to triage bugs ### What is your fix for the problem, implemented in this PR? My fix... - I added some documentation to synthesize the response's that I got from the Bundler Slack group on how people go about triaging. Specifically, I asked how contributors go about incorporating the user's environment into their own when debugging. ### Why did you choose this fix out of the possible options? I chose this fix because... - Hopefully this additional documentation can help squash more bugs πŸ•· - Let me know what y'all think, always happy to incorporate changes.
| * | Add Bug Triage StrategiesPatricia Arbona2017-08-032-0/+9
|/ /
* | Auto merge of #5916 - arbonap:pa-fix-broken-link, r=colby-swandaleThe Bundler Bot2017-08-031-1/+1
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Update broken link in bundle-gem man page Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - This resolves [Issue#296](https://github.com/bundler/bundler-site/issues/296) on bundler/bundler-site. ### What was your diagnosis of the problem? My diagnosis was... - @olleolleolle came across a broken link in the `bundle gem` docs. Specifically, the `bundle-config` link in the "See Also" section was broken. ### What is your fix for the problem, implemented in this PR? My fix... - I fixed the broken link πŸ’ ### Why did you choose this fix out of the possible options? I chose this fix because... - To make navigating around the bundler docs a pleasant experience.
| * | Update broken link in bundle-gem man pagePatricia Arbona2017-08-021-1/+1
| | |
* | | Auto merge of #5919 - koic:fix_template_of_gemsrb, r=colby-swandaleThe Bundler Bot2017-08-031-1/+1
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the tempalte of gems.rb ### What was the end-user problem that led to this PR? Building from source using `rake bundler_2:install`. (#5901) ```console % bundle version 2.0.0.dev (2017-08-02 commit f184adfc1)% cd /tmp % bundle init Writing new gems.rb to /private/tmp/gems.rb % cat gems.rb # frozen_string_literal: true # A sample gems.rb source "https://rubygems.org" git_source(:github) {|repo_name| "https://github.com/#{repo_name}" } # gems "rails" ``` Editing the comment out of `# gems "rails"` and executing it resulted in an error. ```console % bundle install [!] There was an error parsing `gems.rb`: Undefined local variable or method `gems' for Gemfile. Bundler cannot continue. # from /private/tmp/gems.rb:8 # ------------------------------------------- # > gems "rails" # ------------------------------------------- ``` ### What was your diagnosis of the problem? [Like the Gemfile](https://github.com/bundler/bundler/blob/c4ae85f01edca72d48c617be07440c2cd600ab4f/lib/bundler/templates/Gemfile#L7), use `gem` method. ### What is your fix for the problem, implemented in this PR? Use `gem` method instead of `gems` method.
| * | | Fix the tempalte of gems.rbKoichi ITO2017-08-031-1/+1
| | | |
* | | | Auto merge of #5917 - arbonap:pa-doc-specs, r=colby-swandaleThe Bundler Bot2017-08-031-2/+2
|\ \ \ \ | |_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Include `bin/rspec` in directions to test doc changes Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - I was updating documentation in a `.ronn` file. I struggled a bit when I tried testing my doc changes. When I tried running `rspec ./spec/commands/help_spec.rb` and `rspec ./spec/quality_spec.rb`, I kept getting this error: ``` uninitialized constant Bundler::EnvironmentPreserver::BUNDLER_PREFIX (NameError) Did you mean? Bundler::BundlerError ``` ### What was your diagnosis of the problem? My diagnosis was... - @colby-swandale told me that I should run`bin/rspec` instead of `rspec`. It worked 🎈 ### What is your fix for the problem, implemented in this PR? My fix... - To add in `bin/rspec` to the ["Writing docs for man pages" documentation](https://github.com/bundler/bundler/blob/master/doc/documentation/WRITING.md) ### Why did you choose this fix out of the possible options? I chose this fix because... - Hopefully this could help someone else save some time on their hands and prevent them from getting the same error I did.[[selfie-0 uploading...]] [[selfie-1 uploading...]] [[selfie-2 uploading...]]
| * | | Include `bin/rspec` in directions to test doc changesPatricia Arbona2017-08-021-2/+2
|/ / /
* | | Auto merge of #5911 - arbonap:pa-setup-docs, r=segiddinsThe Bundler Bot2017-08-021-0/+11
|\ \ \ | |/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Explain RUBYOPT in Debugging section of Setup docs Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - I was looking through `SETUP.md` and I didn't understand why I had to set the `RUBYOPT` env variable to `-rpry dbundle`. I was wondering why I couldn't just normally use `require 'pry'; binding.pry` to debug Bundler's codebase. ### What was your diagnosis of the problem? My diagnosis was... - To add more explanations to the Setup documentation. ### What is your fix for the problem, implemented in this PR? My fix... - I asked @indirect and @jules2689 all about `RUBYOPT`, and why we have to set `RUBYOPT` to require pry and require this path to Bundler. I used their wonderful explanations and tried synthesizing their knowledge in a way that made sense to those who aren't familiar with `RUBYOPT`. I also included background context as to why Bundler chooses to do this. ### Why did you choose this fix out of the possible options? I chose this fix because... - Having background context of how and why Bundler combines `RUBYOPT` with `dbundle` in the Setup docs will help new contributors ramp-up. - I also think having this knowledge will (hopefully) help new contributors not put in `require 'pry'; binding.pry` like they may be used to in other Ruby projects. ✨ Let me know what y'all think of this first draft πŸ‘ŒπŸΌ. I'm happy to incorporate changes.
| * | Explain RUBYOPT in Debugging section of Setup docsPatricia Arbona2017-08-011-0/+11
|/ /
* | Auto merge of #5902 - bundler:seg-molinillo-0.6, r=segiddinsThe Bundler Bot2017-08-0129-390/+830
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Update to Molinillo 0.6.0 ### What was the end-user problem that led to this PR? The problem was Molinillo 0.5.x was so old :P ### What was your diagnosis of the problem? My diagnosis was we needed to _fix every single known bug_, while also improving performance. ### What is your fix for the problem, implemented in this PR? My fix is to behold the awesomeness of https://github.com/CocoaPods/Molinillo/releases/0.6.0, and introduce it to Bundler.
| * | Update vendored Molinillo to 0.6.1seg-molinillo-0.6Samuel Giddins2017-08-012-15/+3
| | |
| * | Update vendored Molinillo to 0.6.0Samuel Giddins2017-08-0121-161/+629
| | |
| * | Update prerelease logic for new molinilloSamuel Giddins2017-08-012-8/+4
| | |
| * | Update conservative update specs for being able to revert to the locked ↡Samuel Giddins2017-08-011-8/+4
| | | | | | | | | | | | resolution
| * | Use compound matchers in resolving specSamuel Giddins2017-08-012-4/+3
| | |
| * | [DepProxy] Allow == nilSamuel Giddins2017-08-011-0/+1
| | |
| * | Update the resolving spec for debug output having changedSamuel Giddins2017-08-011-2/+4
| | |
| * | [Resolver] Use Molinillo’s built in conflict message with treesSamuel Giddins2017-08-011-91/+68
| | |
| * | Refactor SpecGroup to not subclass ArraySamuel Giddins2017-08-013-102/+113
| | |
| * | Refactor the GVP specs to not assume spec groups are arraysSamuel Giddins2017-08-011-15/+17
| | |
* | | Auto merge of #5909 - koic:print_only_version_number, r=segiddinsThe Bundler Bot2017-08-016-5/+34
|\ \ \ | |/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [2.0] Print only version number from `bundler --version` ### What was the end-user problem that led to this PR? (Fix #4706) The behavior was different from [Bundler 2 RFC](https://github.com/bundler/rfcs/pull/6/files#diff-0579edc0374571613cd01f0a42c19bdaR38). ### What was your diagnosis of the problem? When it is `bundler_2_mode`, it behaves as written in RFC. ### What is your fix for the problem, implemented in this PR? This PR will switch `Bundler.feature_flag` using `bundler_2_mode`. ### Why did you choose this fix out of the possible options? Since #4708 does not use `Bundler.feature_flag`, then compatibility with Bundler 1 will be lost. In addition, this PR adds specs for Bundler 2.
| * | [2.0] Print only version number from `bundler --version`Koichi ITO2017-08-016-5/+34
| |/ | | | | | | Fix https://github.com/bundler/bundler/issues/4706.
* | Auto merge of #5900 - bundler:seg-env-eval-gemfile, r=colby-swandaleThe Bundler Bot2017-08-016-10/+86
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [Env] Print all gemfiles that have been evaled ### What was the end-user problem that led to this PR? The problem was we'd get error reports like https://github.com/bundler/bundler/issues/5381 where the env output was insufficient to reproduce the issue because `eval_gemfile` had been used. ### What was your diagnosis of the problem? My diagnosis was we needed to print out all gem files that had gone through `eval_gemfile` ### What is your fix for the problem, implemented in this PR? My fix keeps track of all eval'ed gem files and prints them all out in the env report.
| * | [Env] Print all gemfiles that have been evaledseg-env-eval-gemfileSamuel Giddins2017-07-276-10/+86
| | |
* | | Auto merge of #5907 - ↡The Bundler Bot2017-07-311-0/+1
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | roberts1000:fix-magic-comment-style-for-gemspec-template, r=colby-swandale Add new line after newgem gemspec template ### What was the end-user problem that led to this PR? When using `bundle gem` to create a new gem, the created gemspec file starts with ``` # coding: utf-8 lib = File.expand_path('../lib', __FILE__) ``` Rubocop recommends having a blank line after magic comments (like `# coding: utf-8`). ### What was your diagnosis of the problem? The rubocop style guide recommends having a blank line after "magic comments" ([see link](https://github.com/bbatsov/ruby-style-guide#separate-magic-comments-from-code)). The main bundler.gemspec follows this recommendation, but the .gemspec, that is created for new gems, does not. ### What is your fix for the problem, implemented in this PR? Update the template file to have a blank line after the magic comment. ### Why did you choose this fix out of the possible options? It's what rubocop wants.
| * | | Add new line after new gem gemspec templateRobert Soly2017-07-311-0/+1
| | |/ | |/|
* | | Auto merge of #5903 - arbonap:pa-delete-aliases, r=segiddinsThe Bundler Bot2017-07-311-2/+2
|\ \ \ | |/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Delete redundant aliases from bundle outdated command πŸ’« Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - I was looking at the `bundle outdated` command in `cli.rb`, and noticed some flag aliases were the same as the original flags. ### What was your diagnosis of the problem? My diagnosis was... - I thought these aliases weren't necessary ### What is your fix for the problem, implemented in this PR? My fix... - I deleted two aliases from `bundle outdated`. - I ran `rake spec:deps && bin/rspec spec/commands/outdated_spec.rb` locally and all tests passed. ### Why did you choose this fix out of the possible options? I chose this fix because... - to DRY up `cli.rb` πŸ‘ŒπŸΌ
| * | Delete redundant aliases from bundle outdated commandPatricia Arbona2017-07-271-2/+2
| |/
* | Auto merge of #5906 - arbonap:pa-add-aliases, r=colby-swandaleThe Bundler Bot2017-07-286-10/+14
|\ \ | |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add missing flag aliases to docs 🌟 Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - I noticed in the documentation that a lot of aliases for flags were missing. ### What was your diagnosis of the problem? My diagnosis was... - I added missing aliases for flags into the documentation. ### What is your fix for the problem, implemented in this PR? My fix... - I added the missing flag aliases into the docs, so that users can more easily access this information. πŸ’– - I also added in a line describing how RSpec is the default testing framework ### Why did you choose this fix out of the possible options? I chose this fix because... - more bundler knowledge for bundler users is helpful πŸ’
| * Add missing flag aliases to docsPatricia Arbona2017-07-286-10/+14
|/
* Auto merge of #5901 - bundler:seg-bundler-2-tasks, r=indirectThe Bundler Bot2017-07-262-0/+18
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add tasks to make running in 2.0 mode easier ### What was the end-user problem that led to this PR? The problem was it was too difficult to run Bundler in 2.0 mode. ### What was your diagnosis of the problem? My diagnosis was we needed a way to install a dev version of 2.0, and also to run commands as bundler 2. ### What is your fix for the problem, implemented in this PR? My fix adds a `bin/bundle2` bin stub that runs bundler as 2.0.0.dev, and a `bundle_2:install` rake task that will install `bundler-2.0.0.dev`
| * Add a task to install a bundler 2 dev gemseg-bundler-2-tasksSamuel Giddins2017-07-261-0/+10
| |
| * Add bundle2 binstubSamuel Giddins2017-07-261-0/+8
|/
* Auto merge of #5896 - bundler:seg-globally-cache-git-repos, r=indirectThe Bundler Bot2017-07-254-15/+21
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [Source::Git] Cache repos globally ### What was the end-user problem that led to this PR? Closes https://github.com/bundler/bundler/issues/5895. The problem was: > Now that we have a "global" user-wide cache of gems, it would be great to also have a user-wide place to keep git repos. In 1.x, the default system path provided that user-wide cache, and losing it because of the default 2.0 path will slow down installs with git repos by a lot. ### What was your diagnosis of the problem? My diagnosis was we needed to change the git cache location to something global ### What is your fix for the problem, implemented in this PR? My fix uses the global "~/.bundle/cache" directory to cache git gems ### Why did you choose this fix out of the possible options? I chose this fix because it means the cache (which only contains _downloaded_ information, nothing evaluated at runtime) can be shared across every app and every ruby version
| * [Source::Git] Cache repos globallyseg-globally-cache-git-reposSamuel Giddins2017-07-254-15/+21
| |
* | Auto merge of #5892 - arbonap:pa-gemfile-doc-changes, r=colby-swandaleThe Bundler Bot2017-07-251-0/+31
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add more background explanations to bundle-install and gemfile docs Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... - I thought the `bundle-install` and `gemfile` docs could have more background explanations. ### What was your diagnosis of the problem? My diagnosis was... - Given more clarifying background information regarding some Ruby-specific domain knowledge, this can help those newer to the Ruby community. ♦️ ### What is your fix for the problem, implemented in this PR? My fix... - I gave a general overview of various Ruby implementations, and explained what git submodules and gemspecs are. ### Why did you choose this fix out of the possible options? I chose this fix because... - @rubymorillo and @indirect gave me the go-ahead to give more background information about these topics πŸ‘πŸ½ - please let me know what y'all think, I'm happy to incorporate edits
| * | Add bundle-install and gemfile background explanationsPatricia Arbona2017-07-231-0/+31
| | |
* | | Auto merge of #5897 - bundler:colby/gem-empty-name-validation, r=segiddinsThe Bundler Bot2017-07-253-0/+13
|\ \ \ | |_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add clear error message when using a gem in the Gemfile with an empty name ### What was the end-user problem that led to this PR? Users are allowed to add a gem to the Gemfile with an empty string as the name but Bundler won't find a gem with an empty and does not make this clear as possible to the end user. Using a Gemfile like: ``` source 'https://rubygems.org' gem '', git: 'https://github.com/rack/rack' ``` Bundler will print the following: ``` β€Ί dbundler install Fetching https://github.com/rack/rack Fetching gem metadata from https://rubygems.org/. Fetching gem metadata from https://rubygems.org/........... Fetching gem metadata from https://rubygems.org/... Could not find gem '' in https://github.com/rack/rack (at master@8693330). The source does not contain any versions of '' ``` ### What is your fix for the problem, implemented in this PR? Check the name of the gem in the DSL. If the name is empty print a clear message to the user that using an empty name is not allowed ``` [!] There was an error parsing `Gemfile`: an empty gem name is not valid. Bundler cannot continue. # from /Users/c/Desktop/Projects/testapp/Gemfile:2 # ------------------------------------------- # source 'https://rubygems.org' > gem '', git: 'https://github.com/rack/rack' # ------------------------------------------- ```
| * | add more clear error message for user when adding a gem in the Gemfile with ↡colby/gem-empty-name-validationColby Swandale2017-07-253-0/+13
|/ / | | | | | | an empty name
* | Auto merge of #5860 - bundler:seg-validate-settings, r=segiddinsThe Bundler Bot2017-07-2546-187/+518
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce the notion of settings validation & default install to ./.bundle Thanks so much for the contribution! To make reviewing this PR a bit easier, please fill out answers to the following questions. ### What was the end-user problem that led to this PR? The problem was... ### What was your diagnosis of the problem? My diagnosis was... ### What is your fix for the problem, implemented in this PR? My fix... ### Why did you choose this fix out of the possible options? I chose this fix because...
| * | Ensure sudo specs are run against system gems for consistencyseg-validate-settingsSamuel Giddins2017-07-241-0/+3
| | |
| * | [Validator] Inspect array of symbols for 1.8.7 compatibilitySamuel Giddins2017-07-241-1/+1
| | |
| * | Untangle the system gem path and the default bundle path in the specsSamuel Giddins2017-07-2423-85/+123
| | |
| * | fixup! Add unit tests for the settings validatorSamuel Giddins2017-07-231-2/+2
| | |
| * | Hash#to_h does not exist on old rubiesSamuel Giddins2017-07-231-1/+1
| | |
| * | Add unit tests for the settings validatorSamuel Giddins2017-07-233-6/+117
| | |