Update custom errors with Exception#full_message

[NeimadTL]

What was the end-user or developer problem that led to this PR?

Not an issue per se but more of an enhancement related to issue #8067

What is your fix for the problem, implemented in this PR?

Make use of Exception#full_message instead of Exception#message so the cause and the backstrace show up.

Make sure the following tasks are checked

• Describe the problem / feature
• Write tests for features and bug fixes
• Write code to solve the problem
• Make sure you follow the current code style and write meaningful commit messages without tags

[NeimadTL]

@deivid-rodriguez , now I wondering whether I should get rid of the "Backtrace" part of the message on L140-124 as well as on L#213-215 ? Thanks for reviewing when you'll be ready.

[deivid-rodriguez]

@NeimadTL Can you show how these errors look now? It seems to me that they are going to start showing redundant information now, because Exception#full_message includes the exception class name and backtrace.

So, yeah, regarding your question, yes, I think Exception#full_message should replace both the class name and the backtrace.

[NeimadTL]

Hey @deivid-rodriguez, thanks for your feedback.

Here's what I've got so far :

According to me (but please correct me if I'm wrong), the ran spec seems to be the correct one to see the message but the thing is, even by changing the the full_message variable (here and here) with something different (say, full_message= 'azerty'), I get the same message as in the screenshot. I'm missing something here, and I don't what it is. A bit of help would be highly appreciated. Thank you.

[deivid-rodriguez]

What debug print are you adding so that running the spec prints that output?

My suggestion was more to try a realworld case. To do this, you can do the following. Create a sample app and Gemfile, for example,

source "https://rubygems.org"

gem "rake"

Configure it to install to a local path, for example, bundle config path foo, and then install it with bundle install.

The installed rake code should be at ./foo/ruby/3.4.0/gems/rake-13.2.1/lib/rake.rb or similar.

Now edit that file to force a GemRequireError, for example, manually add require "something-that-does-not-exist" to the top of that file.

Now you should be able to see the new error by running:

$ ruby -I /path/to/your/clone/of/rubygems/bundler/lib -rbundler -e'Bundler.require'

And you can compare it with the error that your globally installed bundler shows:

$ ruby -rbundler -e'Bundler.require'

[NeimadTL]

Hi @deivid-rodriguez, thanks for your help. Can you explain to me how to do that part please:

Configure it to install to a local path, for example, bundle config path foo