Ruby 中调试堆栈级别太深
我的 ruby 程序遇到了堆栈级别太深（SystemStackError）错误，以 datamapper 结束：
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/collection.rb:510:in `each'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/comparison.rb:616:in `map'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/comparison.rb:616:in `expected'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/comparison.rb:461:in `matches?'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/operation.rb:498:in `matches?'
from /usr/local/lib/ruby/gems/1.8/gems/extlib-0.9.15/lib/extlib/inflection.rb:103:in `any?'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/operation.rb:159:in `each'
from /usr/local/lib/ruby/1.8/set.rb:195:in `each'
... 5188 levels...
有没有办法调试它？喜欢探索隐藏的 5188 个关卡吗？ Ruby-debug 无法帮助我，内置 ruby 分析器因“[FATAL] 无法分配内存”而终止
I have ruby program that is running into a stack level too deep (SystemStackError) error, ending on datamapper:
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/collection.rb:510:in `each'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/comparison.rb:616:in `map'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/comparison.rb:616:in `expected'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/comparison.rb:461:in `matches?'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/operation.rb:498:in `matches?'
from /usr/local/lib/ruby/gems/1.8/gems/extlib-0.9.15/lib/extlib/inflection.rb:103:in `any?'
from /usr/local/lib/ruby/gems/1.8/gems/dm-core-1.0.0/lib/dm-core/query/conditions/operation.rb:159:in `each'
from /usr/local/lib/ruby/1.8/set.rb:195:in `each'
... 5188 levels...
Is there a way to debug it? Like looking into the hidden 5188 levels? Ruby-debug was unable to help me, and the builtin ruby profiler dies with "[FATAL] failed to allocate memory"
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
除了没有找到可以处理堆栈溢出的 ruby profiler 之外，问题还在于滥用 Datamapper 的惰性
应用程序正在获取文本属性 (这是惰性的）在一个查询（也是惰性的）内，在另一个查询（这也是惰性的） ）。尽管所有这些惰性通常会将 N+1 查询变成 O(k) 查询（实际上通常意味着 4 个查询），但这最终会产生堆栈溢出。
在 rake、rspec 和 netbeans 环境之外以简单的 ruby 脚本执行溢出测试，让我更清楚地看到了罪魁祸首。
仍然希望红宝石分析器能够帮助我。
The problem, besides finding no ruby profiler that can deal with stack overflows, was abusing Datamapper lazyness
The application was getting a text property (which is lazy) inside a query (which is also lazy), inside another query (which is lazy as well). Even though all this lazyness usually turn a N+1 query into a O(k) query (which actually usually means 4 queries), this turned out to generate a stack overflow.
Executing the overflowing test as a simple ruby script, outside rake, rspec and netbeans environment, allowed me to see the culprit more clearly.
Still wished ruby profilers could have helped me with it.