ISSUE-3572 try fix stack overflow

[zhang2014]

I hereby agree to the terms of the CLA available at: https://databend.rs/dev/policies/cla/

Summary

fix stack overflow.

Changelog

• Bug Fix

Related Issues

#3572 Partial fixes(SQL parser may still stack overflow)

Test Plan

Unit Tests

Stateless Tests

[databend-bot]

Thanks for the contribution!
I have applied any labels matching special text in your PR Changelog.

Please review the labels and make any necessary changes.

[vercel]

This pull request is being automatically deployed with Vercel (learn more).
To see the status of your deployment, click below or on the icon next to each commit.

🔍 Inspect: https://vercel.com/databend/databend/8jMkZZ7y1JCYj8k5F5d3kW3JK5UL
✅ Preview: https://databend-git-fork-zhang2014-fix-issue-3572-databend.vercel.app

[Deployment for 4447d08 canceled]

[sundy-li]

Need a test case to cover the issue.

[xudong963]

I have tried to use maybe_grow() to solve stack overflow in another database, but it can't completely solve the problem, only to prevent panic or limit users' behavior. I think your idea can fully solve stack overflow, nice work!

[xudong963]

If I read correctly, you try to avoid stack overflow by manually replacing the stack with the heap. You are essentially simulating a stack behavior. Maybe we can fully avoid recursing by using iterating? And VecDeque may be a good structure.

[zhang2014]

Maybe we can fully avoid recursing by using iterating? And VecDeque may be a good structure.

We want to support both preorder traversal(pre_visit) and postorder traversal(post_visit), which makes me think of the stack first, maybe you have a better idea, please let me know. : D

[mergify]

This pull request has merge conflicts that must be resolved before it can be merged. @zhang2014 please rebase it 🙏

[zhyass]

Need resolve the conflicts.

[zhang2014]

Need resolve the conflicts.

Let's convert PR to draft first and I will continue to this work until it's ready.
BTW: resolve conflicts is hard(because the FunctionFactory::get has some different).

[zhang2014]

Need a test case to cover the issue.
Need resolve the conflicts.

@zhyass @sundy-li Done.

[codecov-commenter]

Codecov Report

Merging #3603 (4447d08) into main (b712e6d) will increase coverage by 0%.
The diff coverage is 70%.

@@          Coverage Diff           @@
##            main   #3603    +/-   ##
======================================
  Coverage     60%     60%            
======================================
  Files        697     705     +8     
  Lines      37230   37787   +557     
======================================
+ Hits       22367   22709   +342     
- Misses     14863   15078   +215     

Impacted Files	Coverage Δ
common/planners/src/plan_rewriter.rs	39% <50%> (-4%)	⬇️
common/planners/src/plan_expression_rewriter.rs	63% <60%> (ø)
common/planners/src/plan_expression_common.rs	63% <78%> (+7%)	⬆️
query/src/optimizers/optimizer_constant_folding.rs	85% <81%> (-2%)	⬇️
common/planners/src/plan_expression.rs	81% <100%> (-2%)	⬇️
common/planners/src/plan_expression_chain.rs	68% <100%> (+3%)	⬆️
common/planners/src/plan_expression_visitor.rs	100% <100%> (+14%)	⬆️
.../functions/src/aggregates/aggregate_arg_min_max.rs	58% <0%> (-12%)	⬇️
common/functions/src/aggregates/aggregate_sum.rs	85% <0%> (-4%)	⬇️
...ns/src/aggregates/aggregate_combinator_distinct.rs	68% <0%> (-4%)	⬇️
... and 34 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0ef869a...4447d08. Read the comment docs.

[databend-bot]

Wait for another reviewer approval

[databend-bot]

CI Passed
Reviewers Approved
Let's Merge
Thank you for the PR @zhang2014