Skip to content

Latest commit

 

History

History
99 lines (64 loc) · 3.29 KB

README.md

File metadata and controls

99 lines (64 loc) · 3.29 KB

Boosting the declarative part

The declarative DB query is so much more than just the Contracts object!

To be precise, it is:

  1. Contracts
  2. Functions
  3. Modifiers
  4. Instructions
  5. StateVariables

Each has many methods, and more importantly, you can combine them in the same query, filtering out unnecessary entities and improving the subsequent arbitrary logic part.

Example

Take a look at this glide from the examples on the website:

from glider import *


def query():

    instructions = (
        Functions()
        .without_properties([MethodProp.HAS_MODIFIERS])
        .instructions()
        .with_callee_function_name("selfdestruct")
        .exec(100)
    )

    return instructions

It steps down from Functions to Instructions and uses both. In the first step, it gets rid of the functions with any modifiers, and in the second step, it queries all the instructions containing selfdestruct call from the remaining functions.

Up-n-Down

Think of the combinations as layers where the contracts are the highest level and instructions are the lowest. You can jump between these layers however you like, one at a time. The final jump will determine what you'll get as a result.

All the possible transitions (pay attention to the directions):

flowchart TD
    Contracts <--> Functions
    Contracts -.-> StateVariables
    Contracts -.-> Modifiers

    Functions <--> Instructions
    Functions <-.-> Modifiers

    Modifiers <--> Instructions
    Modifiers --> Contracts
Loading

*Weak dash links mean that you can't transit directly but need to loop and go through each single instance. For example, you can get StateVariables from a Contract instance. Similarly, you can get Modifiers from a Function and Functions from a Modifier.

E.g., if you add .functions() to the glide above, you'll get List[Function], where each is without modifiers and contains selfdestruct as one of their instructions.

from glider import *


def query():

    functions = (
        Functions()
        .without_properties([MethodProp.HAS_MODIFIERS])
        .instructions()
        .with_callee_function_name("selfdestruct")
        .functions()
        .exec(100)
    )

    return functions

Optimization notes

I have a few tips for you when it comes to the performance during combinations:

  1. Try to avoid stepping up because it requires much more time (e.g., Functions -> Contracts).

  2. If you need to step up, try to wrap your query like a sandwich, even if you don't need to use any methods from an upper layer. For example, if you need to do some logic with Instructions but want to get functions as a result, then:

    1. Initiate the query with Functions
    2. Step down to Instructions
    3. Perform your operations
    4. Get back to Functions (like in the glide above)

    This trick will cut out all instructions not in functions beforehand (like instructions from modifiers); thus, your operations will be performed on fewer instances.

  3. It's also useful to think about the size of the data you operate with and the complexity of operations. It's better to filter out as much as you can with simple operations before getting to complex ones like filtration by function signature.

Mix and experiment!

Read next: Debug Technique