Results 75 issues of Roman Leventov

My export resulted in a lot of errors like the following: ``` Unknown markup type 10 { type: 10, start: 317, end: 336 } Unknown markup type 10 { type:...

bug

When exporting a story with links that start with bold, and then bold continues after the end of the link, importer generates Markdown formatting in wrong order, that is, ```...

Overriding `Object.finalize()` method is very bad for GC. Instead, `sun.misc.Cleaner`, or intermediate that automatically detects the presence of `java.lang.ref.Cleaner` and uses it if present (something like this: https://github.com/OpenHFT/Chronicle-Core/blob/067e963a0b79775d9dee33cb13b5fa5b7e49683e/src/main/java/net/openhft/chronicle/core/cleaner/impl/reflect/ReflectionBasedByteBufferCleanerService.java, for example)...

help wanted

I think there could be a section with code reviewing checklists. - https://github.com/isocpp/CppCoreGuidelines - https://github.com/code-review-checklists/java-concurrency These two that know about. https://www.infoq.com/articles/API-Design-Joshua-Bloch can also be considered an API design checklist. Perhaps...

```julia using Interpolations, Zygote itp = interpolate(([0.0,1.0],), [1.0, 2.0], Gridded(Linear())) f(c) = itp[c] f'(0.5) ERROR: ArgumentError: invalid index: 0.5 of type Float64 Stacktrace: [1] to_index(::Float64) at ./indices.jl:297 [2] to_index(::Array{Float64,1}, ::Float64)...

needs adjoint

It's useful for implementing LRU caches. To avoid blowing the number of methods, a boolean flag could be added to automatically "move to first" on any writes.

### Description When one clicks on the "link-chain" button to the left of any header in a Markdown document, Github just navigates to this page section (the URL in browser...

enhancement
help wanted
small

https://fluence.network/ says: > Current software model favors data and value extraction, so the largest platforms earn the highest returns, hire top developers and perpetuate dominance. Fluence turns competition into collaboration...

recent

```julia function loss1(p) f(x,p,t) = 1 prob = DiscreteProblem(f, 0, (1,10), p) sol = solve(prob, FunctionMap(scale_by_time = true), saveat=[1,2,3]) return sum(sol) end DiffEqFlux.sciml_train(loss1,[1],ADAM(0.05),maxiters = 20) ``` Error in line https://github.com/SciML/DiffEqSensitivity.jl/blob/c31e529853c2c45cab583e03b08c24af99c50c2a/src/adjoint_common.jl#L41:...

It seems that API for working with array slices is missing, that makes users to write inefficient code like the following: https://github.com/FasterXML/jackson-datatypes-collections/blob/c374783ff962041d003125cbd7b617bab5d16bff/eclipse-collections/src/main/java/com/fasterxml/jackson/datatype/eclipsecollections/deser/BaseCollectionDeserializer.java#L199-L218