[WONTFIX] nimpretty infers wrong indentation

[timotheecour]

proc foo_verylong_verylong_verylong_verylong_verylong_verylong_verylong(a1 = 1,
    a2 = "asdfasdf", a3 = "asdfasdf"): auto =
  discard

nimpretty indents as:

proc foo_verylong_verylong_verylong_verylong_verylong_verylong_verylong(
        a1 = 1,
    a2 = "asdfasdf", a3 = "asdfasdf"): auto =
    discard

(ie, it wrongly inferred indentation of 4, whereas it was 2)

EDIT: s/margin/indentation/ (was a typo)

[Araq]

It doesn't infer it at all, that's why. ;-)

[timotheecour]

it does:
this diff:

diff --git a/compiler/layouter.nim b/compiler/layouter.nim
index 9f9b6fc8e..fe6125d7e 100644
--- a/compiler/layouter.nim
+++ b/compiler/layouter.nim
@@ -43,6 +43,7 @@ proc openEmitter*(em: var Emitter, cache: IdentCache;
   let fullPath = Absolutefile config.toFullPath(fileIdx)
   em.indWidth = getIndentWidth(fileIdx, llStreamOpen(fullPath, fmRead),
                                cache, config)
+  echo ("em.indWidth:", em.indWidth)
   if em.indWidth == 0: em.indWidth = 2
   em.config = config
   em.fid = fileIdx

produces this output:

(Field0: "em.indWidth:", Field1: 4)
so getIndentWidth infers 4, when it should've inferred 2 in this example; somehow it got "tricked" by the argument reflow

a not-very-robust way to infer in this example is to take the 1st argument reflow (line 2: 4 spaces) and divide by 2; however that's only ok if it was obtained by a previous invocation of nimpretty;
a more robust way is to infer by ignoring argument reflow and infer it from 1st indentation it finds after that (line 3=>2 spaces)

[Araq]

It infers the indentation, not the "margin".

[timotheecour]

duh... was a typo; i meant indentation; the bug still holds despite my typo :)

[dom96]

Why close this? It seems like a genuine bug

[timotheecour]

@dom96 because of this comment from @narimiran #10629 (comment)

Btw, sorry to be blunt, but I will not consider/review your PRs until the count of your open issues is above 150. Close some least important ones (which affect the least amount of Nim users), this will show some good will from your side, and then I'll change my mind.

I went through all my open issues and they're all valid (IMO), but I had to reduce to < 150 to avoid this stalemate.

[krux02]

@timotheecour Please show some respect. You are supposed to reduce your issues to set priorities on them. They are not all valid. I am sure I can find some that are invalid or written in a very obscure way that I have no idea what the real problem is. But the point is, you should do it and don't continue to waste our time on it. Only keep the important and issues open and close the issues that are not important. Then it would be really nice of you, when from the other 144 issues if you would check if they conform the issues template and rework them to make them easier processable. You still have 144 issues from 1256 issues in total. That is more that 10% from you alone.