bptato

> The auto detect should be an option in the options panel, so that a user does not have to use the command line option in order to enable it....

> Well, adding a command line option should at least update the man page > and probably also the MANUAL.html. Wait what? I thought we decided we don't need a...

When I wrote "We can live without that for now", I meant it. It isn't that relevant with just two queries, especially when one of those is just a fallback....

I've tested the latest version: * Piping to stdin works now, thanks. * It seems initImage gets called twice for some reason, so detection is performed twice too :/ I...

Ah, no, by "support" I meant "don't print an error and stop looping" :P SyncTERM (at least the version I have, from a few months ago) doesn't do private color...

> afaict activeImage is only set to TRUE once and never set back to FALSE, so if w3m is started and initImage called without auto-detection, and the user then enables...

QJS itself is fairly complete. Most significant omission I've found so far is that it has no global proxies, so WindowProxy may be impossible to implement. I haven't tried too...

Reduced to ```nim # nim r --mm:arc -d:danger -d:lto --cc:clang --nimcache:test z.nim 1 1 proc c_atoi(s: cstring): cint {.importc: "atoi", header: "".} var cmdLine {.importc: "cmdLine".}: cstringArray proc main() =...

Update: I haven't been able to reproduce the crash of the original program with gcc. I must have somehow enabled strict aliasing with gcc too when I first reported this...

I don't see why constness would matter. The core problem is that for `seq[T]`, ARC generates code like this: ```c struct tySequence__qwqHTkRvwhrRyENtudHQ7g { NI len; tySequence__qwqHTkRvwhrRyENtudHQ7g_Content* p; }; struct tySequence__qwqHTkRvwhrRyENtudHQ7g_Content...