Currently PrintfTarget emits a trailing \0. See https://dxr.mozilla.org/mozilla-central/rev/b06968288cff469814bf830aa90f1c84da490f61/mozglue/misc/Printf.cpp#884 However this is only really of use to SprintfState; other users don't want this. It seems better to emit to not emit this, and to have SprintfState deal with its own needs.

Comment on attachment 8842199 [details] Bug 1341880 - do not emit trailing \0 in PrintfState; https://reviewboard.mozilla.org/r/115862/#review118044 ::: mozglue/misc/Printf.h:123 (Diff revision 1) > ~SprintfState() { > this->free_(mBase); > } > > + bool vprint(const char* format, va_list ap_list) { > + return mozilla::PrintfTarget::vprint(format, ap_list) && append("", 1); I can see bystanders raising eyebrows on the append("", 1) (which is perfectly correct, just non obvious to casual readers ; especially, I can see someone asking themselves where that comes from, look at blame, see the corresponding changeset, and be confused that the patch replaces an emit("\0", 1) with an append("", 1)). Can you add a comment as to why it works?

Comment on attachment 8842199 [details] Bug 1341880 - do not emit trailing \0 in PrintfState; https://reviewboard.mozilla.org/r/115862/#review118050

Pushed by ttromey@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/133ad85684a1 do not emit trailing \0 in PrintfState; r=glandium