Handling argc==0 in the kernel

Posted Jan 28, 2022 20:07 UTC (Fri) by khim (subscriber, #9252)
In reply to: Handling argc==0 in the kernel by larkey
Parent article: Handling argc==0 in the kernel

You found the right page, just, perhaps, not the right quote. I think that one should close that discussion about meanings of different words:

> The wording, in particular the use of the word should, requires a Strictly Conforming POSIX Application to pass at least one argument to the exec function, thus guaranteeing that argc be one or greater when invoked by such an application.

I don't think explanations can be more clear than that.

You don't need to dig around for strict meanings of different words, authors of the standard did that for you.

Handling argc==0 in the kernel

Posted Jan 28, 2022 21:13 UTC (Fri) by larkey (guest, #104463) [Link]

But in what way does this contradict what I say? "should" means that this is a recommendation, i.e., a *strictly* conforming application needs to fulfill this criterion. But POSIX only uses should for argv[0] being the program name, but they use *shall* for argv ≠ NULL. Which is precisely my point. The latter is a *must do*, the former is not as strict as a requirement.