mirrors/cosmopolitan
1
0
Fork 0
mirror of https://github.com/jart/cosmopolitan.git synced 2025-05-22 21:32:31 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Improve documentation

Browse source 
The Cosmo API documentation page is pretty good now
https://justine.lol/cosmopolitan/documentation.html
This commit is contained in:
Justine Tunney 2020-12-27 07:02:35 -08:00
parent 13437dd19b
commit 1bc3a25505
367 changed files with 2542 additions and 26178 deletions
Download patch file Download diff file Expand all files Collapse all files

34
libc/stdio/printf.c
View file

@ -25,39 +25,39 @@
* Cosmopolitan supports most of the standard formatting behaviors
* described by `man 3 printf`, in addition to the following:
*
* - %jd, %jx, etc. are {,u}intmax_t which in Cosmopolitan is 128-bit.
* - `%jd`, `%jx`, etc. are {,u}intmax_t which in Cosmopolitan is 128-bit.
*
* - %'d or %,d may be used to insert thousands separators. The prior is
* - `%'d` or `%,d` may be used to insert thousands separators. The prior is
* consistent with C; the latter is consistent with Python.
*
* - %m inserts strerror(errno) into the formatted output. This is
* - `%m` inserts strerror(errno) into the formatted output. This is
* consistent with glibc, musl, and uclibc.
*
* - %n inserts "\n" on non-Windows and "\r\n" on Windows. This is the
* - `%n` inserts "\n" on non-Windows and "\r\n" on Windows. This is the
* same behavior as Java. It's incompatible with glibc's behavior,
* since that's problematic according to Android's security team.
*
* - %hs converts UTF-16/UCS-2 → UTF-8, which can be helpful on Windows.
* - `%hs` converts UTF-16/UCS-2 → UTF-8, which can be helpful on Windows.
* Formatting (e.g. %-10hs) will use monospace display width rather
* than string length or codepoint count.
*
* - %ls (or %Ls) converts UTF-32 → UTF-8. Formatting (e.g. %-10ls) will
* use monospace display width rather than string length.
* - `%ls` (or `%Ls`) converts UTF-32 → UTF-8. Formatting (e.g. %-10ls)
* will use monospace display width rather than string length.
*
* - The %#s and %#c alternate forms display values using the standard
* IBM standard 256-letter alphabet. Using %#.*s to specify length
* will allow true binary (i.e. with NULs) to be formatted.
* - The `%#s` and `%#c` alternate forms display values using the
* standard IBM standard 256-letter alphabet. Using `%#.*s` to specify
* length will allow true binary (i.e. with NULs) to be formatted.
*
* - The %'s and %'c alternate forms are Cosmopolitan extensions for
* - The `%'s` and `%'c` alternate forms are Cosmopolitan extensions for
* escaping string literals for C/C++ and Python. The outer quotation
* marks can be added automatically using %`s. If constexpr format
* marks can be added automatically using ``%`s``. If constexpr format
* strings are used, we can avoid linking cescapec() too.
*
* - The backtick modifier (%`s and %`c) and repr() directive (%r) both
* ask the formatting machine to represent values as real code rather
* than using arbitrary traditions for displaying values. This means
* it implies the quoting modifier, wraps the value with {,u,L}['"]
* quotes, displays NULL as "NULL" rather than "(null)", etc.
* - The backtick modifier (``%`s`` and ``%`c``) and repr() directive
* (`%r`) both ask the formatting machine to represent values as real
* code rather than using arbitrary traditions for displaying values.
* This means it implies the quoting modifier, wraps the value with
* {,u,L}['"] quotes, displays NULL as "NULL" rather than "(null)".
*
* @see palandprintf() for intuitive reference documentation
* @see {,v}{,s{,n},{,{,x}as},f,d}printf