CVS log for src/tests/lib/libcurses/slave/curses_commands.c

Up to [cvs.NetBSD.org] / src / tests / lib / libcurses / slave

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

tests/libcurses: fix grammar in comment (since yesterday)

tests/libcurses: clean up comments

tests/libcurses: fix argument handling for mvwget_wch

There's currently no test for that function, therefore no functional
change.

tests/libcurses: unexport argument handling functions

No functional change.

tests/libcurses: KNF for while loop

tests/libcurses: remove redundant argument numbers

Since all arguments are processed in increasing order, there is no need
to add the redundant argument numbers.  Most of the curses functions
have less than 5 arguments, which makes it easy enough to count the ARG
macros.

Changes to curses_commands.c:

* replace ^(\tARG_\w+\()\d(?:, |) with $1
* replace (define ARG_\w+\()i,\s with $1
* replace args\[i\] with *args++
* replace \(i\) with ()
* replace \(void\)0 with args++

The wrong argument count in cmd_mvwget_wch is still detected by
lint.lua, as it was before.  There is no test yet that covers this
function.

Reapply fix for big-endian environments, which was lost when GSoC results
were merged...

http://www.nerv.org/netbsd/?q=id:20200620T075016Z.3584036ccf31f69ee76ea4a02e9be30ff081df21

> Fix false positive for mvscanw tests on big endian machines.
>
> When conversion specifier is not a derivative form of "%s", retrieve
> input as 32bit integer, and then convert to string literal. Then we
> can avoid interpretation from ASCII code to integer, which is
> apparently byte-order depended.

When mvscanw(3) fails, string is not modified.
Let's ignore the 2nd result for this case.

s/&string/string/ for mvscanw(3); string is already pointer to buffer.

tests/libcurses: clean up inclusion guards, function declarations

No functional change.

tests/libcurses: remove excess empty lines

tests/libcurses: query function arguments in consistent order

tests/libcurses: don't access args directly

By providing declarative syntax for accessing the arguments, the
unnecessarily detailed boilerplate code is hidden.  This allows easy
inspection by tools and humans, to check for typos and other mistakes.

tests/libcurses: replace ARG_CHTYPE_STRING with ARG_CHTYPE

All uses of the previous macro did not treat the argument as a string or
array of chtype, but as a single chtype.  It's strange that the previous
code arbitrarily split the access to the argument by first storing it as
a pointer and then dereferencing it.

No functional change.

tests/libcurses: add shortcut macros for SCREEN and unsigned int

tests/libcurses: remove redundant comments

There is nothing surprising about call2, call3 or call4.

tests/libcurses: reduce boilerplate in function dispatcher

No functional change.  The generated code for GCC 9.3 on NetBSD 9.99.80
x86_64 is exactly the same as before.

tests/libcurses: reduce boilerplate in function dispatcher

This makes the code more declarative and easier to reason about.

The generated code stays exactly the same.

tests/libcurses: fix scanf parameter for unsigned int

tests/libcurses: fix typo in license text

Merge in code from Google Summer of Code project which dramatically
increases the number of tests performed.

Thanks to Naman Jain <jnaman806@gmail.com> for his excellent work on
this GSoC project.

Fix false positive for mvscanw tests on big endian machines.

When conversion specifier is not a derivative form of "%s", retrieve
input as 32bit integer, and then convert to string literal. Then we
can avoid interpretation from ASCII code to integer, which is
apparently byte-order depended.

Sync with HEAD

Start supporting wide char tests.

Fix argument parsing for mvinsch, last arg is a chtype not a string.

Resync to 2012-11-19 00:00:00 UTC

sync with head

* Fix mvscanw return
* Fix the *vline family of calls, one argument is chtype not int

- add support for getparyx and getyx calls
- allow a NULL pointer to be returned to the director
- add support for passing back a single chtype character
- fix some indentation
- fix a lot of curses command calls that were just plain wrong
- don't try to allocate storage for a NULL parameter, it doesn't need it

fix bug where we read the same argument twice. Name the timeout val the
same 'tval' in both timeout routines.

Catchup with rmind-uvmplock merge.

Turn warns on for all tests and fix all the bugs.

Apply patch provided by mrg to fix bad calls to vw_printw and vw_scanw

Fix obvious bug.
(note: this file still doesn't build on amd64)

Add atf for libcurses with some tests