_known_hosts_real tests were failing after commit 81794a9:
FAIL: Hosts should be put in COMPREPLY
FAIL: Hosts should have username prefix and colon suffix
This is because tcl's get_hosts() is now doing a unique sort, but bash
_known_hosts_real() IS returning duplicates relying on bash's compgen/complete
to remove the duplicates). Fixed by calling _get_hosts -unsorted.
Commit 8227e76 was failing on these `chown' tests in bash-4.1:
FAIL: Check preserve special chars in funky\ user:Debia<TAB>
FAIL: Check preserve special chars in funky\.user:Debia<TAB>
FAIL: Check preserve special chars in fu\ nky.user:Debia<TAB>
FAIL: Check preserve special chars in f\ o\ o\.\bar:Debia<TAB>
FAIL: Check preserve special chars in foo\_b\ a\.r\ :Debia<TAB>
because a removal pattern is expanded:
$ a=\\b
$ w=\\
$ echo ${a#$w} # Doesn't work
\b
$ echo ${a#"$w"} # Ok
b
Prohibit word creation of characters if they're excluded from
$COMP_WORDBREAKS. For example, with ':' included in $COMP_WORDBREAKS,
'a b:' should split to 'b' and ':'. With ':' excluded from
$COMP_WORDBREAKS, 'a b:' should split to 'b:', NOT 'b' and ':'.
Also made `match_items()' more strict about matching
space/newline/prompt after the last item.
Added options to match_items():
- end-newline
- end-prompt
- end-space
and transferred them to other functions.
Function `assert_complete()' now has a `-nospace' option to explicitly
disallow a space after a completion...
Function `assert_bash_list()' now expects a single item to be followed
by a newline.
File locations are prefixed with `$::srcdir' so that `make distcheck' can
execute the test suite using a relative path.
The current working directory is removed from the test-suite-bash-prompt.
Furthermore, no more dynamic creation of files in dir $::srcdir/fixtures since
this dir is read-only during `make distcheck'. Instead create dynamic files in
$TESTDIR/tmp
The test suite uses these "directory" variables
tcl bash description
------------ ----------- ------------------------------------
$::srcdir $SRCDIR where `fixtures' reside, relative
$::srcdirabs $SRCDIRABS where `fixtures' reside, absolute
$::TESTDIR $TESTDIR where `runtest' is invoked, absolute
Steps to reproduce the problem:
$ LANG=POSIX ./run unit/_parse_help.exp
...
Running ./unit/_parse_help.exp ...
FAIL: short + long
FAIL: short + long, slash separated
Cause:
When comparing list items, `assert_bash_list()' expects the real list to be
sorted, whereas the output of `_parse_help' is unsorted.
These particular two tests were failing because they suffered the
following LANG-dependant sort-difference:
$ cat t.txt
-m
--n
$ LANG=en_US.UTF-8 sort t.txt
-m
--n
$ LANG=POSIX sort t.txt
--n
-m
Solution:
Made to default for `assert_bash_list' more-intuitive: unsorted, and added an
option `-sort' to explicitly enable sorting.
I felt uncomfortable adding yet another optional argument, so I seized this
opportunity to move subsequent optional arguments to single optional arguments.
E.g.:
assert_bash_list {expected cmd {test ""} {prompt /@} {size 20}} {
has now become:
# ...
# @param list $args Options:
# -nosort Compare list unsorted. Default is sorted
# -prompt Bash prompt. Default is `/@'
# -chunk-size N Compare list N items at a time. Default
# is 20.
assert_bash_list {expected cmd test {args {}}
(and the `test' argument has become mandatory).
This provides a bit of case insensivity to file extension matching,
allowing for example _filedir foo to match both *.foo and *.FOO. Note
that this is not real case insensivity; mixed case extensions like
*.FOO.gz are not matched by _filedir foo.gz so those need to be
handled like before this change.
This fixes tilde completion when doing for instance `ls ~<TAB>'. See
also: Alioth #312613.
Added _tilde() helper function.
Contrary to _expand() and __expand_tilde_by_ref(), _tilde() also leaves the
tilde (~) intact. Replacement of the tilde has been a complaint in other
bugreports, especially since plain bash completion (without bash-completion
installed) doesn't expand the tilde when doing a `ls ~<TAB>'.
To run the tests:
./run completion/ls.exp
./run unit/_tilde.exp
Usage: _get_comp_words_by_ref [OPTIONS] [VARNAMES]
Available VARNAMES:
cur Return cur within varname "cur"
prev Return prev within varname "prev"
words Return words within varname "words"
cword Return cword within varname "cword"
Available OPTIONS:
-n EXCLUDE Characters out of $COMP_WORDBREAKS which should NOT be
considered word breaks. This is useful for things like scp
where we want to return host:path and not only path, so we
would pass the colon (:) as -n option in this case. Bash-3
doesn't do word splitting, so this ensures we get the same
word on both bash-3 and bash-4.
-c VARNAME Return cur within specified VARNAME
-p VARNAME Return prev within specified VARNAME
-w VARNAME Return words within specified VARNAME
-i VARNAME Return words within specified VARNAME
Example usage:
$ _get_comp_words_by_ref -n : cur prev
regardless of $cur beginning with ' or not"), because this is failing
tests:
FAIL: completing f a\'b/ should return i
FAIL: completing f a\"b/ should return i
FAIL: f a\$b/ should show completions
FAIL: f a\\b/ should show completions
FAIL: completing f2 a\'b/ should return i
FAIL: completing f2 a\"b/ should return i
FAIL: f2 a\$b/ should show completions
FAIL: f2 a\\b/ should show completions
- Fix _filedir to check for availability of COMP_WORDS (_filedir runs from
within a completion instead of the command line) when doing a `complete -p
${COMP_WORDS[0]'
- Fix _filedir usage comment
- Enhanced _filedir tests
- Added _filedir test "completing with filter '.e1' should show completions"
This solves the following problems:
- now one function call suffices instead of two (_get_cword; _get_pword) if
subsequent words need to be retrieved. Also more than two words can be
retrieved at once, e.g.: _get_comp_words_by_ref cur prev prev2 prev3
Also this prevents passing of `wordbreakchars' to differ in calls to
`_get_cword' and `_get_pword', e.g.: _get_comp_words_by_ref -n : cur prev
- passing by reference, no subshell call necessary anymore
- _get_pword now also takes into account the cursor position
Added testsuite proc `assert_no_output()'
Word of caution:
The passing-arguments-by-ref system in bash doesn't work if the new variable is
also declared local. For example:
t() {
local a
# ...
eval $1=b
}
a=c; t a; echo $a # Outputs "c", should be "b"
# Variable "a" is 'forbidden'
To make name collissions like this less likely to happen, but make the real
function still use readable variables, I've wrapped the `*_by_ref'
functions within an additional layer using variables prefixed with double
underscores (__). For example:
_t() {
# Readable variables can still be used here
local a
# ...
eval $1=b
}
t() {
local __a
_t __a
eval $1=\$__a
}
a=c; t a; echo $a # Outputs "b"
# Variable "__a" is 'forbidden'
Now only more obfuscated variables (starting with double prefix (__)) are
forbidden to use.
The linefeed between the final "d" and "}" seems to be significant for
these versions of sed. And they are quite anal about comments, so
move them outside of the sed statement here.
Directories `a\b' and `a*b' are now created run-time on a
non-Cygwin/Windows system only. Tests concerning these directories will
also be run only on a non-Cygwin/Windows system.
(Alioth #312191)
On bash-3, completing a b c: with COMP_WORDBREAKS -= : would result in the
current word being bc: (Alioth #312190)
Added unit test, run via:
./run unit/_get_cword.exp
This fixes completing filenames containing single quote (') on bash-4.
Also added emulation of `-o filenames' to _filedir.
Added tests for _filedir.
Fixed array assignment within __reassemble_comp_words_by_ref().
The COMP_ settings of the last test of _count_args() were falling through to
the first test of _get_cword(), causing it to fail.
Unset COMP_ variables within the unit tests setup & teardown.
Fixed last test at `_count_args.exp' by solving conflicting settings of
COMP_CWORD and COMP_POINT.
Steps to reproduce the problem:
$ ./runUnit _count_args.exp _get_cword.exp
...
Running ./unit/_count_args.exp ...
Running ./unit/_get_cword.exp ...
ERROR Unexpected output from bash command "_get_cword should run without errors":
bash: $index: substring expression < 0
...
If a word is made up of multiple word separator characters:
$ a b::<TAB> # CWORDS are: a, b, and :: (correct)
__reassemble_comp_words_by_ref() couldn't handle this. It assumed CWORDS were:
$ a b::<TAB> # CWORDS: a, b, : and : (but they're not)
Added test case for this. To run the automated tests:
./runUnit _get_cword.exp
Actually enhanced __get_cword3 to _get_cword, and removed __get_cword4.
__get_cword4 could handle chars to exclude from COMP_WORDBREAKS, but
failed with partial quoted arguments (e.g. "a 'b c|", | = cursor
position). This was no problem till bash-4.0.35, because bash < 4.0.35
also returned partial quoted arguments incorrectly. See also:
http://www.mail-archive.com/bug-bash@gnu.org/msg06094.html
Now that bash-4.0.35 returns quoted arguments ok, __get_cword3 is
enhanced to also handle chars to exclude from COMP_WORDBREAKS. Because
__get_cword3 also handles partial quoted arguments correctly, this makes
__get_cword3 suitable for bash-4 as well.
Code in `unit_start()' and `completion_start()' is merged and put central in
lib/library.exp:
- start_bash()
- source_bash_completion()
- init_tcl_bash_globals()
Global string variable `bash_versinfo_0' is replaced with list variable
`BASH_VERSINFO'.
Expands only tilde (~), if first char, in variable.
This function displays bash's capabilities of passing a variable by
reference (variable indirection) which allows us to avoid using a
subshell. As far as I can see it works surprisingly well?
To run the automated test:
./runUnit __expand_tilde_by_ref.exp
Also fixed some testsuite issues regarding list splitting.
- Added code comments to _get_cword, __get_cword3 & __get_cword4
- (testsuite) Added tests for _get_cword
- (testsuite) Bugfixes assert_bash_exec() && match_items()
Bash-4 splits COMP_WORDS using characters from COMP_WORDBREAKS, but has
a bug where quoted words are also splitted, see:
http://www.mail-archive.com/bug-bash@gnu.org/msg06095.html
__get_cword3 is used for bash-2/3 and __get_cword4 is used for bash-4.
__get_cword4 handles well temporarily disabling of COMP_WORDBREAK
characters, but fails quoted words (a 'b c) and subshells (a $(b c).
See the expected failures when running the automated tests.
__get_cword3 does a better job of returning quoted words.
To run the automated tests on bash-3/4:
$ ./runUnit _get_cword.exp [--tool_exec <path to bash-3/4 binary>]
