mirror of
https://git.hardenedbsd.org/hardenedbsd/HardenedBSD.git
synced 2024-11-16 23:57:54 +01:00
.. | ||
all | ||
append.test | ||
assocd.test | ||
async.test | ||
case.test | ||
clock.test | ||
cmdAH.test | ||
cmdInfo.test | ||
concat.test | ||
dcall.test | ||
defs | ||
dstring.test | ||
env.test | ||
error.test | ||
eval.test | ||
event.test | ||
exec.test | ||
expr.test | ||
fhandle.test | ||
fileName.test | ||
for.test | ||
format.test | ||
get.test | ||
history.test | ||
if.test | ||
incr.test | ||
info.test | ||
interp.test | ||
io.test | ||
ioCmd.test | ||
join.test | ||
license.terms | ||
lindex.test | ||
link.test | ||
linsert.test | ||
list.test | ||
llength.test | ||
load.test | ||
lrange.test | ||
lreplace.test | ||
lsearch.test | ||
lsort.test | ||
misc.test | ||
parse.test | ||
pid.test | ||
pkg.test | ||
proc.test | ||
README | ||
regexp.test | ||
remote.tcl | ||
rename.test | ||
scan.test | ||
set.test | ||
socket.test | ||
source.test | ||
split.test | ||
string.test | ||
subst.test | ||
switch.test | ||
trace.test | ||
unknown.test | ||
uplevel.test | ||
upvar.test | ||
while.test |
Tcl Test Suite -------------- SCCS: @(#) README 1.6 96/04/17 10:51:11 This directory contains a set of validation tests for the Tcl commands. Each of the files whose name ends in ".test" is intended to fully exercise one or a few Tcl commands. The commands tested by a given file are listed in the first line of the file. You can run the tests in two ways: (a) type "make test" in ../unix; this will run all of the tests. (b) start up tcltest in this directory, then "source" the test file (for example, type "source parse.test"). To run all of the tests, type "source all". In either case no output will be generated if all goes well, except for a listing of the tests.. If there are errors then additional messages will appear in the format described below. Note: don't run the tests as superuser, since this will cause several of the tests to fail. The rest of this file provides additional information on the features of the testing environment. This approach to testing was designed and initially implemented by Mary Ann May-Pumphrey of Sun Microsystems. Many thanks to her for donating her work back to the public Tcl release. Definitions file: ----------------- The file "defs" defines a collection of procedures and variables used to run the tests. It is read in automatically by each of the .test files if needed, but once it has been read once it will not be read again by the .test files. If you change defs while running tests you'll have to "source" it by hand to load its new contents. Test output: ------------ Normally, output only appears when there are errors. However, if the variable VERBOSE is set to 1 then tests will be run in "verbose" mode and output will be generated for each test regardless of whether it succeeded or failed. Test output consists of the following information: - the test identifier (which can be used to locate the test code in the .test file) - a brief description of the test - the contents of the test code - the actual results produced by the tests - a "PASSED" or "FAILED" message - the expected results (if the test failed) You can set VERBOSE either interactively (after the defs file has been read in), or you can change the default value in "defs". Selecting tests for execution: ------------------------------ Normally, all the tests in a file are run whenever the file is "source"d. However, you can select a specific set of tests using the global variable TESTS. This variable contains a pattern; any test whose identifier matches TESTS will be run. For example, the following interactive command causes all of the "for" tests in groups 2 and 4 to be executed: set TESTS {for-[24]*} TESTS defaults to *, but you can change the default in "defs" if you wish. Saving keystrokes: ------------------ A convenience procedure named "dotests" is included in file "defs". It takes two arguments--the name of the test file (such as "parse.test"), and a pattern selecting the tests you want to execute. It sets TESTS to the second argument, calls "source" on the file specified in the first argument, and restores TESTS to its pre-call value at the end. Batch vs. interactive execution: -------------------------------- The tests can be run in either batch or interactive mode. Batch mode refers to using I/O redirection from a UNIX shell. For example, the following command causes the tests in the file named "parse.test" to be executed: tclTest < parse.test > parse.test.results Users who want to execute the tests in this fashion need to first ensure that the file "defs" has proper values for the global variables that control the testing environment (VERBOSE and TESTS).