Is it possible to enable exit on error behavior in an interactive Tcl shell? - automation

I need to automate a huge interactive Tcl program using Tcl expect.
As I realized, this territory is really dangerous, as I need to extend the already existing mass of code, but I can't rely on errors actually causing the program to fail with a positive exit code as I could in a regular script.
This means I have to think about every possible thing that could go wrong and "expect" it.
What I currently do is use a "die" procedure instead of raising an error in my own code, that automatically exits. But this kind of error condition can not be catched, and makes it hard to detect errors especially in code not written by me, since ultimately, most library routines will be error-based.
Since I have access to the program's Tcl shell, is it possible to enable fail-on-error?
EDIT:
I am using Tcl 8.3, which is a severe limitation in terms of available tools.
Examples of errors I'd like to automatically exit on:
% puts $a(2)
can't read "a(2)": no such element in array
while evaluating {puts $a(2)}
%
% blublabla
invalid command name "blublabla"
while evaluating blublabla
%
As well as any other error that makes a normal script terminate.
These can bubble up from 10 levels deep within procedure calls.
I also tried redefining the global error command, but not all errors that can occur in Tcl use it. For instance, the above "command not found" error did not go through my custom error procedure.

Since I have access to the program's Tcl shell, is it possible to
enable fail-on-error?
Let me try to summarize in my words: You want to exit from an interactive Tcl shell upon error, rather than having the prompt offered again?
Update
I am using Tcl 8.3, which is a severe limitation in terms of available
tools [...] only source patches to the C code.
As you seem to be deep down in that rabbit hole, why not add another source patch?
--- tclMain.c 2002-03-26 03:26:58.000000000 +0100
+++ tclMain.c.mrcalvin 2019-10-23 22:49:14.000000000 +0200
## -328,6 +328,7 ##
Tcl_WriteObj(errChannel, Tcl_GetObjResult(interp));
Tcl_WriteChars(errChannel, "\n", 1);
}
+ Tcl_Exit(1);
} else if (tsdPtr->tty) {
resultPtr = Tcl_GetObjResult(interp);
Tcl_GetStringFromObj(resultPtr, &length);
This is untested, the Tcl 8.3.5 sources don't compile for me. But this section of Tcl's internal are comparable to current sources, tested using my Tcl 8.6 source installation.
For the records
With a stock shell (tclsh), this is a little fiddly, I am afraid. The following might work for you (though, I can imagine cases where this might fail you). The idea is
to intercept writes to stderr (this is to where an interactive shell redirects error messages, before returning to the prompt).
to discriminate between arbitrary writes to stderr and error cases, one can use the global variable ::errorInfo as a sentinel.
Step 1: Define a channel interceptor
oo::class create Bouncer {
method initialize {handle mode} {
if {$mode ne "write"} {error "can't handle reading"}
return {finalize initialize write}
}
method finalize {handle} {
# NOOP
}
method write {handle bytes} {
if {[info exists ::errorInfo]} {
# This is an actual error;
# 1) Print the message (as usual), but to stdout
fconfigure stdout -translation binary
puts stdout $bytes
# 2) Call on [exit] to quit the Tcl process
exit 1
} else {
# Non-error write to stderr, proceed as usual
return $bytes
}
}
}
Step 2: Register the interceptor for stderr in interactive shells
if {[info exists ::tcl_interactive]} {
chan push stderr [Bouncer new]
}
Once registered, this will make your interactive shell behave like so:
% puts stderr "Goes, as usual!"
Goes, as usual!
% error "Bye, bye"
Bye, bye
Some remarks
You need to be careful about the Bouncer's write method, the error message has already been massaged for the character encoding (therefore, the fconfigure call).
You might want to put this into a Tcl package or Tcl module, to load the bouncer using package req.
I could imagine that your program writes to stderr and the errorInfo variable happens to be set (as a left-over), this will trigger an unintended exit.

Related

Asynchronous reading of an stdout

I've wrote this simple script, it generates one output line per second (generator.sh):
for i in {0..5}; do echo $i; sleep 1; done
The raku program will launch this script and will print the lines as soon as they appear:
my $proc = Proc::Async.new("sh", "generator.sh");
$proc.stdout.tap({ .print });
my $promise = $proc.start;
await $promise;
All works as expected: every second we see a new line. But let's rewrite generator in raku (generator.raku):
for 0..5 { .say; sleep 1 }
and change the first line of the program to this:
my $proc = Proc::Async.new("raku", "generator.raku");
Now something wrong: first we see first line of output ("0"), then a long pause, and finally we see all the remaining lines of the output.
I tried to grab output of the generators via script command:
script -c 'sh generator.sh' script-sh
script -c 'raku generator.raku' script-raku
And to analyze them in a hexadecimal editor, and it looks like they are the same: after each digit, bytes 0d and 0a follow.
Why is such a difference in working with seemingly identical generators? I need to understand this because I am going to launch an external program and process its output online.
Why is such a difference in working with seemingly identical generators?
First, with regard to the title, the issue is not about the reading side, but rather the writing side.
Raku's I/O implementation looks at whether STDOUT is attached to a TTY. If it is a TTY, any output is immediately written to the output handle. However, if it's not a TTY, then it will apply buffering, which results in a significant performance improvement but at the cost of the output being chunked by the buffer size.
If you change generator.raku to disable output buffering:
$*OUT.out-buffer = False; for 0..5 { .say; sleep 1 }
Then the output will be seen immediately.
I need to understand this because I am going to launch an external program and process its output online.
It'll only be an issue if the external program you launch also has such a buffering policy.
In addition to answer of #Jonathan Worthington. Although buffering is an issue of writing side, it is possible to cope with this on the reading side. stdbuf, unbuffer, script can be used on linux (see this discussion). On windows only winpty helps me, which I found here.
So, if there are winpty.exe, winpty-agent.exe, winpty.dll, msys-2.0.dll files in working directory, this code can be used to run program without buffering:
my $proc = Proc::Async.new(<winpty.exe -Xallow-non-tty -Xplain raku generator.raku>);

Is there a bgerror hook equivalent for uncaught errors out of the event loop in TCL?

In Tcl, when an uncaught error occurs in the background in the event loop, the function bgerror will be called if one exists (or any function registered via interp bgerror).
When an unknown command is encountered, the unknown command will be called.
Is there a similar mechanism for uncaught errors outside of the event loop that make their way to the top level? The default behavior from tclsh is just to print the error on stderr, but I would like a specific proc to be called instead without having to put the entire code in one big catch.
The default behaviour, if a Tcl error makes it's way to the top of the main execution path, is to print the errorInfo trace to standard error and, if running as a script, exit with a non-zero result code. If you want to override this, you need to either write your own entry code/main loop in C (when you can do anything you want when Tcl_Eval() returns TCL_ERROR), or wrap your code in a catch or try. You can use uplevel #0 within such Tcl code to ensure that the trapping code doesn't appear to be on the stack as seen by the rest of your code.
For example:
proc printStackOnError {script} {
try {
uplevel "#0" $script
} on error {msg opt} {
set stackinfo [dict get $opt -errorinfo]
# remove last three uninteresting lines; they're just part of this procedure's machinery
puts stderr [join [lrange [split $stackinfo "\n"] 0 end-3] "\n"]
}
}
Demonstrating (interactively, with Tcl 8.6):
% printStackOnError {eval {error foo}}
foo
while executing
"error foo"
("eval" body line 1)
invoked from within
"eval {error foo}"
You're advised to make sure you either print or log the errorInfo when an error is trapped at that level. Otherwise any bugs in your code are going to be unreasonably difficult to fix.

Ghostscript for PS integrity test: terminate at EOF, return error unless stack is empty

To test the integrity of PostScript files, I'd like to run Ghostscript in the following way:
Return 1 (or other error code) on error
Return 0 (success) at EOF if stack is empty
Return 1 (or other error code) otherwise
I could run gs in the background, and use a timeout to force termination if gs hangs with items left on the stack. Is there an easier solution?
Ghostscript won't hang if you send files as input (unless you write a program which enters an infinite loop or otherwise fails to reach a halting state). Having items on any of the stacks won't cause it to hang.
On the other hand, it won't give you an error if a PostScript program leaves operands on the operand stack (or dictionaries on the dictionary stack, clips on the clip stack or gstates on the graphics state stack). This is because that's not an error, and since PostScript interpreters normally run in a job server loop its not a problem either. Terminating the job returns control to the job server loop which does a save and restore round the total job, thereby clearing up anything left behind.
I'd suggest that if you really want to do this you need to adopt the same approach, you need to write a PostScript program which executes the PostScript program you want to 'test', then checks the operand stack (and other stacks if required) to see if anything is left. Note that you will want to execute the test program in a stopped context, as an error in the course of the program will clearly potentially leave stuff lying around.
Ghostscript returns 0 on a clean exit and a value less than 0 for errors, if I remember correctly. You would need to use signalerror in your test framework in order to raise an error if items are left at the end of a program.
[EDIT]
Anything supplied to Ghostscript on the command line by either -s or -d is defined in systemdict, so if we do -sInputFileName=/test.pdf then we will find in systemdict a key /InputFileName whose value is a string with the contents (/test.pdf). We can use that to pass the filename to our program.
The stopped operator takes an executable array as an argument, and returns either true or false depending on whether an error occurred while executing the array (3rd Edition PLRM, p 697).
So we need to run the program contained in the filename we've been given, and do it in a 'stopped' context. Something like this:
{InputFileName run} stopped
{
(Error occurred\n) print flush
%% Potentially check $error for more information.
}{
(program terminated normally\n) print flush
%% Here you could check the various stacks
} ifelse
The following, based 90% on KenS's answer, is 99% satisfactory:
Program checkIntegrity.ps:
{Script run} stopped
{
(\n===> Integrity test failed: ) print Script print ( has error\n\n) print
handleerror
(ignore this error which only serves to force a return value of 1) /syntaxerror signalerror
}{
% script passed, now check the stack
count dup 0 eq {
pop (\n===> Integrity test passed: ) print Script print ( terminated normally\n\n) print
} {
(\n===> Integrity test failed: ) print Script print ( left ) print
3 string cvs print ( item(s) on stack\n\n) print
Script /syntaxerror signalerror
} ifelse
} ifelse
quit
Execute with
gs -q -sScript=CodeToBeChecked.ps checkIntegrity.ps ; echo $?
For the last 1% of satisfaction I would need a replacement for
(blabla) /syntaxerror signalerror
It forces exit with return code 1, but is very verbous and distracts from the actual error in the checked script that is reported by handleerror. Therefore a cleaner way to exit(1) would be welcome.

tcl tcltest unknown option -run

When I run ANY test I get the same message. Here is an example test:
package require tcltest
namespace import -force ::tcltest::*
test foo-1.1 {save 1 in variable name foo} {} {
set foo 1
} {1}
I get the following output:
WARNING: unknown option -run: should be one of -asidefromdir, -constraints, -debug, -errfile, -file, -limitconstraints, -load, -loadfile, -match, -notfile, -outfile, -preservecore, -relateddir, -singleproc, -skip, -testdir, -tmpdir, or -verbose
I've tried multiple tests and nothing seems to work. Does anyone know how to get this working?
Update #1:
The above error was my fault, it was due to it being run in my script. However if I run the following at a command line I got no output:
[root#server1 ~]$ tcl
tcl>package require tcltest
2.3.3
tcl>namespace import -force ::tcltest::*
tcl>test foo-1.1 {save 1 in variable name foo} {expr 1+1} {2}
tcl>echo [test foo-1.1 {save 1 in variable name foo} {expr 1+1} {2}]
tcl>
How do I get it to output pass or fail?
You don't get any output from the test command itself (as long as the test passes, as in the example: if it fails, the command prints a "contents of test case" / "actual result" / "expected result" summary; see also the remark on configuration below). The test statistics are saved internally: you can use the cleanupTests command to print the Total/Passed/Skipped/Failed numbers (that command also resets the counters and does some cleanup).
(When you run runAllTests, it runs test files in child processes, intercepting the output from each file's cleanupTests and adding them up to a grand total.)
The internal statistics collected during testing is available in AFACT undocumented namespace variables like ::tcltest::numTests. If you want to work with the statistics yourself, you can access them before calling cleanupTests, e.g.
parray ::tcltest::numTests
array set myTestData [array get ::tcltest::numTests]
set passed $::tcltest::numTests(Passed)
Look at the source for tcltest in your library to see what variables are available.
The amount of output from the test command is configurable, and you can get output even when the test passes if you add p / pass to the -verbose option. This option can also let you have less output on failure, etc.
You can also create a command called ::tcltest::ReportToMaster which, if it exists, will be called by cleanupTests with the pertinent data as arguments. Doing so seems to suppress both output of statistics and at least most resetting and cleanup. (I didn't go very far in investigating that method.) Be aware that messing about with this is more likely to create trouble than solve problems, but if you are writing your own testing software based on tcltest you might still want to look at it.
Oh, and please use the newer syntax for the test command. It's more verbose, but you'll thank yourself later on if you get started with it.
Obligatory-but-fairly-useless (in this case) documentation link: tcltest

proc_open interaction

Here's what I'm trying to achieve: open a shell (korn or bash, doesn't matter), from that shell, I want to open a ssh connection (ssh user#host). At some point it is likely to happen I will be prompted for either a password or I might be asked whether or not I'm sure I want to connect (offending keys).
Before anyone asks: yes, I am aware there is a plugin for ssh2 exec calls, but the servers I'm working on don't support it, and are unlikely to do so.
Here's what I've tried so far:
$desc = array(array('pipe','r'),array('pipe','w'));//used in all example code
$p = proc_open('ssh user#host',$desc,$pipes);
if(!is_resource($p)){ die('#!#$%');}//will omit this line from now on
sleep(1);//omitting this,too but it's there every time I need it
Then I tried to read console output (stream_get_contents($pipes[1])) to see what I have to pass next (either password, yes or return 'connection failed: '.stream_get_contents($pipes[1]) and proc_close $p.
This gave me the following error:
Pseudo-terminal will not be allocated because stdin is not a terminal.
So, I though ssh was called in the php:// io-stream context, seems a plausible explanation of the above error.
Next: I though about my first SO question and decided it might be a good idea to open a bash/ksh shell first:
$p = proc_open('bash',$desc,$pipes);
And take it from there, but I got the exact same error message, only this time, the script stopped running but ssh did run. So I got hopeful, then felt stupid and, eventually, desperate:
$p=proc_open('bash && ssh user#host',$desc,$pipes);
After a few seconds wait, I got the following error:
PHP Fatal error: Allowed memory size of 134217728 bytes exhausted (tried to allocate 133693440 bytes)
The Call Stack keeps bringing up the stream_get_contents line, even in my last desperate attempt:
#!/path/to/bin/php -n
<?php
$p = proc_open('bash && ssh user#host',array(array('pipe','r'),array('pipe','w')),$ps);
if (!is_resource($p))
{
die('FFS');
}
usleep(10);
fwrite($ps[0],'yes'."\n");
fflush($ps[0]);
usleep(20);
fwrite($ps[0],'password'."\n");
fflush($ps[0]);
usleep(20);
fwrite($ps[0],'whoami'."\n");
fflush($ps[0]);
usleep(2);
$msg = stream_get_contents($ps[1]);
fwrite($ps[0],'exit'."\n");
fclose($ps[0]);
fclose($ps[1]);
proc_close($p);
?>
I know, its a mess, a lot of fflush and redundancy, but the point is: I know this connection will first prompt me for offending keys, and then ask a password. My guess is the stream in $pipes[1] holds the ssh connection, hence it's content is huge. what I need then, is a pipe inside a pipe... is this even possible? I must be missing something, what good is a pipe if this isn't possible...
My guess is the proc_open command is wrong to begin with, (error: Broken pipe). But I really can't see any other way around the first error... any thoughts? Or follow up questions if the above rant isn't at all clear (which it probably isn't).
Before anyone asks: yes, I am aware there is a plugin for ssh2 exec
calls, but the servers I'm working on don't support it, and are
unlikely to do so.
There are actually two. The PECL module, which is a PITA that most servers don't have installed anyway and phpseclib, a pure PHP SSH2 implementation. An example of its use:
<?php
include('Net/SSH2.php');
$ssh = new Net_SSH2('www.domain.tld');
if (!$ssh->login('username', 'password')) {
exit('Login Failed');
}
echo $ssh->exec('pwd');
echo $ssh->exec('ls -la');
?>