Develop Reference

JavaScript hanging writing errors from a .ps1 script

Because of reasons, I have a situation in which I need to use a .js file to call a .bat file, which calls a .ps1 file. It may sound crazy, but yes, I have to launch the .ps1 file this way.
My problem is that, the .js file is hanging after writing a variable number of errors. The number of errors that causes the script to hang changes between scripts, but seems to be consistent within a script.
I have included sanitized scripts below.
Every time I run this particular combo using cscript (cscript c:\temp\test.js "c:\temp\test3.bat"), the script stops after writing "Checking device 10 of 100." to the log file and does not return me to the prompt. If I run the .bat file from the same command prompt, the script runs as expected. If, on line 41 of the .ps1 script, I change the message type from "Error" to "Verbose" (for example), the script runs fine and completes as expected.
It seems that the .js file does not like something about "Write-Error $Message" from the Out-PsLogging function but:
I cannot tell what it doesn't like
I cannot tell why it stops writing errors after (in this case) 10 errors
I have seen this behavior on Windows 10 and Server 2019 (ps 5.1).
Code:
test.js
var wshShell = new ActiveXObject("WScript.Shell");
var cmdLine = WScript.Arguments(0);
var result = {};
result.stdout = "";
result.stderr = "";
result.exitCode = -1;
if (!wshShell) {
WScript.Echo("Can not run command");
}
var oExec;
try {
oExec = wshShell.Exec(cmdLine);
WScript.Echo("Ran the command - " + cmdLine);
WScript.Echo(oExec.Status);
var line;
WScript.Echo("About to start while loop");
while (oExec.Status != 1) {
WScript.Echo(oExec.Status);
while (!oExec.StdOut.AtEndOfStream) {
WScript.Echo("Inside the second while loop");
line = oExec.StdOut.ReadLine();
result.stdout += line + "\n";
WScript.Echo(line + "\n");
WScript.Echo(oExec.StdOut.AtEndOfStream);
}
while (!oExec.StdErr.AtEndOfStream) {
WScript.Echo("Inside the third while loop");
line = oExec.StdErr.ReadLine();
result.stderr += line + "\n";
WScript.Echo(line + "\n");
WScript.Echo(oExec.StdErr.AtEndOfStream);
}
WScript.Sleep(100);
}
}
catch (e) {
WScript.Echo("Unable to run command - " + cmdLine);
WScript.Echo(e);
}
test3.bat
cmd.exe /C Powershell.exe -Command "&{C:\temp\test3.ps1 -Verbose}"
test3.ps1
Function Out-PsLogging {
[CmdletBinding(DefaultParameterSetName = 'SessionOnly')]
param(
[Parameter(Mandatory, ParameterSetName = 'File')]
[System.IO.FileInfo]$LogPath,
[Parameter(Mandatory)]
[string]$Message,
[Parameter(Mandatory)]
[ValidateSet('Info', 'Warning', 'Error', 'Verbose', 'First')]
[string]$MessageType
)
$logType = "LogFile"
Switch ($logType) {
"LogFile" {
Switch ($MessageType) {
"Info" { [System.IO.File]::AppendAllLines([string]$LogPath, [string[]]$Message); Write-Host $Message }
"Warning" { [System.IO.File]::AppendAllLines([string]$LogPath, [string[]]$Message); Write-Warning $Message }
"Error" { [System.IO.File]::AppendAllLines([string]$LogPath, [string[]]$Message); Write-Error $Message }
"Verbose" { [System.IO.File]::AppendAllLines([string]$LogPath, [string[]]$Message); Write-Verbose $Message -Verbose }
"First" { [System.IO.File]::WriteAllLines($LogPath, $Message); Write-Verbose $Message -Verbose }
}
}
}
}
$a = 0
$array = #(1..100)
$LogPath = 'C:\temp\log3.txt'
$message = ("{0}: Beginning {1}." -f ([datetime]::Now).ToString("yyyy-MM-dd`THH:mm:ss"), $myinvocation.MyCommand)
Out-PsLogging -LogPath $LogPath -MessageType First -Message $message
$itGlueConfigsMatchingPolicy = Foreach ($device in $array) {
$a++
$message = ("{0}: Checking device {1} of {2}." -f ([datetime]::Now).ToString("yyyy-MM-dd`THH:mm:ss"), $a, $array.count)
Out-PsLogging -LogPath $LogPath -MessageType Error -Message $message
}; Remove-Variable a -Force
$message = ("{0}: Completed {1}." -f ([datetime]::Now).ToString("yyyy-MM-dd`THH:mm:ss"), $myinvocation.MyCommand)
Out-PsLogging -LogPath $LogPath -MessageType Verbose -Message $message

Turns out the issue is a bug in wscript: https://groups.google.com/forum/#!topic/microsoft.public.scripting.wsh/kIvQsqxSkSk

Problems with displaying text in Javascript "N�O�T�I�C�E�", php displays it with no problem

PHP creates the text from metadata from image file which stores it in variable of the same name of IMG source.
No Problems
JavaScript displays the gallery to show images, scroll & enlarge.
During enlargement it gives description of image from source attribute but always displays the text with � after every character. But PHP doesn't.
I've tried encoding which you will see from code & & echo header("Content-Type: text/html; charset=ISO-8859-1");
PHP:
$path = "gallery/";
$objs = new RecursiveIteratorIterator(new
RecursiveDirectoryIterator($path),
RecursiveIteratorIterator::SELF_FIRST);
$num = 0;
foreach ($objs as $pic){
$pic = str_replace('gallery/', '', $pic);
if ($pic == '..' or $pic == '.'){
continue;
}
$exif = exif_read_data("$DIR/gallery/$pic", 0, true);
$pic = str_replace('.JPG', '', $pic);
${$pic} = $exif['IFD0']['Comments']; ///// php Variable
echo "<script> var $pic = '".$exif['IFD0']['Comments']."';
var $pic = utf8_encode($pic);</script>";
// JS of variable of same name.
}
Sloppy but Works
JQuery:
$('#mSide').on('click',function(){
var src = $('#mPic').attr('src');
var v = src.replace('/gallery/','');
v = v.replace('.JPG','');
v = unescape(encodeURIComponent(v));
$('#fullPic').show();
$('#fullPic').append('<img class=\"fPic\" src=\"'+src+'\"
height=\"90%\" widdth=\"90%\" style=\"margin-left:0px;\" />');
$('#fullPic').append('<span id=\"picSummary\"
style=\"color:#FFFFFF;\" >testing <textarea > '+ window[v] +'</textarea>
</span>');
$('#fullPic').prepend('<input class=\"closeFullP headr\"
value=\"Close\" READONLY><br class=\"closeFullP\">');
});
Sloppy but Works and window[v] is the call
Should display: this is the new val = NOTICE: This e-mail message may contain legally privileged and/or confidential information. If you are not the intended recipient, you are hereby notified that any dissemination of the contents of this message is strictly prohibited. If you have received this message in error, please immediately notify the sender at ########### and delete all copies of this e-mail message and its attachments.
Not: N�O�T�I�C�E�:� �T�h�i�s� �e�-�m�a�i�l�
�m�e�s�s�a�g�e� �m�a�y� �c�o�n�t�a�i�n�
�l�e�g�a�l�l�y� �p�r�i�v�i�l�e�g�e�d� �a�n�d�/�o�r�
�c�o�n�f�i�d�e�n�t�i�a�l� �i�n�f�o�r�m�a�t�i�o�n�.�
�I�f� �y�o�u� �a�r�e� �n�o�t� �t�h�e�
�i�n�t�e�n�d�e�d� �r�e�c�i�p�i�e�n�t�,...............

Never use utf8_encode() or utf8_decode() they make awful assumptions and happily/silently corrupt your data. Always use a function that allows you to explicitly specify your input and output encodings.
The data you're extracting from EXIF is encoded as UTF16-LE, and your page is... something else. I can't tell what and you should probably figure that out and be consistent.
Suggested reading: UTF-8 all the way through
If your page is encoded as ISO-8859-1:
$proper = mb_convert_encoding($input, 'ISO-8859-1', 'UTF-16-LE');
If your page is encoded as UTF-8:
$proper = mb_convert_encoding($input, 'UTF-8', 'UTF-16-LE');
Ref: https://secure.php.net/manual/en/function.mb-convert-encoding.php

Live iPerf Output Printed to HTML with PHP/Ajax

I've been working on a web interface for iPerf and it's going pretty well. My goal is to have the output of the commands be streamed live on the page. It works for iPerf2 commands but when I run iPerf3, it only shows the output once the entire test is complete.
Based on my research I've done on this site and others, I'm thinking it might have something to do with the buffer. I've been messing around with flushing the buffer at different times or setting the 'fread' length to different values but I can't get it to act the same as regular iperf commands. I think this because instead of running the 'popen' on the iperf command itself, I used popen on a python script that would run the iperf command. This still returned the same issue. It only shows the output on the web page once the entire test is complete.
Here is my code:
phpQ.php
<!DOCTYPE html>
<html>
<body>
<div>
Server Domain/IP Address: <input id="address" type="text"><br>
<input id="run" type="button" value="iPerf"><br><br>
</div>
<div id="result"></div>
<script>
function updateText(address) {
var ajax = new XMLHttpRequest();
ajax.onreadystatechange = function() {
if (this.readyState == 3) {
var old_value = document.getElementById("result").innerHTML;
document.getElementById("result").innerHTML = this.responseText;
}
};
if (purpose == 1) {
var url = 'ajaxQ.php?address='+address;
ajax.open('GET', url,true);
ajax.send();
}
}
document.getElementById("run").onclick = function(){
address = document.getElementById("address").value;
purpose = 1;
updateText(address);
}
</script>
</body>
</html>
ajaxQ.php - to see the difference, change "iperf -c" to "iperf3 -c" in the $iperfCmd variable.
<?php
function liveExecuteCommand($cmd,$address)
{
while (# ob_end_flush()); // end all output buffers if any
// tells the user what command is going to run
echo "<pre>Running Command: '".$cmd."'</pre>";
// open the command to run and read output
$proc = popen("$cmd", 'r');
$live_output = "";
$complete_output = "";
while (!feof($proc))
{
# flush();
$live_output = fread($proc,4096);
// adds the live output to the complete output
$complete_output = $complete_output . $live_output;
// prints the live output
echo "<pre>$live_output</pre>";
}
sleep(1);
// close the process
pclose($proc);
echo "<pre>------------------------------------------------------------\nAll Done!</pre>";
}
// this happens if the iPerf button is pressed
if (isset($_GET['address'])) {
$address = $_GET['address'];
$iperfCmd = "iperf -c ".$address." -i 1 -t 5";
liveExecuteCommand($iperfCmd,$address);
}
else{
echo "No post request";
}
?>
FYI, I am running the iperf client on an 'Ubuntu 16.04.2 LTS' CLI server and the iperf server on an 'Ubuntu 16.04 LTS' desktop server.
Thanks for any help!

Phantomjs cannot open a page that browser opens

I have a Phantomjs script that tries to open a url. phantomjs returns this error:
Unable to load resource (request ID:undefinedURL:http://foo.bar/tree/nav_value/27)
Error code: 203. Description: Error downloading http://foo.bar/tree/nav_value/27 - server replied: Not Found
But when I open the url http://foo.bar/tree/nav_value/27 with chrome browser, there's no problem and the page is loaded correctly!
This is the script:
// Read the Phantom webpage '#intro' element text using jQuery and "includeJs"
"use strict";
var page = require('webpage').create();
var system = require('system');
if (system.args.length != 2) {
console.log("please pass 2 argument")
}
var company_id = system.args[1]
console.log("c", company_id)
page.onConsoleMessage = function(msg) {
console.log("message", msg);
};
page.onResourceError = function(resourceError) {
console.log('Unable to load resource (request ID:' + resourceError.id + 'URL:' + resourceError.url + ')');
console.log('Error code: ' + resourceError.errorCode + '. Description: ' + resourceError.errorString);
};
page.onError = function(msg, trace) {
console.log("error", msg)
}
var nav_value;
page.open("http://foo.bar/tree/nav_value/27", 'post', 'username=navid&password=test', function(status) {
if (status === "success") {
page.includeJs("http://ajax.googleapis.com/ajax/libs/jquery/1.6.1/jquery.min.js", function() {
page.evaluate(function() {
nav_value = parseInt($("#value").text());
});
phantom.exit(0);
});
} else {
phantom.exit(1);
}
});
EDIT:
Something odd happens. When I run this code with phantomjs on windows on another machine it works. But on Ubuntu it returns the error!
The url that phantomjs is trying to open is on the same server. (Ubuntu)
What is the problem?

Not sure this will help, but I have some ideas that helped me figure out problems with PhantomJS in the past.
First, as you say it works on another machine, you may want to test other versions of PhantomJS, by downloading the executable and specifying the path on your Python script. Version 1.9.8 helped me with bypassing some security restrictions in the past (I also left some settings in case it may interest).
driver = webdriver.PhantomJS(
executable_path='/path/to/the/downloaded/phantomjs19',
# you can specify args, such as:
service_args=[
'--ignore-ssl-errors=true',
'--ssl-protocol=any',
'--web-security=false',
],
# and also other capabilities:
desired_capabilities={
'phantomjs.page.settings.resourceTimeout': '5000',
'phantomjs.page.settings.userAgent': (
"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/53 "
"(KHTML, like Gecko) Chrome/15.0.87"
),
},
)
You may also try to see if upgrading Selenium helps.
pip install selenium --upgrade
Another idea that might help to understand what is happening is to try to print a screenshot and log the page source before the error happens. You can do it like:
# Set the window size to something appropriate for your tests.
driver.set_window_size(900, 800)
driver.save_screenshot('screen.png')
# Check if the page source matches your expectations.
with open('temp.html', 'w') as f:
f.write(driver.page_source)
Please, let me know if this helps!

fwrite error - problems writing serial output rapidly

Firstly, I apologize if this issue has been posted before, I couldn't find the answer anywhere.
I'm writing a one page website that uses JavaScript, Ajax and PHP to take key presses and send them out via serial port.
I'm testing it on EasyPHP and WAMP, I'm yet to test on a Linux server.
My problem is that when keys are pressed rapidly I get the error below.
Warning: fopen(com4): failed to open stream: Permission denied in \keypresstest.php on line 7
If I press keys slowly, around 500ms intervals, it all works fine.
Furthermore, following the error above, PHP does not set my $fp variable, so it also returns errors:
Warning: fwrite() expects parameter 1 to be resource, boolean given in \keypresstest.php on line 9
Warning: fclose() expects parameter 1 to be resource, boolean given in \keypresstest.php on line 10
I think that there is slow 'release' of the serial port resource which it's defining as 'Resource id #3' and if I could some how speed this up or spawn another instance when needed i might be able to resolve the issue.
I hope someone reading this might have some advice.
My code is below, I've removed some components to keep this code more concise
HTML File
<script> <!-- call php script here -->
function sendKey(str) {
var xmlhttp = new XMLHttpRequest();
xmlhttp.onreadystatechange = function() {
if (xmlhttp.readyState == 4 && xmlhttp.status == 200) {
document.getElementById("txtHint").innerHTML = document.getElementById("txtHint").innerHTML + xmlhttp.responseText;
}
}
xmlhttp.open("GET", "keypresstest.php?q=" + str, true);
xmlhttp.send();
}
</script>
<script> // page js key listener script
document.addEventListener('keydown', function(event) { // things happen as keys are pressed
if (!arrayContainsCheck(keysHeld, event.keyCode)) { // check if the key is already being held, this is so keydown events don't continue to propagate whilst a key is held
if(event.keyCode == 38 || event.keyCode == 87) { // up/w keys, send output to php script "f"
sendKey("f");
} keysHeld.push(event.keyCode);
}
});
document.addEventListener('keyup', function(event) {
document.getElementById("lightSwitch").style.backgroundColor = "blue";
if(event.keyCode == 38 || event.keyCode == 87) { // up/w keys, send output to php script "g"
document.getElementById("upArrow").style.border = "1px solid red";
sendKey("g");
}
keyRelease(keysHeld, event.keyCode);
});
</script>
PHP File
<?php
$q = $_REQUEST["q"]; // get the q parameter from URL
// Output q parameter to com4 serial port
echo "q is $q <br />";
echo "defining fp... <br />";
$fp = fopen("com4", "w");
echo "fp defined as " . $fp . " .<br />";
fwrite($fp, $q);
fclose($fp);
?>