How to check condition if some custom field empty?
For example, it's possible to check that qa_contact is not set.
sub object_end_of_set_all {
my ($self, $args) = #_;
my $object = $args->{'object'};
if ($object->{'bug_status'} eq 'RESOLVED') {
if ($object->{'qa_contact'} eq "") {
ThrowUserError("empty_qa_contact");
}
}
}
Is there is same way for custom field e.g. cf_test ?
I know that to save custom field in variable, need to:
my $test = new Bugzilla::Field({ name => 'cf_test' });
Which method can be used to get its value or check if it's not empty ?
Found out two problems:
1. After executing ThrowUserError("...") any object (cf_test or even bugzilla fields) is always NULL.
2. If to use Hook "object_end_of_set_all" for custom field then only a cached values are showed for a custom field.
The answer is to use different Hook:
sub bug_end_of_update {
my ($self, $args) = #_;
my ($bug, $old_bug, $timestamp, $changes) = #$args{qw(bug old_bug timestamp changes)};
if ($bug->bug_status eq 'RESOLVED') {
if ($bug->cf_test eq "") {
ThrowUserError("test_is_empty");
}
}
}
Related
# Object.pm
sub update {
$table = $self->DB_TABLE;
...
}
The update function is triggered when a value is updated and seems to be executed multiple times by other files whose relevant parts look like:
# Status.pm
use constant DB_TABLE => 'Status';
# Flag.pm
use constant DB_TABLE => 'flag';
I don't know the inner workings of this project, but modified Flag.pm and Object.pm as below because I need to use a different table for updating flag.
# Flag.pm
use constant DB_TABLE => 'flag';
use constnat DB_UPDATE_TABLE => '<Table to use when updating flag>';
# Object.pm
sub update {
my $table = undef;
if($self->DB_UPDATE_TABLE) {
$table = $self->DB_UPDATE_TABLE;
} else {
$table = $self->DB_TABLE;
}
}
When I triggered sub update, I get
Can't locate object method "DB_UPDATE_TABLE" via package "<Status.pm>" at Object.pm.
Is there any way I can check if DB_UPDATE_TABLE exists in each file without error? I can add the following line to Status.pm, but there are a couple dozen of files like Status.pm.
use constant DB_UPDATE_TABLE => '';
I don't know why it is $self->DB_TABLE not $self->{DB_TABLE} but with the assumption that it is a method... tried the following, but it also had its own error.
if( my $ref = eval { $self->can( DB_UPDATE_TABLE ) } ) {
$table = $self->DB_UPDATE_TABLE;
} else {
$table = $self->DB_TABLE;
}
Bareword "DB_UPDATE_TABLE" not allowed while "strict subs" in use at Object.pm => I couldn't find the part to set 'strict subs'
Super close!
$self->can( DB_UPDATE_TABLE )
should be
$self->can("DB_UPDATE_TABLE")
In Javascript, I am sending a blob using XHR by the following code:
var v=new FormData();
v.append("EFD",new Blob([...Uint8Array...]));
var h=new XMLHttpRequest();
h.setRequestHeader("Content-type","multipart/form-data; charset=utf-8");
h.open("POST","...url...");
h.send(v);
In the server, I have created in Perl the following function, that suppose to implement CGI->param and CGI->upload:
# QS (Query String) receive in argument string for single parameter or array of many required parameters.
# If string been supplied: Return the value of the parameter or undef if missing.
# If array been supplied, a hash will be returned with keys for param names and their corresponding values.
# If the first argument is undef, then return hash with ALL available parameters.
sub QS {
my $b=$ENV{'QUERY_STRING'};
if($ENV{'REQUEST_METHOD'} eq "POST") {
read(STDIN,$b,$ENV{'CONTENT_LENGTH'}) or die "E100";
}
my $e=$_[0]; my $t=&AT($e); my $r={}; my #q=split(/&/,$b);
my %p=(); if($t eq "A") { %p=map { $_=>1 } #{$e}; }
foreach my $i(#q) {
my ($k,$s)=split(/=/,$i); $s=~tr/+//; $s=~s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;
if($t eq "") { $r->{$k}=$s; }
elsif($t eq "A") { if($p{$k}) { $r->{$k}=$s; } }
elsif($k eq $_[0]) { return $s; }
}
return $r;
}
# AT is a function for determining type of an object, and also a quck way to distinguish between just a string and a number.
sub AT {
if(!defined $_[0]) { return ""; } my $v=ref($_[0]);
if($v eq "") { return ($_[0]*1 eq $_[0])?"N":"S"; }
my $k={"ARRAY"=>"A","HASH"=>"H"};
return $k->{$v}||$_[0]->{_obt}||$v;
}
So in the main program it will be called as:
my $EFD=&FW::QS("EFD"); # FW = The module name where QS and AT are.
When I issuing the POST from the client, the script in the server does not pop-up any errors, and does not terminates - it continues to run and run and run.... Endlessly.... Consuming 100% CPU time and 100% memory - without any explanation.
I have these in the beginning of the script, though:
use strict;
use warnings;
use diagnostics;
but it still behave in such a way that I need to kill the script in order to terminate it...
Anyone know what I did wrong...? No infinite loop here, as far as I know... If I change the Blob to regular classic way of "...url...?EFD=dhglkhserkhgoi" then it works just fine, but I want a Blob....
Thanks a lot
This QS function is only usable for POSTs with an application/x-www-urlencoded body, which yours isn't.
In a shortcode I can limit wp_query results by custom field values.
Example:
[my-shortcode meta_key=my-custom-field meta_value=100,200 meta_compare='IN']
And obviously it's possible to use multiple custom fields in a wp_query like WP_Query#Custom_Field_Parameters
But how can I use multiple custom fields in my shortcode? At the moment I do pass all the shortcode parameters with $atts.
On of a few different solutions might be to use a JSON encoded format for the meta values. Note that this isn't exactly user-friendly, but certainly would accomplish what you want.
You would of course need to generate the json encoded values first and ensure they are in the proper format. One way of doing that is just using PHP's built in functions:
// Set up your array of values, then json_encode them with PHP
$values = array(
array('key' => 'my_key',
'value' => 'my_value',
'operator' => 'IN'
),
array('key' => 'other_key',
'value' => 'other_value',
)
);
echo json_encode($values);
// outputs: [{"key":"my_key","value":"my_value","operator":"IN"},{"key":"other_key","value":"other_value"}]
Example usage in the shortcode:
[my-shortcode meta='[{"key":"my_key","value":"my_value","operator":"IN"},{"key":"other_key","value":"other_value"}]']
Which then you would parse out in your shortcode function, something like so:
function my_shortcode($atts ) {
$meta = $atts['meta'];
// decode it "quietly" so no errors thrown
$meta = #json_decode( $meta );
// check if $meta set in case it wasn't set or json encoded proper
if ( $meta && is_array( $meta ) ) {
foreach($meta AS $m) {
$key = $m->key;
$value = $m->value;
$op = ( ! empty($m->operator) ) ? $m->operator : '';
// put key, value, op into your meta query here....
}
}
}
Alternate Method
Another method would be to cause your shortcode to accept an arbitrary number of them, with matching numerical indexes, like so:
[my-shortcode meta-key1="my_key" meta-value1="my_value" meta-op1="IN
meta-key2="other_key" meta-value2="other_value"]
Then, in your shortcode function, "watch" for these values and glue them up yourself:
function my_shortcode( $atts ) {
foreach( $atts AS $name => $value ) {
if ( stripos($name, 'meta-key') === 0 ) {
$id = str_ireplace('meta-key', '', $name);
$key = $value;
$value = (isset($atts['meta-value' . $id])) ? $atts['meta-value' . $id] : '';
$op = (isset($atts['meta-op' . $id])) ? $atts['meta-op' . $id] : '';
// use $key, $value, and $op as needed in your meta query
}
}
}
Currently have an small perl script what for the given username fetch his email address from the ActiveDirectory using Net::LDAP.
The search part is the following:
my $user = "myuser";
my $mesg = $ldap->search(
base => "dc=some,dc=example,dc=com",
filter => '(&(sAMAccountName=' . $user . ')(mail=*))', #?!?
);
for my $entry ($mesg->entries) {
my $val = $entry->get_value('mail');
say "==$val==";
}
Working ok.
How i should modify the above statement to fetch all available information for the given user myuser? I'm looking to get an perl-ish data structure, such something like next:
my $alldata = search(... all info for the given $user ... );
say Dumper $alldata; #hashref with all stored informations for the $user
It is probably dead simple - but i'm an total AD & LDAP-dumb person...
Edit: When I dump out the $msg->entries (what is an LADP::Entry object) got something, but i'm not sure than it contains everything or only the part of the stored data...
I've done something similar, and I use this to query LDAP:
my $ldapResponse = $ldap->search(base => $base, filter => $filter, attrs => $attrs);
And then this to parse it:
if ($ldapResponse && $ldapResponse->count()) {
$ldapResponse->code && die $ldapResponse->error;
my %domainNames = %{$ldapResponse->as_struct};
foreach my $domainName (keys %domainNames) {
my %ldapResponse;
my %dnHash = %{$domainNames{$domainName}};
foreach my $attr (sort(keys %dnHash)) {
# Note that the value for each key of %dnHash is an array,
# so join it together into a string.
my $value = join(" ", #{$dnHash{$attr}});
$ldapResponse{$attr} = $value;
}
// Dump/use %ldapResponse
}
}
I've never tried to use the ldap->entries in your code, but the above works for me!
I explicitly specify a(long) list of attributes ($attr), but perhaps that's optional as your example shows, and you can get ALL LDAP fields by just skipping that arg to search().
For an application I'd like to create some kind of setup-steps. In one of the steps the database configuration is written to the application.ini file. This all works, but something very strange happens: All the paths to the directories (library, layout, ...) are changed from paths with APPLICATION_PATH . to full paths. As you can imagine, this isn't very systemfriendly. Any idea how I can prevent that?
I update the application.ini with this code:
# read existing configuration
$config = new Zend_Config_Ini(
$location,
null,
array('skipExtends' => true,
'allowModifications' => true));
# add new values
$config->production->doctrine->connection = array();
$config->production->doctrine->connection->host = $data['server'];
$config->production->doctrine->connection->user = $data['username'];
$config->production->doctrine->connection->password = $data['password'];
$config->production->doctrine->connection->database = $data['database'];
# write new configuration
$writer = new Zend_Config_Writer_Ini(
array(
'config' => $config,
'filename' => $location));
$writer->write();
Since Zend_Config_Ini uses the default ini scanning mode (INI_SCANNER_NORMAL), it will parse all options and replace constants with their respective values. What you could do, is call parse_ini_file directly, using the INI_SCANNER_RAW mode, so the options aren't parsed.
ie. use
$config = parse_ini_file('/path/to/your.ini', TRUE, INI_SCANNER_RAW);
You will get an associative array that you can manipulate as you see fit, and afterwards you can write that back with the following snippet (from the comments):
function write_ini_file($assoc_arr, $path, $has_sections=FALSE) {
$content = "";
if ($has_sections) {
foreach ($assoc_arr as $key=>$elem) {
$content .= "[".$key."]\n";
foreach ($elem as $key2=>$elem2) {
if(is_array($elem2))
{
for($i=0;$i<count($elem2);$i++)
{
$content .= $key2."[] = ".$elem2[$i]."\n";
}
}
else if($elem2=="") $content .= $key2." = \n";
else $content .= $key2." = ".$elem2."\n";
}
}
}
else {
foreach ($assoc_arr as $key=>$elem) {
if(is_array($elem))
{
for($i=0;$i<count($elem);$i++)
{
$content .= $key2."[] = ".$elem[$i]."\n";
}
}
else if($elem=="") $content .= $key2." = \n";
else $content .= $key2." = ".$elem."\n";
}
}
if (!$handle = fopen($path, 'w')) {
return false;
}
if (!fwrite($handle, $content)) {
return false;
}
fclose($handle);
return true;
}
ie. call it with :
write_ini_file($config, '/path/to/your.ini', TRUE);
after manipulating the $config array. Just make sure you add double quotes to the option values where needed...
Or alternatively - instead of using that function - you could try writing it back using Zend_Config_Writer_Ini, after converting the array back to a Zend_Config object, I guess that should work as well...
I'm guess you could iterate over the values, checking for a match between the value of APPLICATION_PATH, and replacing it with string literal APPLICATION_PATH.
That is if you know that APPLICATION_PATH contains the string '/home/david/apps/myapp/application' and you find a config value '/home/david/apps/myapp/application/views/helpers', then you do some kind of replacement of the leading string '/home/david/apps/myapp/application' with the string 'APPLICATION_PATH', ending up with 'APPLICATION_PATH "/views/helpers"'.
Kind of a kludge, but something like that might work.
This is a long shot - but have you tried running your Zend_Config_Writer_Ini code while the APPLICATION_PATH constant is not defined? It should interpret it as the literal string 'APPLICATION_PATH' and could possibly work.