NAME
Remedy::ARSTools - a perl wrapper to the ARSperl project, providing a simplified object interface with field definition caching.
SYNOPSIS
use Remedy::ARSTools;
#create a new object with a new field definition data cache
my $Remedy = new Remedy::ARSTools(
Server => $server_host_or_ip,
User => $username,
Pass => $password,
ConfigFile => $file_to_cache_field_definition_data,
Schemas => [ 'list', 'of', 'schema names', 'to get', 'field data for' ]
) || die ($Remedy::ARSTools::errstr);
#create a ticket
my $ticket_number = $Remedy->CreateTicket(
Schema => $schema_name,
Fields => {
'fieldName1' => "value1",
'fieldName2' => "value2,
... etc ...
}
) || die $Remedy->{'errstr'};
#merge ticket
my $ticket_number = $Remedy->MergeTicket(
Schema => $schema_name,
MergeCreateMode => "Overwrite",
Fields => {
'fieldName1' => "value1",
'fieldName2' => "value2,
... etc ...
}
) || die $Remedy->{'errstr'};
#modify a ticket
$Remedy->ModifyTicket(
Schema => $schema_name,
Ticket => $ticket_number,
Fields => {
'fieldName1' => "value1",
'fieldName2' => "value2,
... etc ...
}
) || die $Remedy->{'errstr'};
#query for tickets
$tickets = $Remedy->Query(
Schema => $schema_name,
QBE => $qbe_string,
Fields => ['array', 'of', 'fieldNames', 'to', 'retrieve']
) || die $Remedy->{'errstr'};
#delete a ticket
$Remedy->DeleteTicket(
Schema => $schema_name,
Ticket => $ticket_number
) || die $Remedy->{'errstr'};
$parsed_diary = $Remedy->ParseDiary(
Diary => $raw_diary_data_from_database,
) || die $Remedy->{'errstr'};
$Remedy->Destroy();
OVERVIEW & DEPENDENCIES
First things first, you need ARSperl installed for this module to work. ARSperl is the perl interface to the Remedy C API, and provides all the "magic" of talking to your Remedy server. This module is a perl wrapper that sits atop ARSperl. The purpose of this module is to provide a nice, simplified interface to ARSperl that is independent of the particular version of ARSperl and the Remedy C API that you have installed.
You will need the following items to be installed prior to attempting to use this module:
- Remedy C API
-
This comes as part of your Remedy server installation. This API is proprietary, and owned by the Remedy corporation (or BMC, or Peregrin or whom ever owns them this week). You can usually find this under the 'api' directory under the remedy installation directory on your remedy server. The Remedy C API is required by the ARSperl installation.
- ARSperl
-
as mentioned earlier, this is the perl interface to the Remedy C API. You can download ARSperl from your local CPAN mirror, or also from the sourceforge project page:
http://sourceforge.net/projects/arsperl/
- Data::DumpXML
-
this perl module is available from your local CPAN mirror. It is used to serialize field definition data into a configuration file.
A NOTE ON FIELD DEFINITION DATA
Remedy assigns a unique 'field_id' to each field in a schema. In order to do pretty much anything with that field in the Remedy API, you must know the field_id rather than the name. For instance 'entry_id' is typically field_id '1', however it gets a lot more complicated from there. Additionally, Remedy implements fields with enumerated values in a unique way, assigning an integer to each enumerated value starting at 0. For instance, 'Status' = "New" = 0. One must also know the enum value corresponding to the 'human readable' value when performing operations using the API.
This module attempts to hide all of that, allowing you to reference fields directly by name, and enumerated field values by their 'human readable' (string) value (rather than by integer). However, to do so, the module needs to maintain a mapping of field id's and enumerated values. The mapping can be loaded from the remedy server when you create a Remedy::ARSTools object, however, this is a rather time-consuming task, and is also network intensive. As an alternative, you can specify a special file in which the object will store field definition data. This file acts as a field definition data cache, and it's contents are automatically updated.
Use of an external file in which to cache field definition data is highly recommended for speed improvments, but is not completely necessary. The 'penalty' for not using the file, is that it takes much longer to instantiate new objects.
new
This is, of course, the object constructor. Upon failure, the method will return the undef value, and an error message will be written to $Remedy::ARSTools::errstr. At object creation, the field definition data is loaded either directly from the remedy server, or from the provided config file.
syntax
my $Remedy = new Remedy::ARSTools([ options ]) || die $Remedy::ARSTools::errstr;
options
the following options are accepted by the new() function:
- Server (required)
-
this is the hostname or ip address of the remedy server to which access is desired.
- User (required)
-
the 'Login Name' of the Remedy account to be used for access to 'Server'
- Pass (required)
-
the password for 'User'
- ConfigFile
-
this is the full path and filename of the file in which field definition data should be cached (and which may already contain field definition data).
- LoginOverride
-
if a non-zero value is specified for this option, the function will not attempt to login to the remedy server until a function requiring it is called.
- Port
-
if specified, will instruct the C API to communicate with the Remedy server only on the specified TCP port.
- RPCNumber
-
if specified, will instruct the C API to communicate with the Remedy server using only the specified RPC port (note only supported where ARSPerl > 8.001 is installed, also note, RPCNumber and Port are mutually exclusive).
- Language
-
The language the user is using. If not specified, the default will be used.
- AuthString
-
It's here because it's in ARSPerl, and it's in ARSperl because it's in the C API. It "has something to do with the Windows Domain", according to the ARSperl documentation. You can specify it here, and it'll be passed on to ARSperl, if you know what to do with it.
- ReloadConfigOK
-
if 0 or the undef value are supplied on this argument, the module will not attempt to update cached field definition data in the specified config, if it is found to be out of date.
- GenerateConfig
-
if 0 or the undef value are supplied on this argument, an error is generated if the specified ConfigFile does not already contain field definition data. The specified ConfigFile will not be created if it does not already exist.
- TruncateOK
-
If a non-zero value is specified on this argument, functions which write data into remedy will silently truncate field data values if they are too long to fit in their specified fields. This is a short cut to setting this option individually on every function call.
LoadARSConfig
This function loads field definition data from the 'ConfigFile' specified in the object, or directly from the Remedy server (if 'ConfigFile' dosen't exist, or the internal 'staleConfig' flag is set).
Normally, this function is called only internally, but it can be used externally, to force an object to reload it's field definition data.
syntax
$Remedy->LoadARSConfig() || die $Remedy->{'errstr'};
ARSLogin
This function connects to the remedy server specified by the 'Server' in the object, obtaining a "control token" from the remedy server. If the object is already logged into the Remedy server, this function will return without doing anything. If the object's internal 'staleLogin' flag is set true, or if the object is not yet connected to the Remedy server (such as when 'loginOverride' is specified at object instantiation), the function will connect.
syntax
$Remedy->ARSlogin() || die $Remedy->{'errstr'};
Destroy
This is the object destructor. This function releases the "control token" back to the Remedy server, clearing the user's session. This also completely destroys the object.
syntax
$Remedy->Destroy();
CheckFields
This function checks a hash containing field name and value pairs against field definition data. If a field value is too long, it is truncated (if the object's TruncateOK is set), otherwise an error is returned. Also string values provided for enum fields are converted to their integer values.
This function is unique, in that if no errors are found, the undef value is returned, with the string "ok" on the object's errstr. If errors are found a string containing a concatenation of all errors found in the field list is returned. If a more serious error is encountered (not relating to field values), then the undef value is returned with a string other than "ok" on the object's errstr.
This is most definitely called internally, though it can be useful externally for data validation.
syntax
my $errors = $Remedy->CheckFields( [ options ] ) || do {
die $Remedy->{'errstr'} if ($remedy->{'errstr'} ne "ok");
};
if ($errors !~/^\s*/){ print $errors, "\n"; }
options
the CheckFields function accepts the following options
- Schema (required)
-
the name of the schema in which the fields that values should be checked for exist.
- Fields (required)
-
a hash reference in the form of { 'field_name' => $value ... }, where each 'field_name' refers to a field in 'Schema', and each $value represents a value for the field.
NOTE: the referenced hash will be modified (values truncated, or strings translated to integers for enum fields)
CreateTicket
Create a new record in the specified Schema containing the specified field values.
syntax
my $ticket_number = $Remedy->CreateTicket( [ options ] ) || die $Remedy->{'errstr'};
options
the following options are accepted by the CreateTicket function
- Schema
-
the name of the schema in which the record should be created
- Fields
-
a hash reference in the form of { 'field_name' => $value ... }, where each 'field_name' is the name of a field in 'Schema' and each $value is a value to place in that field.
ModifyTicket
Change the specified field values in the specified record, in the specified Schema
syntax
$Remedy->ModifyTicket( [ options ] ) || die $Remedy->{'errstr'};
options
the following options are accepted by the ModifyTicket function:
- Ticket
-
the 'ticket number' (or 'entry id', or 'record number' ... field id number 1, that is) of the record that we wish to modify.
- Schema
-
the name of the schema in which 'Ticket' exists
- Fields
-
a hash reference in the form of { 'field_name' => $value ... }, where each 'field_name' is the name of a field in 'Ticket' and $value is the value to set on that field.
DeleteTicket
Remove the specified record from the specified Schema. Obviously, this will fail if the 'User' specified at instantiation, does not have administrator permissions.
syntax
$Remedy->DeleteTicket( [ options ] ) || die $Remedy->{'errstr'};
options
- Ticket
-
the 'ticket number' (or 'entry id', or 'record number' ... field id number 1, that is) of the record that we wish to delete.
- Schema
-
the name of the schema in which 'Ticket' exists
Query
Return selected field values from Tickets matching the specified query string in the specified Schema. It should be noted that having external processes query through the ARS API presents a lot of overhead on the server, is slower than a direct SQL query to the underlying database, and at least in my opinion, should be avoided. However, If you're here, I'll presume you have your reasons ;-).
Data is returned as an array reference. Each element of the array is a hash reference, representing a ticket which matched the specified query string. The hash reference is in the form of { 'field_name' => $value ... }, where each 'field_name' is the name of a field in the ticket and $value is the value for that field.
syntax
my $tickets = $Remedy->Query( [ options ] ) || die $Remedy->{'errstr'};
options
the Query function accepts the following arguments:
- Schema
-
the name of the schema that you want to return matching records from
- QBE
-
this is the "Query By Example" string, or 'query string' or "that thing you type in the 'Search Criteria' line when you click the 'Advanced' button in the client". You know what I'm talking about probably. Just remember, it's not exactly the same thing as an SQL 'where' clause.
- Fields
-
An array contianing the list of field names corresponding to selected field values we'd like returned from matching records in 'Schema'. You may find it helpful to build the array reference inline with the function call like so:
Fields => ['field1','field2','field3']
ParseDiary
This is one of those routines that dosen't seem to fit in anywhere perfectly. I included it in Remedy::ARSTools, because ... well, it's dern handy, and this seemed to be as good of a place as any for it. If you took my advice from the notes on 'Query' and decided to do your queries directly from the database used by your remedy server, then you might run into some problems with diaries. Remedy stores the diary as one big ol' text field. Which is a problem, because as you are probably aware, the diary has a timestamp and user associated with each entry. So what you get when you select a diary field from your database, is each diary entry separated by some trash. This 'trash' is the username and timestamp. So what this does is to parse a raw diary format into the same format as you would get if you had queried a diary field via ARS (i.e. the same format as returned by ARS::getField.). That format is an array reference. Each element in the array is, in turn, a hash reference. Each nested hash contains three fields 'timestamp', 'user', and 'value'. Oh yeah, and the Array is sorted chronologically, with the earliest entries first. Here's another look at what the data-structure looks like:
\@DIARY = [
{ 'timestamp' => $date, 'user' => $user, 'value' => $diary_entry }
...
];
syntax
my $diary_entries = $Remedy->ParseDiary( [ options ] ) || die $Remedy->{'errstr'};
options
the following options are accepted by the ParseDiary function:
- Diary
-
a big ol' text string from the database containing an unparsed diary
- ConvertDate
-
if a non-zero value is specified on this option, the timestamp field of each diary entry will be converted from 'epoch' time to the local time zone set in the 'locale' on your computer.
MergeTicket
Pretty much the same thing as CreateTicket, but with a merge transaction, allowing you all of the freedoms and responsibilities that come with that (for heaven's sake: be careful, mmmkay?).
The returned value is one of two things. If you are in MergeMode = 'Create' or 'Error', this will return the entry_id (i.e. "ticket number, aka "field id = 1") of the record you just merged. If you are in MergeMode = "Overwrite", AND the entry_id value you specified already exists this will contain the string "overwritten" (you may now see my point about being careful).
syntax
my $ticket_number = $Remedy->MergeTicket( [options] ) || die $Remedy->{'errstr'}
options
- Schema
-
the name of the schema (i.e. "form") in which the record should be merged.
- Fields
-
a hash reference in the form of { 'field_name' => $value ... }, where each 'field_name' is the name of a field in 'Schema' and each $value is a value to place in that field. Just like CreateTicket
- MergeCreateMode
-
this controls what happens when the record you want to merge has the same value for field_id = 1 as an existing record. That is to say, what happens in the situation where you specify an existing ticket number on Fields. There are three values:
- Error
-
Throw an error and exit if there is an existing record in the spefified Schema with the same ticket number
- Create
-
Create a new record with a different ticket number. So, basically ignore your speficied field_id = 1 (ticket number) value and create a new record with the rest of the field values, and return the new ticket number.
- Overwrite
-
This is the funny biznezz. If you specify this option, it will overwrite the existing ticket number with your values, and the function will return "overwritten" instead of a ticket number. You have been warned :-).
- AllowNullFields
-
if you specify a non-null value on this option, it will give the API permission to bypass required fields (excepting of course the ARS system fields ... 'status', 'short description', yadda yadda).
- SkipFieldPatternCheck
-
if you specify a non-null value on this option, it will give the API permission to bypass field pattern checking (menus, etc). Obviously, it will not let you set out of range enums and the like, but you'd never get that far anyhow ... this module would throw a field check error before that, but I digress. Set this if you want to set goofy field values and get away with it.
EXAMPLES
#create a new ticket in Users schema my $ticket_number = $Remedy->CreateTicket( Schema => "User", Fields => { 'Login Name' => "sbsqrpnts", 'Password' => "tar-t4r-s4us3", 'Group List' => "fryCooks jellyFishers", 'Full Name' => "Squarepants, Sponge B.", 'Email Address' => 'sbsqrpnts@krustykrab.com', 'License Type' => "Fixed", 'Assigned To' => "sbsqrpnts" } ) || die ($Remedy->{'errstr'});
#query for tickets my $tickets = $Remedy->Query( Schema => "Users", QBE => "'Login Name' = \"sbsqrpnts\"", Fields => [ "Request ID", "Login Name" ] ) || die ($Remedy->{'errstr'});
#modify a ticket $Remedy->modifyTicket( Ticket => $tickets->[0]->{'Request ID'}, Schema => "User", Fields => { 'Full Name' => "SpongeBob Squarepants" } ) || die ($Remedy->{'errstr'});
#delete a ticket $Remedy->DeleteTicket( Schema => "User", Ticket => $tickets->[0]->{'Request ID'} ) || die ($Remedy->{'errstr'});
#log out $Remedy->Destroy();
AUTHOR
Andrew N. Hicox <andrew@hicox.com> Sutio BootyQuake http://www.hicox.com
LICENSE
This module is released under the licensing terms of Perl itself. http://www.arsperl.org/artistic.txt