The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.

NAME

App::Dochazka::REST - Dochazka REST server

VERSION

Version 0.115

Development status

Dochazka is currently a Work In Progress (WIP). Do not expect it to do anything useful.

SYNOPSIS

This is the top-level module of the Dochazka REST server.

    use App::CELL qw( $CELL $log $meta $site );
    use App::Dochazka::REST;
    use Carp;

    my $REST = App::Dochazka::REST->init( sitedir => '/etc/dochazka' );
    croak( $REST->{init_status}->text ) unless $REST->{init_status}->ok;

Read on for documentation.

DESCRIPTION

This is App::Dochazka::REST, the Perl module that implements the REST interface, data model, and underlying database of Dochazka, the open-source Attendance/Time Tracking (ATT) system.

Dochazka as a whole aims to be a convenient, open-source ATT solution. Its reference implementation runs on the Linux platform.

Dochazka architecture

There is more to Dochazka than App::Dochazka::REST, of course. Dochazka REST is the "server component" of Dochazka, consisting of a Plack/PSGI web server (implemented using Web::Machine) and a data model.

Once App::Dochazka::REST is installed, configured, and running, a client will be need in order to actually use Dochazka.

Though no client yet exists, two are planned: a command-line interface (App::Dochazka::CLI) and a web front-end (App::Dochazka::WebGUI). Stand-alone report generators and other utilities that may or may not ever be implemented can also be thought of as clients.

REST INTERFACE

App::Dochazka::REST attempts to present a RESTful interface to potential clients. For a description of what this means in practice, see ....

In the HTTP request, the client should provide an Accept: header specifying either HTML (text/html) or JSON (application/json). If neither is specified, the response body will be in HTML.

The REST interface consists of a number of resources. The resources are documented in the REST server itself, and can be explored using a web browser. For example, if the URL of your Dochazka installation is http://dochazka.site, that's the place to start. The response to that URL will show all the top-level resources. Then, if you want to explore a resource further, you go to http://dochazka.site/[RESOURCE].

The following top-level resources are used to create, read, update, and delete attendance data:

  • employee

  • privhistory

  • schedhistory

  • schedule

  • activity

  • interval

  • lock

Request-response cycle

(In order to protect user passwords from network sniffing and other nefarious activities, the server may be set up to accept HTTPS requests only. These are then decrypted to make HTTP requests. This document ignores this important implementation detail. If this bothers you, you can treat 'HTTP' as an abbreviation for 'HTTP and/or HTTPS'.)

Incoming HTTP requests are handled by App::Dochazka::REST::Resource, which inherits from Web::Machine::Resource. The latter uses Plack to implement a PSGI-compliant stack.

Web::Machine uses a "state-machine" approach to implementing the HTTP 1.1 standard. The request is processed by running it through the state machine: the request goes in, and the response comes out. Each "cog" of the state machine is a Web::Machine::Resource method that can be overrided by a child module. In our case, this module is App::Dochazka::REST::Resource, and the methods it overrides can be seen in its source code.

The behavior of the resulting web server can be summarized, in a hand-wavingly way, as follows:

  • UTF-8 assumed

    The server assumes all incoming requests are encoded in UTF-8, and it encodes all of its responses in UTF-8 as well.

  • Allowed methods test

    One of the first things the server looks at, when it receives a request, is the method. Only certain HTTP methods, such as 'GET' and 'POST', are accepted. If this test fails, a "405 Method Not Allowed" response is sent.

  • Internal and external authentication

    All incoming requests are subject to HTTP Basic Authentication. The credentials entered by the user can be authenticated against an external database (LDAP), and internal database (PostgreSQL 'employees' table), or both. For details, see "AUTHENTICATION". If authentication fails, a "401 Unauthorized" response is sent. This should not be confused with the next step ("Authorization/ACL check").

    In a web browser, repeated failed authentication attempts are typically associated with repeated display of the credentials dialog (and no other indication of what is wrong, which can be confusing to users but is probably a good idea, because any error messages could be abused by attackers).

  • Authorization/ACL check

    After the request is authenticated (i.e. associated with a known employee), the server examines the resource being requested and compares it with the employee's privilege level. If the privilege level is too low for the requested operation, a "403 Forbidden" response is sent.

  • Test for resource existence

    The last test a request undergoes on its quest to become a response is the test of resource existence. If the request is asking for a non-existent resource, e.g. http://dochazka.site/employee/curent, it cannot be fulfilled and a "404 Not Found" response will be sent.

    As some readers might already have guessed, the server already knows whether or not the resource exists in the previous step, Authorization/ACL check. Each stop on the quest can only generate a single error message, however. To deal with this quandary, the ACL check for non-existent resources is set to always pass. This causes requests for non-existent resources to whizz right through the ACL check, only to be caught by the Test for resource existence.

  • Response generation

    # FIXME need some verbiage here

DATA MODEL

This section describes the App::Dochazka::REST data model. Conceptually, Dochazka data can be seen to exist in the following classes of objects:

  • Policy (parameters set when database is first created)

  • Employee (an individual employee)

  • Privhistory (history of changes in an employee's privilege level)

  • Schedule (a schedule)

  • Schedhistory (history of changes in an employee's schedule)

  • Activities (what kinds of work are recognized)

  • Intervals ("work", "attendance", and/or "time tracked")

  • Locks (determining whether a reporting period is locked or not)

These classes are described in the following sections.

Policy

Dochazka is configurable in a number of ways. Some configuration parameters are set once at installation time and, once set, can never be changed -- these are referred to as "site policy" parameters. Others, referred to as "site configuration parameters" or "site params", are set in configuration files such as Dochazka_SiteConfig.pm (see "SITE CONFIGURATION") and can be changed more-or-less at will.

The key difference between site policy and site configuration is that site policy parameters cannot be changed, because changing them would compromise the referential integrity of the underlying database.

Site policy parameters are set at installation time and are stored, as a single JSON string, in the SitePolicy table. This table is rendered effectively immutable by a trigger.

For details, see App::Dochazka::REST::Model::Policy.

Employee

Dochazka is an Attendance and Time Tracking application. To simplify the matter, "Attendance and Time" can be replaced by the word "Work". We could also call Dochazka a "Work Tracking" application. Because "work" is usually done by "employees", all users of Dochazka are referred to as "employees" regardless of their actual legal status. You could even say that "employee" is the Dochazka term for "user".

Employees are distinguished by an internal employee ID number (EID), which is assigned by Dochazka itself when the employee record is created.

Other than the EID, Dochazka need not record any other employee identification data. That said, Dochazka has three optional employee identification fields (full name, nick, email address), which some sites may wish to use, but these can be left blank if needed or desired by the site. Dochazka does not verify the contents of these fields.

Dochazka doesn't care about the employee's identification information for two principal reasons: first, "Dochazka is not an address book" (there are other, better systems -- such as LDAP -- for that); and second, privacy.

For details, see App::Dochazka::REST::Model::Employee.

Privhistory

Dochazka has four privilege levels: admin, active, inactive, and passerby:

  • admin -- employee can view, modify, and place/remove locks on her own attendance data as well as that of other employees; she can also administer employee accounts and set privilege levels of other employees

  • active -- employee can view her own profile, attendance data, modify her own unlocked attendance data, and place locks on her attendance data

  • inactive -- employee can view her own profile and attendance data

  • passerby -- employee can view her own profile

Dochazka's privhistory object is used to track changes in an employee's privilege level over time. Each time an employee's privilege level changes, a Dochazka administrator (i.e., an employee whose current privilege level is 'admin'), a record is inserted into the database (in the privhistory table). Ordinary employees (i.e. those whose current privilege level is 'active') can read their own privhistory.

Thus, with Dochazka it is possible not only to determine not only an employee's current privilege level, but also to view "privilege histories" and to determine employees' privilege levels for any date (timestamp) in the past.

For details, see App::Dochazka::REST::Model::Privhistory and "When history changes take effect".

Schedule

In addition to actual attendance data, Dochazka sites may need to store schedules. Dochazka defines the term "schedule" as a series of non-overlapping "time intervals" (or "timestamp ranges" in PostgreSQL terminology) falling within a single week. These time intervals express the times when the employee is "expected" or "supposed" to work (or be "at work") during the scheduling period.

Example: employee "Barb" is on a weekly schedule. That means her scheduling period is "weekly" and her schedule is an array of non-overlapping time intervals, all falling within a single week.

In its current form, Dochazka is only capable of handling weekly schedules only. Some sites, such as hospitals, nuclear power plants, fire departments, and the like, might have employees on more complicated schedules such as "one week on, one week off", alternating day and night shifts, "on call" duty, etc.

Dochazka can still be used to track attendance of such employees, but if their work schedule cannot be expressed as a series of non-overlapping time intervals contained within a contiguous 168-hour period (i.e. one week), then their Dochazka schedule should be set to NULL.

For details, see App::Dochazka::REST::Model::Schedule.

Schedhistory

The schedhistory table contains a historical record of changes in the employee's schedule. This makes it possible to determine an employee's schedule for any date (timestamp) in the past, as well as (crucially) the employee's current schedule.

Every time an employee's schedule is to change, a Dochazka administrator must insert a record into this table. (Employees who are not administrators can only read their own history; they do not have write privileges.) For more information on privileges, see "AUTHORIZATION".

For details, see App::Dochazka::REST::Model::Schedhistory.

Activity

While on the job, employees "work" -- i.e., they engage in various activities that are tracked using Dochazka. The activities table contains definitions of all the possible activities that may be entered in the intervals table.

The initial set of activities is defined in the site install configuration (DOCHAZKA_ACTIVITY_DEFINITIONS) and enters the database at installation time. Additional activities can be added later (by administrators), but activities can be deleted only if no intervals refer to them.

Each activity has a code, or short name (e.g., "WORK") -- which is the primary way of referring to the activity -- as well as an optional long description. Activity codes must be all upper-case.

For details, see App::Dochazka::REST::Model::Activity.

Interval

Intervals are the heart of Dochazka's attendance data. For Dochazka, an interval is an amount of time that an employee spends doing an activity. In the database, intervals are represented using the tsrange range operator introduced in PostgreSQL 9.2.

Optionally, an interval can have a long_desc (employee's description of what she did during the interval) and a remark (admin remark).

For details, see App::Dochazka::REST::Model::Interval.

Lock

In Dochazka, a "lock" is a record in the "locks" table specifying that a particular user's attendance data (i.e. activity intervals) for a given period (tsrange) cannot be changed. That means, for intervals in the locked tsrange:

  • existing intervals cannot be updated or deleted

  • no new intervals can be inserted

Employees can create locks (i.e., insert records into the locks table) on their own EID, but they cannot delete or update those locks (or any others). Administrators can insert, update, or delete locks at will.

How the lock is used will differ from site to site, and some sites may not even use locking at all. The typical use case would be to lock all the employee's attendance data within the given period as part of pre-payroll processing. For example, the Dochazka client application may be set up to enable reports to be generated only on fully locked periods.

"Fully locked" means either that a single lock record has been inserted covering the entire period, or that the entire period is covered by multiple locks.

Any attempts (even by administrators) to enter activity intervals that intersect an existing lock will result in an error.

Clients can of course make it easy for the employee to lock entire blocks of time (weeks, months, years . . .) at once, if that is deemed expedient.

For details, see App::Dochazka::REST::Model::Lock.

CAVEATS

Weekly schedules only

Unfortunately, the weekly scheduling period is hard-coded at this time. Dochazka does not care what dates are used to define the intervals -- only that they fall within a contiguous 168-hour period. Consider the following contrived example. If the scheduling intervals for EID 1 were defined like this:

    "[1964-12-30 22:05, 1964-12-31 04:35)"
    "[1964-12-31 23:15, 1965-01-01 03:10)"

for Dochazka that would mean that the employee with EID 1 has a weekly schedule of "WED/22:05-THU/04:35" and "THU/23:15-FRI/03:10", because the dates in the ranges fall on a Wednesday (1964-12-30), a Thursday (1964-12-31), and a Friday (1964-01-01), respectively.

When history changes take effect

The effective field of the privhistory and schedhistory tables contains the effective date/time of the history change. This field takes a timestamp, and a trigger ensures that the value is evenly divisible by five minutes (by rounding). In other words,

    '1964-06-13 14:45'

is a valid effective timestamp, while

    '2014-01-01 00:00:01'

will be rounded to '2014-01-01 00:00'.

INSTALLATION

Installation is the process of creating (setting up, bootstrapping) a new Dochazka instance, or "site" in Dochazka terminology.

It entails the following steps:

  • Server preparation

    Dochazka REST needs hardware (either physical or virtualized) to run on. The hardware will need to have a network connection, etc. Obviously, this step is entirely beyond the scope of this document.

  • Software installation

    Once the hardware is ready, the Dochazka REST software and all its dependencies are installed on it. This could be accomplished by downloading and unpacking the tarball (or running git clone) and following the installation instructions, or, more expediently, by installing a packaged version of Dochazka REST if one is available (see https://build.opensuse.org/package/show/home:smithfarm/perl-App-Dochazka-REST).

  • PostgreSQL setup

    One of Dochazka REST's principal dependencies is PostgreSQL server (version 9.2 or higher). This needs to be installed (should happen automatically when using the packaged version of App::Dochazka::REST). Steps to enable it:

        bash# chkconfig postgresql on
        bash#  systemctl start postgresql.service
        bash# su - postgres
        bash$ psql postgres
        postgres-# ALTER ROLE postgres WITH PASSWORD 'mypass';
        ALTER ROLE

    At this point, we exit psql and, still as the user postgres, we edit pg_hba.conf. Using our favorite editor, we change the METHOD entry for local so it looks like this:

        # TYPE  DATABASE   USER   ADDRESS     METHOD
        local   all        all                password

    Then, as root, we restart the postgresql service:

        bash# systemctl restart postgresql.service
  • Site configuration

    Before the Dochazka REST database can be initialized, we will need to tell App::Dochazka::REST about the PostgreSQL superuser password that we set in the previous step. This is done via a site parameter. There may be other site params we will want to set, but the following is sufficient to run the test suite.

    First, create a sitedir:

        bash# mkdir /etc/dochazka

    and, second, a file therein:

        # cat << EOF > /etc/dochazka/Dochazka_SiteConfig.pm
        set( 'DBINIT_CONNECT_AUTH', 'mypass' );
        EOF
        #

    (NOTE: Strictly speaking, this sitedir setup is only needed for database initialization. During normal operation, App::Dochazka::REST connects to the database using the default user dochazka and password dochazka. These are taken from the site parameters DOCHAZKA_DBUSER and DOCHAZKA_DBPASS.)

  • Syslog setup

    It is much easier to administer a Dochazka instance if syslog is running and configured properly to place Dochazka's log messages into a separate file in a known location. In the future, App::Dochazka::REST might provide a syslog_test script to help the administrator complete this step.

  • Database initialization

    In the future, there might be a nifty dochazka-dbinit script to make this process less painful, but for now the easiest way to initialize the database is to clone the git repo from SourceForge and run the test suite:

        bash# cd ~/src
        bash# git clone git://git.code.sf.net/p/dochazka/code dochazka
        ...
        bash# cd dochazka
        bash# perl Build.PL
        bash# ./Build test

    Assuming the previous steps were completed correctly, all the tests should complete without errors.

  • Start the server

    The last step is to start the Dochazka REST service. Maybe, in the future, this will be possible using a command like systemctl start dochazka.service. Right now, though, an executable is run, as root, manually from the bash prompt:

        bash# dochazka-rest --host [HOST] --port 80 --access-log /var/log/dochazka-rest.log

    or, as any user:

        bash$ dochazka-rest
  • Take it for a spin

    Point your browser to the hostname you entered in the previous step, or to http://0:5000/ if you didn't enter a hostname.

The above procedure only includes the most basic steps. Sites with reverse proxies, firewalls, load balancers, connection pools, etc. will need to set those up, as well.

AUTHENTICATION

Since employees do not access the database directly, but only via the App::Dochazka::REST web server, the web server needs to tie all incoming requests to an EID.

All incoming requests are subject to HTTP Basic Authentication. The credentials entered by the user can be authenticated against an external database (LDAP), and internal database (PostgreSQL 'employees' table), or both.

This yields the following possible combinations: internal auth only, external auth only, internal auth followed by external auth, and external auth followed by internal auth. The desired combination can be set in the site configuration.

Current implementation

At the moment, this is accomplished via Web::Machine using HTTP Basic Authentication with a single hardcoded username/password combination demo/demo.

This allows us to use, e.g., curl like this:

    $ curl http://demo:demo@0:5000/

Possible future implementation

This is done when the session is established (see "Session management"). In the site configuration, the administrator associates an LDAP field with either EID or nick. When an employee initiates a session by contacting the server, App::Dochazka::REST first looks up the employee in the LDAP database and determines her EID, either directly or via the employee's nick. If the EID is valid, the password entered by the employee is checked against the password stored in the LDAP database.

Alternatively, App::Dochazka::REST can be configured to authenticate employees against passwords stored in the Dochazka database.

When the REST server registers an incoming request, it first checks to see if it is associated with an active session. If it is, the request is processed. If it is not, the incoming request is authenticated.

Authentication consists of:

  • a check against Dochazka's own list (database) of employees

  • an optional, additional check against an LDAP database

Depending on how the REST server is configured, one of these will include a password check. The server will send the client a session key, etc.

REPORTING

Reporting is a core functionality of Dochazka: for most sites, the entire point of keeping attendance records is to generate reports, at regular (or irregular) intervals, based on those records. One obvious use case for such reports is payroll.

That said, the REST server and its underlying database are more-or-less "reporting neutral". In other words, care was taken to make them as general as possible, to enable Dochazka to be useful in many different site and reporting scenarios.

Thus, in Dochazka a report generator is always implemented either a separate client or as part of a client. Never as part of the server.

METHODS

init

Load site configuration, set up logging, and connect to the database.

init_no_db

Load site configuration and set up logging. Intended for use from the init method as well as from App::Dochazka::REST unit tests that need to connect to the pristine database using connect_db_pristine.

Takes an optional PARAMHASH which is passed to $CELL->load. The App::Dochazka::REST distro sharedir is loaded as the first sitedir, before any sitedir specified in the PARAMHASH is loaded. Call examples:

    my $status = $REST->init_no_db;
    my $status = $REST->init_no_db( verbose => 1 );
    my $status = $REST->init_no_db( sitedir => '/etc/fooapp' );

connect_db_pristine

Connect to a pristine database. This function should be used only for newly created databases. Takes a PARAMHASH with 'dbname', 'dbuser', and 'dbpass'. For username and password, DBINIT_CONNECT_USER and DBINIT_CONNECT_AUTH are used. Returns status object which, on success, will contain the database handle in the payload.

connect_db

Connect to a pre-initialized database and initialize site params. This is the function that should be used in production. Database name, username and password are taken from DOCHAZKA_DBNAME, DOCHAZKA_DBUSER and DOCHAZKA_DBPASS, respectively.

reset_db

Drop and re-create a Dochazka database. Takes database name. Do not call when connected to an existing database. Be very, _very_, _VERY_ careful when calling this function.

create_tables

Takes a database handle, on which it executes all the SQL statements contained in DBINIT_CREATE param.

eid_of_root

Instance method. Returns EID of the 'root' employee.

GLOSSARY OF TERMS

In Dochazka, some commonly-used terms have special meanings:

  • employee -- Regardless of whether they are employees in reality, for the purposes of Dochazka employees are the folks whose attendance/time is being tracked. Employees are expected to interact with Dochazka using the following functions and commands.

  • administrator -- In Dochazka, administrators are employees with special powers. Certain REST/CLI functions are available only to administrators.

  • CLI client -- CLI stands for Command-Line Interface. The CLI client is the Perl script that is run when an employee types dochazka at the bash prompt.

  • REST server -- REST stands for ... . The REST server is a collection of Perl modules running on a server at the site.

  • site -- In a general sense, the "site" is the company, organization, or place that has implemented (installed, configured) Dochazka for attendance/time tracking. In a technical sense, a site is a specific instance of the Dochazka REST server that CLI clients connect to.

AUTHOR

Nathan Cutler, <presnypreklad@gmail.com>

BUGS

Please report any bugs or feature requests to bug-dochazka-rest at rt.cpan.org, or through the web interface at http://rt.cpan.org/NoAuth/ReportBug.html?Queue=App-Dochazka-REST. The author will be notified, and then you'll automatically be notified of progress on your bug as he makes changes.

SUPPORT

The full documentation comes with the distro, and can be comfortable perused at metacpan.org:

    https://metacpan.org/pod/App::Dochazka::REST

You can also read the documentation for individual modules using the perldoc command, e.g.:

    perldoc App::Dochazka::REST
    perldoc App::Dochazka::REST::Model::Activity

Other resources:

LICENSE AND COPYRIGHT

Copyright (c) 2014, SUSE LLC All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

3. Neither the name of SUSE LLC nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.