NAME
    App::Dochazka::REST - Dochazka REST server

VERSION
    Version 0.117

  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 *REST*ful 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-Doch
        azka-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:

    * RT: CPAN's request tracker (report bugs here)
        http://rt.cpan.org/NoAuth/Bugs.html?Dist=App-Dochazka-REST

    * AnnoCPAN: Annotated CPAN documentation
        http://annocpan.org/dist/App-Dochazka-REST

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.

