Koha Coding Standards and Guidelines for Contributors
As in other free and open source software projects, Koha contains code that has been contributed to the project by many different developers. Koha is now a complex piece of software, and we'd like continue to
improve the overall code standards to make it easier for new developers to join the project, and to make existing code easier to maintain. These standards and guidelines are intended to give Koha a reasonably consistent style throughout the code and to ensure maintainability. When you commit new code to Koha, please do your best to comply with these guidelines -- it will increase the chances of your code become part of a stable Koha release. And if you find code in Koha that doesn't comply with these guidelines, please feel free to fix it, just as you would correct any bug. Koha code should conform to the guidelines defined in the perlstyle man page as they apply to code details (indenting, etc.). As for the general structure of t
he code, all Koha code must be separated into three types: These three code types, as well as the Koha naming conventions, are explained in the following sections. All code controlling the "appearance layer" is contained in HTML::Template files. These files are located in the koha-tmpl directory in CVS (currently http://sourceforge.net/cvs/?group_id=16466). They must contain all the code controlling the presentation layer. There are at least two good reasons for this: the look of Koha ca
n be changed by rewriting only the templates, changing nothing in the Perl code itself; and there is a tool to translate templates, so it's the only way to localise our preferred ILS.
Almost all the actual Koha presentation pages use templates. Only a few exceptions (that can be considered bugs) remain at the time of writing. 2.1.1. Templates locationThe koha-tmpl directory contains two different sub-directories: opac-tmpl and intranet-tmpl. Under those two are the themes directories, and one directory for each language -- for example, intranet-tmpl/default/en for English default templates. The templates should be stored in a directory that parallels the location of the Perl scripts. For
example, the Perl scripts located in the $KOHA/acqui.simple directory should use templates located in $KOHA/koha-tmpl/intranet-tmpl/<theme>/<language>/acqui.simple. Exceptions: Scripts located in $KOHA/admin/ use templates located in $KOHA/intranet-tmpl/<theme>/<language>/parameters/ Some scripts in $KOHA/members/ use templates located in $KOHA/intranet-tmpl/<theme>/<language>/ Simple HTML "includes" files for placing standard headers and footers on Koha pages are located in $KOHA/intranet-tmpl/<theme>/<language>/includes/
When modifying templates, you should only modify the
English one manually. All other languages are automatically generated from the English templates through a .po file and the script located in misc/translator/tmpl_process3.pl. The $KOHA/koha-tmpl/<theme>/<language>/help directory contains online help. Under the help directory is a complete copy of the Koha directory structure, with templates using the same name as the "true" templates. These templates must contain online help. The $KOHA/help.pl script is called when the user clicks on "Help," which opens/shows the templates in this directory. PLEASE write online help when you add features. There is usually only one template for a given .pl file. For example, the template related to circulation/returns.pl should be circulation/returns.tmpl. If there are different possible behaviours, they should be separated by a TMPL_IF. For example, $KOHA/koha-tmpl/intranet-tmpl/<theme>/<language>/parameters contains: <!-- TMPL_IF name="add_form" --> that controls what is shown when the user wants to add something (as determined by the Perl script): if ($op) {
$template->param(script_name => $script_name,
$op => 1); # We show only the TMPL_VAR names $op.
} else {
$template->param(script_name => $script_name,
else => 1); # We show only the TMPL_VAR names $op.
}Note that some scripts use two different templates, depending on what the user wants to do. This is sometimes a mistake in Koha design, sometimes not. For example, in $KOHA/opac/opac-search.pl, we can use the opac-search or opac-searchresults template, depending on the step of the search. Both templates are big, so merging them would not be a good idea for maintenance and readability of Koha code. No database (DB) access routines should ever be included in .pl files. All DB access routines should be in the C4 directory, in .pm modules. There is one .pm module for each logical "librarian activity." Fo
r example, everything related to circulation must be in C4/Circulation/Circ2.pm. All inclusion of Perl variables in SQL commands should be done through prepare() statements with ? placeholders, and then using bind() and/or execute() calls. This helps with security and other testing. The C4/Context.pm module contains everything related to technical variables (creating a DB handler, getting a systempreference, etc.). The C4/Koha.pm module contains everything related to things that are useful everywhere in Koha, such as getting a branch list or a branch detail. (At the time of writing, this module probably should be improved. Some subroutines, for instance, are located in other modul
es but would be better here.) Exception: parameters management scripts. For historic reasons, and because they don't do anything complex in the database, all scripts in $KOHA/admin/ deal directly with the database. This is not really a bug, even if it does seem to break the rule. (Think of it as "the exception that proves the rule.") Perl scripts are the scripts that are run from cgi-bin (or mod_perl). They all have (almost) the same behaviour: read parameters; open an HTML::Template file (that checks the user rights to access the page if needed -- if the user doesn't have the correct permi
ssions, show a "sorry no access" page and end); depending on the parameters (often, in a $op variable, but not always), run subroutines from a C4/*.pm package to do something in the database; using the parameters and the results of the previous operations, fill the template; and send the template to the user.
Some scripts are run on the command line. They all are located in the misc directory. The misc directory does not contain any scripts that can be run as cgi-bin. Command line scripts executed without any parameters should return a "help" listing of possible parameters and exit without doing anything. The script names must be related to what they do, and they must be located in a directory that is related to a logical "librarian activity". For example, all scripts related to circulation are in $KOHA/circ/. Exceptions: The acqui.simple directory contains all the code for MARC cataloguing. The directory name obviously is not related to the "librarian activity," but it's historic. The bull directory is related to serials management. Most of the code in this directory was developed in France, and "serials management" is "bulletinage" in French -- but in general, script and directory names should be in English. Some scripts related to borrowers are still in the $KOHA/ directory (pay.pl for example).
Variables can be of three types: All three variables should have the same name. For example, if you get the field issuelength from the issuingrules table, then you should have a <!-- TMPL_VAR name="issuelength" --> in the template, and either a $issuelength or $issuingrules->{'issuelength'} variable i
n the Perl code. Sometimes you must use the same field/variable from two different points of view. In this case, use two variable names, with something distinctive before the "simple" name. For example if you need "librarian name" and "borrower name" from borrowers.surname, use librariansurname and borrowersurname. Scripts and modules in Koha are expected to contain clear and helpful comments to assist other developers who want to work on the code. Despite the availability of external documentation, source code listings should be able to stand on their own, because external documentation may not adequately describe the details and intent of the developer(s). Please make sure you include enough
comments to keep your coding intentions understandable, and try to follow these guidelines: Please write your comments in English or request translation. Write in complete sentences, capitalize the first word, and put two spaces after the end of each sentence (so that the Emacs sentence commands will work). Comments should start with a # and a single space. Comment your code when you do something that someone else may not think is "trivial." In other words, comment anything that is not readily obvious in the code. Each Koha file should begin with a block comment containing: basic information about the file, such as file name, version, date, author: # addbiblio.pl, v 1.52.2.6, 2005/05/19, tipaul the standard Koha copyright message: # Copyright 2000-2005 Katipo C
ommunications
# Copyright 2005 <author or employer>
#
# This file is part of Koha.
#
# Koha is free software; you can redistribute it and/or modify it under the
# terms of the GNU General Public License as published by the Free Software
# Foundation; either version 2 of the License, or (at your option) any later
# version.
#
# Koha is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
# A PARTICULAR PURPOSE. See the GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along with
# Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
# Suite 330, Boston, MA 02111-1307 USA
All block comments should consist of one or more paragraphs built out of complete sentences, and each sentence should end in a period. Block comments generally apply to some (or all) code that follows them, and are indented to the same level as that code. Each line of a block comment starts with a # and a single space (unless it is indented text inside the comment). Paragraphs inside a block comment are separated by a line containing a single #. Block comments are best surrounded by a blank line above and below them. Provide comments at the beginning of every function (unless the function is very short and its purpose is obvious) indicating the function's purpose, assumptions, and limitations. The comments should be a brief introduction to understand why the function exists and what it can do. Document how it does it, too, if that is not obvious. Place a blank line betwee
n a function and its description. Include information about the sorts of arguments it gets, and what the possible values of arguments mean and are used for. Also explain the significance of the return value, if there is one. Use comments on code that consists of loops and logic branches. These are key areas that will assist the reader when reading source code. Try to avoid adding comments at the end of a line of code; end-line comments make code more difficult to read. End-line comments are appropriate, however, when annotating variable declarations. In this case, try to align all end-line comments at a common tab stop. (End-line comments should be separated by at least two spaces from the statement.) When modifying code, always keep the commenting around it up to date. Keeping the comments up-to-date when the code changes is a priority -- comments that contradict the code are worse than no comments! Identify the chang
e with the date and your user name. Avoid using clutter comments, such as an entire line of asterisks. Instead, use white space to separate comments from code. Avoid the use of superfluous or inappropriate comments, such as humorous sidebar remarks. To prevent recurring problems, always use comments on bug fixes and work-around code, especially in this team environment. When you want to leave notes about functionalities to be added, put TODO: in front of the actual comments so developers can easily search for them. When you want to leave notes about bugs to be fixed, put FIXME: in front of the actual comments so developers can easily search for them.
One other type of comment that developers may overlook is the cvs log message added when committing a file, but these comments are also very important. The messages
should never be omitted and should give enough description to let other developers know what you are doing without having to read the code itself. In the case of bug fixes, it is also recommended that the messages include the bug number and the bug title, as well as a brief description of the fix. 4. POD style (in modules)Koha modules should be documented with extensive comments written in POD ("Plain Old Documentation") format. Instructions for the POD markup language can be found in the perlpod man page. POD comments should include as least the following headings: - NAME
The name of the module and a very brief description of the type of functions contained in the module.<
/p> - SYNOPSIS
This may give an example of how to access the module from a script, but it may also contain any general discussion of the purpose or behavior of the module's functions, and discussion of changes made to the module. - DESCRIPTION
If not included in the SYNOPSIS, a general description of the module's purpose should appear here. - FUNCTIONS
Descriptions of each function in the module, beginning with an example of how the function is called and continuing with a description of the behavior of the function. - AUTHOR
This will generally be the "Koha Development Team," but may also be an individual developer who has made significant contributions to the module. Give a contact e-mail address.
For two examples, see the POD comments in the Searc
h.pm module (perldoc Search.pm) and the Biblio.pm module. |