diff options
author | jake%acutex.net <> | 2008-04-04 13:48:06 +0200 |
---|---|---|
committer | jake%acutex.net <> | 2008-04-04 13:48:06 +0200 |
commit | 48c136f57c5a7a8a07a2d04ef22323567c2d77f4 (patch) | |
tree | 496564185705422c26bfb80b0935415679a96c2a /docs | |
parent | 82c370bdfaa089a13cf8a51898a1e0e5ccc9a1d4 (diff) | |
download | bugzilla-48c136f57c5a7a8a07a2d04ef22323567c2d77f4.tar.gz bugzilla-48c136f57c5a7a8a07a2d04ef22323567c2d77f4.tar.xz |
The recommended style for perl (which is the styleguide Bugzilla will use) specifies that else should not be cuddled. This guide, while not specifically saying it, elluded to cuddling else. Fixing that problem.
Diffstat (limited to 'docs')
-rw-r--r-- | docs/en/xml/patches.xml | 398 |
1 files changed, 319 insertions, 79 deletions
diff --git a/docs/en/xml/patches.xml b/docs/en/xml/patches.xml index 11a5ecf11..f5a9ebdfa 100644 --- a/docs/en/xml/patches.xml +++ b/docs/en/xml/patches.xml @@ -1,113 +1,354 @@ <!-- <!DOCTYPE appendix PUBLIC "-//OASIS//DTD DocBook V4.1//EN"> --> + <appendix id="patches" xreflabel="Useful Patches and Utilities for Bugzilla"> - <title>Contrib</title> + <title>Useful Patches and Utilities for Bugzilla</title> - <para> - There are a number of unofficial Bugzilla add-ons in the - <filename class="directory">$BUGZILLA_ROOT/contrib/</filename> - directory. This section documents them. - </para> + <para>Are you looking for a way to put your Bugzilla into overdrive? Catch some of the niftiest tricks here in this section.</para> - <section id="cmdline"> - <title>Command-line Search Interface</title> + <section id="rewrite" xreflabel="Apache mod_rewrite magic"> + <title>Apache <filename>mod_rewrite</filename> magic</title> + <para>Apache's <filename>mod_rewrite</filename> module lets you do some truly amazing things with URL rewriting. Here are a couple of examples of what you can do.</para> + <orderedlist> + <listitem> + <para> + Make it so if someone types + <computeroutput>http://www.foo.com/12345</computeroutput>, + Bugzilla spits back + http://www.foo.com/show_bug.cgi?id=12345. Try setting up + your VirtualHost section for Bugzilla with a rule like + this:</para> + <programlisting> +<![CDATA[ +<VirtualHost 12.34.56.78> +RewriteEngine On +RewriteRule ^/([0-9]+)$ http://foo.bar.com/show_bug.cgi?id=$1 [L,R] +</VirtualHost> +]]> + </programlisting> + + </listitem> + <listitem> + <para>There are many, many more things you can do with + mod_rewrite. As time goes on, I will include many more in + the Guide. For now, though, please refer to the mod_rewrite + documentation at <ulink + url="http://www.apache.org">http://www.apache.org</ulink></para> + </listitem> + </orderedlist> + </section> +<section id="setperl" xreflabel="The setperl.csh Utility"> + <title>The setperl.csh Utility</title> + <para> You can use the "setperl.csh" utility to quickly and + easily change the path to perl on all your Bugzilla files. This + is a C-shell script; if you do not have "csh" or "tcsh" in the + search path on your system, it will not work! + </para> + <procedure> + <step> + <para> + Download the "setperl.csh" utility to your Bugzilla + directory and make it executable. + </para> + <substeps> + <step> + <para> + <computeroutput> + <prompt>bash#</prompt> + <command>cd /your/path/to/bugzilla</command> + </computeroutput> + </para> + </step> + <step> + <para> + <computeroutput> <prompt>bash#</prompt> <command>wget -O + setperl.csh + 'http://bugzilla.mozilla.org/showattachment.cgi?attach_id=10795'</command> </computeroutput> + </para> + </step> + <step> + <para> + <computeroutput> <prompt>bash#</prompt> <command>chmod + u+x setperl.csh</command> </computeroutput> + </para> + </step> + </substeps> + </step> + <step> + <para> + Prepare (and fix) Bugzilla file permissions. + </para> + <substeps> + <step> + <para> + <computeroutput> + <prompt>bash#</prompt> + <command>chmod u+w *</command> + </computeroutput> + </para> + </step> + <step> + <para> + <computeroutput> <prompt>bash#</prompt> <command>chmod + u+x duplicates.cgi</command> </computeroutput> + </para> + </step> + <step> + <para> + <computeroutput> + <prompt>bash#</prompt> + <command>chmod a-x bug_status.html</command> + </computeroutput> + </para> + </step> + </substeps> + </step> + <step> + <para> + Run the script: + </para> + <para> + <computeroutput> <prompt>bash#</prompt> + <command>./setperl.csh /your/path/to/perl</command> + </computeroutput> +<example> + <title>Using Setperl to set your perl path</title> + <para> + <computeroutput> <prompt>bash#</prompt> + <command>./setperl.csh /usr/bin/perl</command> + </computeroutput> + </para> + </example> + </para> + </step> + </procedure> + </section> + + <section id="cmdline"> + <title>Command-line Bugzilla Queries</title> <para> - There are a suite of Unix utilities for searching Bugzilla from the - command line. They live in the - <filename class="directory">contrib/cmdline</filename> directory. - There are three files - <filename>query.conf</filename>, - <filename>buglist</filename> and <filename>bugs</filename>. + Users can query Bugzilla from the command line using this suite + of utilities. </para> - - <warning> - <para> - These files pre-date the templatisation work done as part of the - 2.16 release, and have not been updated. - </para> - </warning> - <para> - <filename>query.conf</filename> contains the mapping from - options to field names and comparison types. Quoted option names - are <quote>grepped</quote> for, so it should be easy to edit this - file. Comments (#) have no effect; you must make sure these lines - do not contain any quoted <quote>option</quote>. + The query.conf file contains the mapping from options to field + names and comparison types. Quoted option names are "grepped" + for, so it should be easy to edit this file. Comments (#) have + no effect; you must make sure these lines do not contain any + quoted "option" </para> - <para> - <filename>buglist</filename> is a shell script that submits a - Bugzilla query and writes the resulting HTML page to stdout. - It supports both short options, (such as <quote>-Afoo</quote> - or <quote>-Rbar</quote>) and long options (such - as <quote>--assignedto=foo</quote> or <quote>--reporter=bar</quote>). - If the first character of an option is not <quote>-</quote>, it is - treated as if it were prefixed with <quote>--default=</quote>. + buglist is a shell script which submits a Bugzilla query and + writes the resulting HTML page to stdout. It supports both + short options, (such as "-Afoo" or "-Rbar") and long options + (such as "--assignedto=foo" or "--reporter=bar"). If the first + character of an option is not "-", it is treated as if it were + prefixed with "--default=". </para> - <para> - The column list is taken from the COLUMNLIST environment variable. - This is equivalent to the <quote>Change Columns</quote> option - that is available when you list bugs in buglist.cgi. If you have - already used Bugzilla, grep for COLUMNLIST in your cookies file - to see your current COLUMNLIST setting. + The columlist is taken from the COLUMNLIST environment variable. + This is equivalent to the "Change Columns" option when you list + bugs in buglist.cgi. If you have already used Bugzilla, use + <command>grep COLUMLIST ~/.netscape/cookies</command> to see + your current COLUMNLIST setting. </para> - <para> - <filename>bugs</filename> is a simple shell script which calls - <filename>buglist</filename> and extracts the - bug numbers from the output. Adding the prefix - <quote>http://bugzilla.mozilla.org/buglist.cgi?bug_id=</quote> - turns the bug list into a working link if any bugs are found. - Counting bugs is easy. Pipe the results through - <command>sed -e 's/,/ /g' | wc | awk '{printf $2 "\n"}'</command> + bugs is a simple shell script which calls buglist and extracts + the bug numbers from the output. Adding the prefix + "http://bugzilla.mozilla.org/buglist.cgi?bug_id=" turns the bug + list into a working link if any bugs are found. Counting bugs is + easy. Pipe the results through <command>sed -e 's/,/ /g' | wc | + awk '{printf $2 "\n"}'</command> </para> - <para> - Akkana Peck says she has good results piping - <filename>buglist</filename> output through + Akkana says she has good results piping buglist output through <command>w3m -T text/html -dump</command> </para> - + <procedure> + <step> + <para> + Download three files: + </para> + <substeps> + <step> + <para> + <computeroutput> <prompt>bash$</prompt> <command>wget -O + query.conf + 'http://bugzilla.mozilla.org/showattachment.cgi?attach_id=26157'</command> </computeroutput> + </para> + </step> + <step> + <para> + <computeroutput> <prompt>bash$</prompt> <command>wget -O + buglist + 'http://bugzilla.mozilla.org/showattachment.cgi?attach_id=26944'</command> </computeroutput> + </para> + </step> + <step> + <para> + <computeroutput> <prompt>bash#</prompt> <command>wget -O + bugs + 'http://bugzilla.mozilla.org/showattachment.cgi?attach_id=26215'</command> </computeroutput> + </para> + </step> + </substeps> + </step> + <step> + <para> + Make your utilities executable: + <computeroutput> + <prompt>bash$</prompt> + <command>chmod u+x buglist bugs</command> + </computeroutput> + </para> + </step> + </procedure> </section> - <section id="cmdline-bugmail"> - <title>Command-line 'Send Unsent Bug-mail' tool</title> - + <section id="quicksearch"> + <title>The Quicksearch Utility</title> <para> - Within the <filename class="directory">contrib</filename> directory - exists a utility with the descriptive (if compact) name - of <filename>sendunsentbugmail.pl</filename>. The purpose of this - script is, simply, to send out any bug-related mail that should - have been sent by now, but for one reason or another has not. + Quicksearch is a new, experimental feature of the 2.12 release. + It consist of two Javascript files, "quicksearch.js" and + "localconfig.js", and two documentation files, + "quicksearch.html" and "quicksearchhack.html" </para> - <para> - To accomplish this task, <filename>sendunsentbugmail.pl</filename> uses - the same mechanism as the <filename>sanitycheck.cgi</filename> script; it - it scans through the entire database looking for bugs with changes that - were made more than 30 minutes ago, but where there is no record of - anyone related to that bug having been sent mail. Having compiled a list, - it then uses the standard rules to determine who gets mail, and sends it - out. + The index.html page has been updated to include the QuickSearch + text box. + </para> + <para> + To take full advantage of the query power, the Bugzilla + maintainer must edit "localconfig.js" according to the value + sets used in the local installation. + </para> + <para> + Currently, keywords must be hard-coded in localconfig.js. If + they are not, keywords are not automatically recognized. This + means, if localconfig.js is left unconfigured, that searching + for a bug with the "foo" keyword will only find bugs with "foo" + in the summary, status whiteboard, product or component name, + but not those with the keyword "foo". + </para> + <para> + Workarounds for Bugzilla users: + <simplelist> + <member>search for '!foo' (this will find only bugs with the + keyword "foo"</member> + <member>search 'foo,!foo' (equivalent to 'foo OR + keyword:foo')</member> + </simplelist> </para> - <para> - As the script runs, it indicates the bug for which it is currently - sending mail; when it has finished, it gives a numerical count of how - many mails were sent and how many people were excluded. (Individual - user names are not recorded or displayed.) If the script produces - no output, that means no unsent mail was detected. + When this tool is ported from client-side JavaScript to + server-side Perl, the requirement for hard-coding keywords can + be fixed. <ulink + url="http://bugzilla.mozilla.org/show_bug.cgi?id=70907">This bug</ulink> has details. </para> + </section> + <section id="bzhacking"> + <title>Hacking Bugzilla</title> <para> - <emphasis>Usage</emphasis>: move the sendunsentbugmail.pl script - up into the main directory, ensure it has execute permission, and run it - from the command line (or from a cron job) with no parameters. + What follows are some general guidelines for changing Bugzilla, and adhering to good coding practice while doing so. We've had some checkins in the past which ruined Bugzilla installations because of disregard for these conventions. Sorry for the lack of formatting; I got this info into the Guide on the day of 2.14 release and haven't formatted it yet. </para> + <literallayout> + +The following is a guide for reviewers when checking code into Bugzilla's +CVS repostory at mozilla.org. If you wish to submit patches to Bugzilla, +you should follow the rules and style conventions below. Any code that +does not adhere to these basic rules will not be added to Bugzilla's +codebase. + + 1. Usage of variables in Regular Expressions + + It is very important that you don't use a variable in a regular + expression unless that variable is supposed to contain an expression. + This especially applies when using grep. You should use: + + grep ($_ eq $value, @array); + + - NOT - + + grep (/$value/, @array); + + If you need to use a non-expression variable inside of an expression, be + sure to quote it properly (using \Q..\E). + +Coding Style for Bugzilla +------------------------- + +While it's true that not all of the code currently in Bugzilla adheres to +this styleguide, it is something that is being worked toward. Therefore, +we ask that all new code (submitted patches and new files) follow this guide +as closely as possible (if you're only changing 1 or 2 lines, you don't have +to reformat the entire file :). + + 1. Whitespace + + Bugzilla's prefered indentation is 4 spaces (no tabs, please). + + 2. Curly braces. + + The opening brace of a block should be on the same line as the statement + that is causing the block and the closing brace should be at the same + indentation level as that statement, for example: + + if ($var) { + print "The variable is true"; + } + else { + print "Try again"; + } + + - NOT - + + if ($var) + { + print "The variable is true"; + } + else + { + print "Try again"; + } + + 3. File Names + + File names for bugzilla code and support documention should be legal across + multiple platforms. \ / : * ? " < > and | are all illegal characters for + filenames on various platforms. Also, file names should not have spaces in + them as they can cause confusion in CVS and other mozilla.org utilities. + + 4. Variable Names + + If a variable is scoped globally ($::variable) its name should be descriptive + of what it contains. Local variables can be named a bit looser, provided the + context makes their content obvious. For example, $ret could be used as a + staging variable for a routine's return value as the line |return $ret;| will + make it blatently obvious what the variable holds and most likely be shown + on the same screen as |my $ret = "";|. + + 5. Cross Database Compatability + + Bugzilla was originally written to work with MySQL and therefore took advantage + of some of its features that aren't contained in other RDBMS software. These + should be avoided in all new code. Examples of these features are enums and + encrypt(). + + 6. Cross Platform Compatability + + While Bugzilla was written to be used on Unix based systems (and Unix/Linux is + still the only officially supported platform) there are many who desire/need to + run Bugzilla on Microsoft Windows boxes. Whenever possible, we should strive + not to make the lives of these people any more complicated and avoid doing things + that break Bugzilla's ability to run on multiple operating systems. + + </literallayout> </section> </appendix> + <!-- Keep this comment at the end of the file Local variables: mode: sgml @@ -123,9 +364,8 @@ sgml-local-ecat-files:nil sgml-minimize-attributes:nil sgml-namecase-general:t sgml-omittag:t -sgml-parent-document:("Bugzilla-Guide.xml" "book" "chapter") +sgml-parent-document:("Bugzilla-Guide.sgml" "book" "chapter") sgml-shorttag:t sgml-tag-region-if-active:t End: --> - |