BAM - Bugzilla automated monitor
=============================

BAM is a tool to create daily statistics of bugs from Bugzilla.
Tool fetches the snapshot of current statuses of selected bugs according to rules defined in the configuration and:
- generate the information about number of unconfirmed / open / resoved / unreleased / unverified / active bugs.
- compare the snapshot with the previous one and generate the information about changes in bug statuses
  (how many bugs have been reported / resolved / released / verified).

The basic idea is that statistics are generated ONLY once a day early in the morning and presents:
- the status in the time when they have been generated.
- changes in bugs' statuses in _previous_ day.
So e.g.:
- the statistics generated on Tuesday morning, week 7 are visible on the web page as statistics from "Week 7, Mon".
- the statistics generated on Monday morning, week 8 are visible on the web page as statistics from "Week 7, Sun"
  (or e.g. "Week 7, Fri-Sun", when statistics are taken in working days only).
Tool is checking wheter metrics has been already feched in current day and dissalow fetching them again. 

However there is also possibility to generate 'online' metrics by defining 'RUN_FEW_TIMES_A_DAY = true' parameter.
Then statistics can be generated eg. every 1 hour / 20 minutes / ? (as you define it in the cron file) and presents:
- the status in the time when they have been generated.
- changes in bugs' statuses in _current_ day.
So e.g.:
- the statistics generated on Tuesday at 10:30, week 7 are visible on the web page as statistics from "Week 7, Tue, 10:30".
- when statistics will be fetched once again at e.g. 13:20, then row "Week 7, Tue, 10:30" will be replaced with row "Week 7, Tue, 13:20"
- last statistic fetched on Tuesday will stay as a dayli statistics in row "Week 7, Tue".



CONFIGURATION:
=============================

1/ Create the user for fetching data from Bugzilla, e.g. bugstats

2/ Copy the content of the 'src' to correct folders

3/ Configure Apache

4/ Create configuration files for statistics you want to fetch.


------------------
Apache:

1/ PHP module must be installed for apache

2/ GD extension must be enabled in php

3/ Fonts (to generate graphs): packages: ttf-dejavu, ttf-dejavu-core
   You can change font used on graphs in lib/static_variables.php

4/ parameters file: lib/static_variables.php
   You must update variables of default strings visible on page.
   If you don't do that then each configurable string will be preceded with "!!" characters.



------------------
perl script:

1/ Prepare configuration files (see example files).
   You can generate the products list file by executing 'check_product_list_of_classification.pl' script with additional parameters.

2/ Add proper line to the /usr/local/bin/fetch_statistics.sh file.

3/ Set credentials to access the Bugzilla in .netrc file of the home directory of the user running perls script.
   Format: machine projects.maemo.org login xxxxx password xxxxxx
   chmod 600 .netrc

4/ Create the 'data' folder for statistics (the one defined under STATISTICS_BASE_PATH variable) and ensure that user running perl scripts has access to write to this folder.

5/ Create the cron file (see example file); define EMAIL [, REPLYTO, proxy, if you need] in the cron file.
   In the example cron executes the /usr/local/bin/fetch_statistics.sh script, in which is defined what reall should be done.


------------------
To create new statistics:

1/ create new config file(s).

2/ add new line to the /usr/local/bin/fetch_statistics.sh file.

================================