bferreirainfo

We could abstract our solution in 5 steps
step 1: receive a goal and starts respective report
step 2: Read log line and identify event type(kill event, suicide event ,etc)
step 3: parse line based on event type
step 4: call respective computing behaviour from report(addKill(killer,killed),etc) passing parsed data.
step 5: parse reports data to output format and output result

Step 1: To package everything we can create a class "QuakeReportApp" that has the "main method" and is the front door of our app, it receives a goal like "game summary, ranking, web ranking " validate it comparing with constants - and as we are going to use these constants after we should put them in class QuakeReportGoals - if that goal is invalid, show error at console, if oks them starts a report X (X = is an implementation of report) determined by the goal.

Steps 2 to - yeah that happened - 5 : is the flux of read/process/output and is common to ours reports so we could abstract it with "ReportBuilderAbstract" and define all (2-5) s

bferreirainfo

<link href="../core-scaffold/core-scaffold.html" rel="import">
<link href="../core-header-panel/core-header-panel.html" rel="import">
<link href="../core-menu/core-menu.html" rel="import">
<link href="../core-item/core-item.html" rel="import">
<link href="../core-icon-button/core-icon-button.html" rel="import">
<link href="../core-toolbar/core-toolbar.html" rel="import">
<link href="../core-menu/core-submenu.html" rel="import">
<link href="../core-menu-button/core-menu-button.html" rel="import">
<link href="../core-icons/core-icons.html" rel="import">
<link href="../paper-tabs/paper-tabs.html" rel="import">