Bug 76215

Summary: webkitpy: make mapping a port name to a port completely data-driven
Product: WebKit Reporter: Dirk Pranke <dpranke>
Component: New BugsAssignee: Dirk Pranke <dpranke>
Status: RESOLVED FIXED    
Severity: Normal CC: abarth, ojan, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Bug Depends on: 76356, 76357, 76360    
Bug Blocks: 75037    
Attachments:
Description Flags
Patch none

Dirk Pranke
Reported 2012-01-12 14:26:56 PST
webkitpy: make mapping a port name to a port completely data-driven
Attachments
Patch (50.05 KB, patch)
2012-01-12 14:38 PST, Dirk Pranke 		no flags
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Dirk Pranke
Comment 1 2012-01-12 14:38:03 PST
Created attachment 122312 [details] Patch
Dirk Pranke
Comment 2 2012-01-12 14:42:14 PST
Okay, this is roughly the last patch needed to finish cleaning up how ports are constructed (at least my best stab at it). It's a bit large; I could break the cleanup in factory.py out into a separate patch but I thought leaving it in would be helpful to show the final end state as you're looking at the other changes (including how port_name and determine_full_port_name are used). With this change, adding a new implementation requires adding one line to factory.py (in the PORT_CLASSES list) plus implementing the actual port class. Also, with the exception of supporting the --chromium flag and the 'chromium-gpu' port name, the only non-data-driven logic in the factory class is the logic for picking a default port for a given platform. Thoughts?
Adam Barth
Comment 3 2012-01-12 14:52:30 PST
I'm excited about the idea behind this patch, but this patch is slightly larger than I can chew on at the moment. I'll look again this evening when things are less hectic.
Dirk Pranke
Comment 4 2012-01-12 15:25:08 PST
no hurry, and I'll be able to try and split it up into two or three chunks if that'll help. Let me know after you take another (quick?) look.
Dirk Pranke
Comment 5 2012-01-12 15:28:54 PST
I will note that this is the last patch in the series so it's not blocking anything else I'm working on. We can take our time w/ this ...
Dirk Pranke
Comment 6 2012-01-17 12:33:47 PST
closing; all dependent bugs have landed.
Note You need to log in before you can comment on or make changes to this bug.