[Webkit-unassigned] [Bug 56286] New: layout test checksums should be embedded in the png as a comment

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Sun Mar 13 20:58:20 PDT 2011


https://bugs.webkit.org/show_bug.cgi?id=56286

           Summary: layout test checksums should be embedded in the png as
                    a comment
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: PC
        OS/Version: Mac OS X 10.5
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Tools / Tests
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: tony at chromium.org
                CC: eric at webkit.org, dglazkov at chromium.org,
                    ojan at chromium.org, mihaip at chromium.org


In bug 55236, Mihai listed a bunch of layout tests where we have .checksums without .pngs and vice versa.  It occurred to me that we can just embed the .checksum value in the pngs as a comment and remove the .checksum files from the tree.

For example, when DRT generates the actual pngs, it could place the checksum in the png comment.  NRWT could be updated to look for a .checksum file and if one isn't found, open the png and read the checksum from it (an iTXt section can be near the beginning of the file: http://en.wikipedia.org/wiki/Portable_Network_Graphics#Ancillary_chunks ).

Pros:
- less files in the checkout (should make svn/git operations faster)
- harder for .checksums and .pngs to get out of sync

Cons:
- harder to view/manually edit the checksum, e.g. it won't be in the diff.  I'm not sure anyone ever actually does this.
- might make NRWT a tad bit slower since we have to check to see if .checksum exists first and we have to scan the png for the checksum, although I bet this will be negligible.

I think I could rolls this out gradually, starting with the chromium port to see if it works.

Thoughts?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list