Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(2)

Issue 5848066: code review 5848066: run.bat: disable test in test\bench\go1 to fix build (Closed)

Can't Edit
Can't Publish+Mail
Start Review
Created:
13 years, 1 month ago by brainman
Modified:
13 years, 1 month ago
Reviewers:
aram
CC:
golang-dev, r
Visibility:
Public.

Description

run.bat: disable test in test\bench\go1 to fix build

Patch Set 1 #

Patch Set 2 : diff -r 48d9395a3b95 https://go.googlecode.com/hg/ #

Patch Set 3 : diff -r 48d9395a3b95 https://go.googlecode.com/hg/ #

Patch Set 4 : diff -r 0449267813c1 https://go.googlecode.com/hg/ #

Unified diffs Side-by-side diffs Delta from patch set Stats (+5 lines, -4 lines) Patch
M src/run.bat View 1 2 1 chunk +5 lines, -4 lines 0 comments Download

Messages

Total messages: 6
brainman
Hello golang-dev@googlegroups.com, I'd like you to review this change to https://go.googlecode.com/hg/
13 years, 1 month ago (2012-03-20 03:48:22 UTC) #1
r
LGTM Really?
13 years, 1 month ago (2012-03-20 03:57:32 UTC) #2
brainman
*** Submitted as http://code.google.com/p/go/source/detail?r=fbc982f5aff1 *** run.bat: disable test in test\bench\go1 to fix build R=golang-dev, r ...
13 years, 1 month ago (2012-03-20 03:59:58 UTC) #3
brainman
On 2012/03/20 03:57:32, r wrote: > Really? I hope my temp fix will work. I ...
13 years, 1 month ago (2012-03-20 04:02:22 UTC) #4
aram
> > I never had to deal with paths that long. I am pretty sure, ...
13 years, 1 month ago (2012-03-20 13:13:58 UTC) #5
aram
13 years, 1 month ago (2012-03-23 11:13:58 UTC) #6
To answer your comments on issue 3358:
>
>
Making the normalization ourselves is a mistake because we can't mimic the 
side effects
of a relative path hitting the I/O manager and because we'd have to ignore 
that the Win32
namespace is private to a session.

If you used the real syscall wrapper (ZwCreateFile) instead of the Win32 
CreateFile,
you could use the path as is and not have the the mentioned normalization 
issues,
nor the path length problem. You'd have the problem of initializing the 
RootDirectory handle
in the OBJECT_ATTRIBUTES structure, but that's easier than making the 
normalization
ourselves.

I'm not advocating for the latter either. I think that the current behavior 
is the correct
thing to do.
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld f62528b