A couple of feature requests

[Alexey Neyman]

Hi all,

1. Is it possible to make the reviews contain remarks (possibly marked as
issues) which are not tied to a particular code location? The process
requirements in our company require a formal description of the changes, which
is also reviewed - and while there is a textbox in the review form to add such
global comments, there is no way to track them - or respond to particular
questions/issues.

I have attached a mock-up of what I mean.

2. On the admin page for a user, is it possible to add an editable list of
review groups the user belongs to?

Regards,
Alexey.

[David Trowbridge]

Alexey,

We're pretty swamped for time right now, so I can't give you any timeframe for when we might get to these (although we can make time for them if your organization would like to help fund their development). We'd like to make sure that we don't forget about them, though.

We already have an open enhancement request that covers feature request number 1: https://code.google.com/p/reviewboard/issues/detail?id=2817

Would you mind attaching your mockup there?

Also, could you file a ticket in the bug tracker for your second request?

Thanks!

-David

--

---
You received this message because you are subscribed to the Google Groups "reviewboard-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard-d...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Matthew Woehlke]

On 2013-04-30 19:34, Alexey Neyman wrote:
> 1. Is it possible to make the reviews contain remarks (possibly marked as
> issues) which are not tied to a particular code location? The process
> requirements in our company require a formal description of the changes, which
> is also reviewed - and while there is a textbox in the review form to add such
> global comments, there is no way to track them - or respond to particular
> questions/issues.

While I support the above feature request (indeed, I think I am watching
the existing issue that David mentioned), the use case you describe
could also be handled by including the commit message or log as part of
the review request (as a fake file addition), which would allow it to be
reviewed in the same manner as code.

This would require your developers to use a wrapper over rbt, but it
isn't too difficult to do this (I am already doing something like this
with git).

--
Matthew

[David Trowbridge]

You're not just watching that issue, you're the one who filed it :)

-David

--

--- You received this message because you are subscribed to the Google Groups "reviewboard-dev" group.

To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard-dev+unsubscribe@googlegroups.com.

[Alexey Neyman]

On Tuesday, April 30, 2013 4:46:22 PM UTC-7, Matthew Woehlke wrote:

On 2013-04-30 19:34, Alexey Neyman wrote:
> 1. Is it possible to make the reviews contain remarks (possibly marked as
> issues) which are not tied to a particular code location? The process
> requirements in our company require a formal description of the changes, which
> is also reviewed - and while there is a textbox in the review form to add such
> global comments, there is no way to track them - or respond to particular
> questions/issues.

While I support the above feature request (indeed, I think I am watching
the existing issue that David mentioned), the use case you describe
could also be handled by including the commit message or log as part of
the review request (as a fake file addition), which would allow it to be
reviewed in the same manner as code.

... and, speaking of 'rbt' utility: I tried to use it, but it just spits the following error message:

[aneyman@build2 src]$ rbt post

CRITICAL:root:can only concatenate list (not "tuple") to list

Is it supposed to do something more useful? :)

Regards,

Alexey.

[Christian Hammond]

Nope, that's all it does! :)

0.5.0 was a completely new codebase (minus post-review), so there's a bunch of stuff we're fixing up. We should be doing 0.5.1 soon.

Can you run with --debug and paste the traceback?

Christian

[Matthew Woehlke]

On 2013-04-30 19:50, David Trowbridge wrote:
> You're not just watching that issue, you're the one who filed it :)

Ah... so I did. :-)

(I guess it /is/ in the number range with most of my other reports.
Seems I was misremembering... and I admit I didn't check. Ah, well :-).)

--
Matthew

[Matthew Woehlke]

On 2013-04-30 22:16, Alexey Neyman wrote:
> ... and, speaking of 'rbt' utility: I tried to use it, but it just spits
> the following error message:
>
> [aneyman@build2 src]$ rbt post
> CRITICAL:root:can only concatenate list (not "tuple") to list
>
> Is it supposed to do something more useful? :)

Probably :-).

Is there a complete traceback associated with this? Or just the above
error? Also, what python version are you using?

--
Matthew

[Alexey Neyman]

Sure:

$ rbt post --debug

DEBUG:root:Checking for a Bazaar repository...

DEBUG:root:Running: bzr info

DEBUG:root:Command exited with rc 3: ['bzr', 'info']

bzr: ERROR: Not a branch: "<WC_PATH>".

---

DEBUG:root:Checking for a CVS repository...

DEBUG:root:Checking for a ClearCase repository...

DEBUG:root:Checking for a Git repository...

DEBUG:root:Running: git rev-parse --git-dir

DEBUG:root:Command exited with rc 128: ['git', 'rev-parse', '--git-dir']

fatal: Not a git repository (or any of the parent directories): .git

---

DEBUG:root:Checking for a Mercurial repository...

DEBUG:root:Running: hg showconfig

DEBUG:root:Running: hg root

DEBUG:root:Command exited with rc 255: ['hg', 'root']

abort: There is no Mercurial repository here (.hg not found)!

---

DEBUG:root:Checking for a Perforce repository...

DEBUG:root:Checking for a Plastic repository...

DEBUG:root:Checking for a Subversion repository...

DEBUG:root:Running: svn info --non-interactive

DEBUG:root:Running: diff --version

DEBUG:root:repository info: Path: <REPO_URL>, Base path: <WC_PATH>, Supports changesets: False

DEBUG:root:Running: svn propget reviewboard:url <REPO_URL>

DEBUG:root:Making HTTP GET request to http://<RB_HOST>/api/

DEBUG:root:Making HTTP GET request to http://<RB_HOST>/api/info/

Traceback (most recent call last):

  File "/usr/local/bin/rbt", line 9, in <module>

    load_entry_point('RBTools==0.5', 'console_scripts', 'rbt')()

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/main.py", line 99, in main

    command.run_from_argv([RB_MAIN] + args)

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/__init__.py", line 150, in run_from_argv

    exit_code = self.main(*args) or 0

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/post.py", line 432, in main

    files=args)

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/utils/diffs.py", line 12, in get_diff

    diff, parent_diff = scmtool.diff(files)

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/clients/svn.py", line 108, in diff

    return (self.do_diff(["svn", "diff", "--diff-cmd=diff"] + files),

TypeError: can only concatenate list (not "tuple") to list

Casting files to list ("list(files)") makes it proceed further, but it fails with:

Traceback (most recent call last):

  File "/usr/local/bin/rbt", line 9, in <module>

    load_entry_point('RBTools==0.5', 'console_scripts', 'rbt')()

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/main.py", line 99, in main

    command.run_from_argv([RB_MAIN] + args)

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/__init__.py", line 150, in run_from_argv

    exit_code = self.main(*args) or 0

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/post.py", line 450, in main

    submit_as=self.options.submit_as)

  File "/usr/local/lib/python2.6/dist-packages/RBTools-0.5-py2.6.egg/rbtools/commands/post.py", line 316, in post_request

    raise CommandError("Error creating review request: %s" % e)

rbtools.commands.CommandError: Error creating review request: The repository path specified is not in the list of known repositories (HTTP 400, API Error 206)

The reason for this, I guess, is somewhat peculiar setup we're using. In our setup, RB is hosted on the same machine as the SVN repository, so to speed up its access to SVN, the repository is configured with a file:///svn/project-svn URL. The developers, though, use a different URL to check-out their working copies: svn://server/svn/project-svn. The post-review was able to match these two URLs, but 'rbt post' apparently is not so capable.

Regards,

Alexey.

[Steven MacLeod]

The easiest way to solve this problem is to set 'REPOSITORY = "<repository-name>"' in your .reviewboardrc, where <repository-name> is the name given to the repository in the Review Board admin panel.

The method for matching was changed in rbt, and I'll look into making it more robust. Writing the new tools was, and still is, a balancing act between having a more maintainable code base and re-using the old battle tested post-review code. There are a number of improvements coming in 0.5.1 which as Christian said, should be ready soon.

--

 
---
You received this message because you are subscribed to the Google Groups "reviewboard-dev" group.

To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard-d...@googlegroups.com.

[Alexey Neyman]

Is there a way to do so via SVN property, similar to reviewboard:url for specifying the URL to the RB itself?

Regards,

Alexey.