On Aug 25, 2009, at 8:29 AM, Leonard Richardson wrote:
> Review: Needs Information
>> Fixes testing infrastructure and packaging problems.
>
> I have two comments.
>
> 1. Why the heck is 'roman' a dependency of lazr.restful? It's not in
> the code and it's such a random thing to need.
I was surprised too, but my tests failed in a clean Python (that is, a
vanilla Python with only the standard library) without it. That said,
I investigated further after you asked. It turns out that docutils
tries to do something tricky, and my version of docutils had broken
the trick. I rebuilt docutils and we're all good. Therefore, I
removed that indirect dependency from the list.
> 2. This comment isn't terribly helpful.
>
> 70 + # We want the initial slash of the path, so we subtract 1 from
> 71 + # the base_url's len.
>
> I'd prefer something like:
>
> # We want just one character of the base_url (the slash that begins
> the path), so we cut
> # off the length of the base_url minus one character.
Yes, good improvement. I took it a little further, as you can see in
the incremental diff below.
=== modified file 'src/lazr/restful/testing/webservice.py'
--- src/lazr/restful/testing/webservice.py 2009-08-20 04:18:09 +0000
+++ src/lazr/restful/testing/webservice.py 2009-08-25 14:04:09 +0000
@@ -211,8 +211,9 @@ 'preceding slash?)', path_or_url)
else:
- # We want the initial slash of the path, so we
subtract 1 from
- # the base_url's len.
+ # The base_url ends with a slash. We want that slash
to begin
+ # the path, so we cut off the length of the base_url
minus one
+ # character. full_url = path_or_url[len(base_url)-1:] full_headers = {'Host': self.domain}
if headers is not None:
On Aug 25, 2009, at 8:29 AM, Leonard Richardson wrote:
> Review: Needs Information
>> Fixes testing infrastructure and packaging problems.
>
> I have two comments.
>
> 1. Why the heck is 'roman' a dependency of lazr.restful? It's not in
> the code and it's such a random thing to need.
I was surprised too, but my tests failed in a clean Python (that is, a
vanilla Python with only the standard library) without it. That said,
I investigated further after you asked. It turns out that docutils
tries to do something tricky, and my version of docutils had broken
the trick. I rebuilt docutils and we're all good. Therefore, I
removed that indirect dependency from the list.
> 2. This comment isn't terribly helpful.
>
> 70 + # We want the initial slash of the path, so we subtract 1 from
> 71 + # the base_url's len.
>
> I'd prefer something like:
>
> # We want just one character of the base_url (the slash that begins
> the path), so we cut
> # off the length of the base_url minus one character.
Yes, good improvement. I took it a little further, as you can see in
the incremental diff below.
Thanks
Gary
=== modified file 'setup.py'
'lazr. enum',
'lazr. lifecycle' ,
'lazr. uri',
'setuptools' ,
'simplejson' ,
'van. testing' ,
--- setup.py 2009-08-20 04:18:09 +0000
+++ setup.py 2009-08-25 13:46:54 +0000
@@ -61,7 +61,6 @@
- 'roman',
=== modified file 'src/lazr/ restful/ testing/ webservice. py' restful/ testing/ webservice. py 2009-08-20 04:18:09 +0000 restful/ testing/ webservice. py 2009-08-25 14:04:09 +0000
'preceding slash?)',
path_ or_url)
full_ url = path_or_ url[len( base_url) -1:]
full_ headers = {'Host': self.domain}
--- src/lazr/
+++ src/lazr/
@@ -211,8 +211,9 @@
else:
- # We want the initial slash of the path, so we
subtract 1 from
- # the base_url's len.
+ # The base_url ends with a slash. We want that slash
to begin
+ # the path, so we cut off the length of the base_url
minus one
+ # character.
if headers is not None: