Re: [opensource-dev] Review Request: Do not fail when no scp command is found, unless it is actually needed to fetch something

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/127/#review274
---



autobuild/common.py


Why not use None here? I think that is more Pythonie.



autobuild/common.py


...and test for None here.


- Alain


On Jan. 28, 2011, 5:20 a.m., Oz Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/127/
> ---
> 
> (Updated Jan. 28, 2011, 5:20 a.m.)
> 
> 
> Review request for Viewer.
> 
> 
> Summary
> ---
> 
> During initialization, if there is no scp or pscp command found then 
> autobuild fails immediately.  This is true whether or not any scp urls need 
> to be used.
> 
> This change modifies the behavior so that a warning is printed if no command 
> is found, but execution proceeds until an scp command is needed, at which 
> time execution fails with an explanatory message.
> 
> This patch can print the warning multiple times - I didn't attempt to 
> suppress the extras.
> 
> 
> Diffs
> -
> 
>   autobuild/common.py f49808fe3c07 
> 
> Diff: http://codereview.secondlife.com/r/127/diff
> 
> 
> Testing
> ---
> 
> I've tested this locally, simulating the error by temporarily modifying the 
> names of the commands to be found for scp.
> 
> I have not checked it on Windows, where the original problem was found.
> 
> 
> Thanks,
> 
> Oz
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: OPEN-8: allow specification of the configuration with env var AUTOBUILD_CONFIGURATION

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/141/#review342
---


I'm not sure AUTOBUILD_CONFIGURATION is a good name for this environment 
variable.  I know this is long, but AUTOUBUILD_DEFAULT_BUILD_CONFIGURATIONS is 
more accurate and descriptive (and one could probably drop the AUTOBUILD if we 
aren't too worried about conflicts with other env variable names)

- Alain


On Feb. 6, 2011, 8:01 p.m., Oz Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/141/
> ---
> 
> (Updated Feb. 6, 2011, 8:01 p.m.)
> 
> 
> Review request for Viewer.
> 
> 
> Summary
> ---
> 
> This patch adds support for specifying the configurations using the 
> AUTOBUILD_CONFIGURATION environment variable (multiple configurations can be 
> specified using commas to separate the values).
> 
> 
> This addresses bug open-8.
> 
> 
> Diffs
> -
> 
>   autobuild/autobuild_base.py 9ee2db08d677 
>   autobuild/autobuild_tool_build.py 9ee2db08d677 
>   autobuild/autobuild_tool_configure.py 9ee2db08d677 
> 
> Diff: http://codereview.secondlife.com/r/141/diff
> 
> 
> Testing
> ---
> 
> Tested manually with and without the command line argument, and with and 
> without the environment variable set; the appropriate configuration values 
> were used.
> 
> 
> Thanks,
> 
> Oz
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: OPEN-7: allow specification of the config-file with env var AUTOBUILD_CONFIG_FILE

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/140/#review343
---


Should AUTOBUILD_CONFIG_FILE be renamed to AUTOBUILD_CONFIG_FILENAME to clarify 
this is a name and not a full file path?  Not sure what would happen if someone 
set AUTOBUILD_CONFIG_FILE=foo/bar/baz/MyAutobuildConfig.xml, but probably not 
what the user expects.


autobuild/configfile.py


This gets called once on package load.  I guess that's fine as long as you 
just use autobuild through the autobuild script, but you might get unexpected 
results if you are calling autobuild functions in your own interactive python 
session. Admittedly that's a corner case... 



autobuild/configfile.py


This isn't a warning.  It is perfectly OK to run autobuild with no script 
file, like, for example when you call autobuild edit to begin configuring an 
autobuild package.


- Alain


On Feb. 6, 2011, 5:39 p.m., Oz Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/140/
> ---
> 
> (Updated Feb. 6, 2011, 5:39 p.m.)
> 
> 
> Review request for Viewer.
> 
> 
> Summary
> ---
> 
> This change allows the environment variable AUTOBUILD_CONFIG_FILE to specify 
> a default config-file, so that the precedence becomes:
> 
>1. the --config-file command line option
>2. the environment variable AUTOBUILD_CONFIG_FILE
>3. "autobuild.xml"
> 
> It also adds an info-level (--verbose) display of the config file name that 
> is being loaded, and a warning level display if the config file does not 
> exist.
> 
> 
> This addresses bug open-7.
> 
> 
> Diffs
> -
> 
>   autobuild/autobuild_tool_build.py 9ee2db08d677 
>   autobuild/autobuild_tool_configure.py 9ee2db08d677 
>   autobuild/autobuild_tool_edit.py 9ee2db08d677 
>   autobuild/autobuild_tool_install.py 9ee2db08d677 
>   autobuild/autobuild_tool_installables.py 9ee2db08d677 
>   autobuild/autobuild_tool_manifest.py 9ee2db08d677 
>   autobuild/autobuild_tool_package.py 9ee2db08d677 
>   autobuild/autobuild_tool_print.py 9ee2db08d677 
>   autobuild/autobuild_tool_uninstall.py 9ee2db08d677 
>   autobuild/configfile.py 9ee2db08d677 
> 
> Diff: http://codereview.secondlife.com/r/140/diff
> 
> 
> Testing
> ---
> 
> Manually tested with and without the command line option, and with and 
> without the environment variable set: confirmed that the correct file name is 
> used per the precedence above.
> 
> Confirmed that the new logging works correctly.
> 
> 
> Thanks,
> 
> Oz
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: open-2 (improved progress messages during dependency checks) and open-31 (more standardized short form option switches)

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/150/#review359
---


"(which really has the dual meaning "something that might be a problem" and 
"something that should be seen at the default logging level")" .  As I mention 
in the comments, I think only the first kind of message should be shown at 
WARNING level because I don't like chat.  Others may disagree.



autobuild/autobuild_main.py


I'd would be a little cleaner to use:
  os.environ.get(AUTOBUILD_LOGLEVEL)
which returns None if the key does not exist in the map.  I prefer None to 
'' for an unset value myself. Or, you can use:
  os.environ.get(AUTOBUILD_LOGLEVEL, '')
if you prefer an empty string.




autobuild/autobuild_tool_install.py


I disagree with upping this to warning.  IMHO, the default mode for 
autobuild shouldn't be chatty.  If you want it to talk to you, set your warning 
level to INFO. I prefer that it just run quietly with no messages if everything 
is going fine.  If we have other messages that are WARNING level but not true 
warnings, I'd prefer they be made info instead.


- Alain


On Feb. 13, 2011, 5:21 a.m., Oz Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/150/
> ---
> 
> (Updated Feb. 13, 2011, 5:21 a.m.)
> 
> 
> Review request for Viewer.
> 
> 
> Summary
> ---
> 
> I combined open-31 into this because that change is so small and simple that 
> it didn't seem worth a separate commit (besides, while testing the rest of 
> this, I really wanted the short form of --verbose).
> 
> open-31 changes the short form of --version to -V, so that the short form of 
> the more commonly used --verbose can be the lower case -v, and adds -n as a 
> short form for --dry-run (as it is used in many versions of make, and in 
> other commands like scp and rsync).
> 
> open-2 corrects the problem that (at least in the viewer build) there is a 
> long delay during which there are no progress messages.  The delay is the 
> configure step while packages are checked, including possibly downloading, 
> extracting, and installing them.  Increasing the logging level with the 
> --verbose or --debug switches did not increase the amount of logging during 
> this phase.
> 
> The fundamental problem proved to be that the package checks were done in 
> recursively invoked instances of autobuild (I am guessing that the process 
> hierarchy is autobuild->cmake->autobuild, but it could have just been 
> autobuild calling itself directly; this fix should work regardless of the 
> depth of the process tree).  The verbosity control options were not passed 
> through to the child processes, and since there could be some other process 
> in between (as cmake above), it seemed more robust (and relies less on 
> correct configuration) to pass the verbosity control to the children through 
> the environment.
> 
> The default verbosity is now read from the AUTOBUILD_LOGGING environment 
> variable, whose contents can be --debug, --verbose, --quiet, or the one 
> letter forms of any of those options (if the environment variable is not set, 
> then the default value is to log warnings and worse, as before).  Any of the 
> options used on the command line override any value from the environment.  
> Regardless of how the verbosity level is established, the environment 
> variable AUTOBUILD_LOGGING is set for all child processes, so that if any of 
> those (directly or indirectly) are another invocation of autobuild, then that 
> recursive child will use the same verbosity level as the parent (unless, of 
> course the recursive invocation uses an explicit option).  It is true that 
> this also allows the user to set a default verbosity in the shell 
> environment, and that works, but it wasn't really the motivation for the 
> change and I did not add anything to the documentation about it.
> 
> With that change made, the options correctly controlled whether or not 
> logging is visible during the package checking phase.  However, the resulting 
> messages had an inconsistent level of detail - some operations (such as 
> uninstall) were very verbose, while others (some of which might take 
> significant time) were logged only at high verbosity levels.  This led to the 
> addition of a few short log messages at the default 'warning' level (which 
> really has the dual meaning "something that might be a problem" and 
> "something that should be seen at the default logging level") in order to 
> make sure that some message is printed before any potentially long operation 
> (downloads, extracts, 

Re: [opensource-dev] Review Request: OPEN-45: Correctly include the pass-through options when displaying the configure and build commands

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/213/#review476
---


I don't like leaking these details into the executable interface.  I'd prefer 
to do this another way, but I'll have to think about precisely what that way is.

- Alain


On March 16, 2011, 8:19 p.m., Oz Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/213/
> ---
> 
> (Updated March 16, 2011, 8:19 p.m.)
> 
> 
> Review request for Viewer and Alain Linden.
> 
> 
> Summary
> ---
> 
> Extended the __call__ method of Executable to pass in the command type as a 
> string and the dry_run flag so that the logging of the command being run 
> could use the infrastructure there to exactly assemble the options and 
> arguments for the logging of the command to be run.  The callers then don't 
> need to do the logging separately or do anything with dry_run other than pass 
> it down to the executable.
> 
> 
> This addresses bug open-45.
> http://jira.secondlife.com/browse/open-45
> 
> 
> Diffs
> -
> 
>   autobuild/autobuild_tool_build.py abc1014d5ad6 
>   autobuild/autobuild_tool_configure.py abc1014d5ad6 
>   autobuild/executable.py abc1014d5ad6 
> 
> Diff: http://codereview.secondlife.com/r/213/diff
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Oz
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: Suggested fix for OPEN-45

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/215/
---

(Updated March 17, 2011, 3:02 p.m.)


Review request for Viewer, Oz Linden and Boroondas Gupte.


Summary
---

Suggested fix for OPEN-45.  I'd like to decouple Executable as much as possible 
from assumptions and state of autobuild.  To that end I added an optional 
argument to the __str__ method to accept an options list to insert into the 
command string.  The command string generated by __call__ and __str__ is now 
identical.  Incidentally I made the environment an optional argument to 
__call__ to make it more general and decouple it a little more from the rest of 
autobuild.


This addresses bug OPEN-45.
http://jira.secondlife.com/browse/OPEN-45


Diffs
-

  autobuild/autobuild_tool_build.py abc1014d5ad6 
  autobuild/autobuild_tool_configure.py abc1014d5ad6 
  autobuild/common.py abc1014d5ad6 
  autobuild/executable.py abc1014d5ad6 

Diff: http://codereview.secondlife.com/r/215/diff


Testing
---


Thanks,

Alain

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: Enable watchdog timer for Beta release crash hunting.

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/337/#review743
---



indra/newview/app_settings/settings.xml
<http://codereview.secondlife.com/r/337/#comment707>

I presume this means trigger the watchdog after 20 seconds of 'freeze'.


- Alain


On June 14, 2011, 2:46 p.m., Jenn wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/337/
> ---
> 
> (Updated June 14, 2011, 2:46 p.m.)
> 
> 
> Review request for Viewer, Alain Linden, Richard Nelson, and Nat Goodspeed.
> 
> 
> Summary
> ---
> 
> Enable watchdog timer for Beta release to get tracebacks for classes of 
> crashes for which we currently have little data.
> 
> Setting timer to 20 seconds. Will either disable or make longer for 
> deployment to the Release channel.
> 
> 
> Diffs
> -
> 
>   indra/newview/app_settings/settings.xml a7c507c7e0f8 
> 
> Diff: http://codereview.secondlife.com/r/337/diff
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jenn
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: Enable watchdog timer for Beta crash hunting.

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/337/#review745
---

Ship it!


- Alain


On June 14, 2011, 2:50 p.m., Jenn wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/337/
> ---
> 
> (Updated June 14, 2011, 2:50 p.m.)
> 
> 
> Review request for Viewer, Alain Linden, Richard Nelson, and Nat Goodspeed.
> 
> 
> Summary
> ---
> 
> Enable watchdog timer for Beta release to get tracebacks for classes of 
> crashes for which we currently have little data.
> 
> Setting timer to 20 seconds. Will either disable or make longer for 
> deployment to the Release channel.
> 
> 
> Diffs
> -
> 
>   indra/newview/app_settings/settings.xml a7c507c7e0f8 
> 
> Diff: http://codereview.secondlife.com/r/337/diff
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jenn
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: Changes to fix CHOP-662.

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/357/#review795
---

Ship it!



indra/llvfs/tests/lldiriterator_test.cpp
<http://codereview.secondlife.com/r/357/#comment794>

Personally, I dislike referencing jira items in code comments.  Its 
referencing something ephemeral in something that will long out live it.


- Alain


On June 20, 2011, 8:21 p.m., Squire Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/357/
> ---
> 
> (Updated June 20, 2011, 8:21 p.m.)
> 
> 
> Review request for Viewer, Oz Linden, Seth ProductEngine, and Alain Linden.
> 
> 
> Summary
> ---
> 
> Changes to fix CHOP-662. CHOP-662 was caused by the recent addition of 
> lldiriterator.cpp which is used to iterate over directories. It takes a mask 
> in the form of a glob string to use as a pattern for which files to iterate 
> over.
> 
> Unfortunately, as originally implemented, it converted the glob to a regex 
> pattern without escaping any legal glob characters which are illegal regex 
> characters. As a result if the passed in mask was an illegal regex it would 
> fail and llerrs would cause a crash.
> 
> In CHOP-662 this happens whenever a group name has, e.g. a '+' character at 
> the beginning. When logging in the viewer would attempt to load the group 
> chat log file which would result in a glob starting with a '+' character and 
> in turn to an illegal regex in lldiriterator.
> 
> Also, the chat code was doing nothing to ensure that illegal glob characters 
> were not being passed to the lldiriterator constructor.
> 
> 
> This addresses bug CHOP-662.
> http://jira.secondlife.com/browse/CHOP-662
> 
> 
> Diffs
> -
> 
>   doc/contributions.txt e8f2a53c3d6e 
>   indra/llvfs/CMakeLists.txt e8f2a53c3d6e 
>   indra/llvfs/lldiriterator.cpp e8f2a53c3d6e 
>   indra/llvfs/tests/lldiriterator_test.cpp PRE-CREATION 
>   indra/newview/lllogchat.cpp e8f2a53c3d6e 
> 
> Diff: http://codereview.secondlife.com/r/357/diff
> 
> 
> Testing
> ---
> 
> Added a unit test for lldiriterator which checks for construction failures on 
> examples of the most common group names which were causing the crashes.
> 
> 
> Thanks,
> 
> Squire
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: Changes to fix CHOP-662.

[Alain Linden]

> On June 21, 2011, 7:57 a.m., Alain Linden wrote:
> > indra/llvfs/tests/lldiriterator_test.cpp, line 46
> > <http://codereview.secondlife.com/r/357/diff/1/?file=3024#file3024line46>
> >
> > Personally, I dislike referencing jira items in code comments.  Its 
> > referencing something ephemeral in something that will long out live it.
> 
> Boroondas Gupte wrote:
> In general I'd agree, but here my interpretation was that this test was 
> inspired by a specific issue. Should in that case the whole issue description 
> be copied into the comment?

The way I see it, the code remains long after the jira is dead and forgotten.  
As much as possible, code commentary should stand alone.  This test makes sense 
regardless of what jira inspired it, so why bother mentioning it?  Years from 
now a developer reading this code won't care about the jira.  That's my opinion.


- Alain


---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/357/#review795
---


On June 20, 2011, 8:21 p.m., Squire Linden wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://codereview.secondlife.com/r/357/
> ---
> 
> (Updated June 20, 2011, 8:21 p.m.)
> 
> 
> Review request for Viewer, Oz Linden, Seth ProductEngine, and Alain Linden.
> 
> 
> Summary
> ---
> 
> Changes to fix CHOP-662. CHOP-662 was caused by the recent addition of 
> lldiriterator.cpp which is used to iterate over directories. It takes a mask 
> in the form of a glob string to use as a pattern for which files to iterate 
> over.
> 
> Unfortunately, as originally implemented, it converted the glob to a regex 
> pattern without escaping any legal glob characters which are illegal regex 
> characters. As a result if the passed in mask was an illegal regex it would 
> fail and llerrs would cause a crash.
> 
> In CHOP-662 this happens whenever a group name has, e.g. a '+' character at 
> the beginning. When logging in the viewer would attempt to load the group 
> chat log file which would result in a glob starting with a '+' character and 
> in turn to an illegal regex in lldiriterator.
> 
> Also, the chat code was doing nothing to ensure that illegal glob characters 
> were not being passed to the lldiriterator constructor.
> 
> 
> This addresses bug CHOP-662.
> http://jira.secondlife.com/browse/CHOP-662
> 
> 
> Diffs
> -
> 
>   doc/contributions.txt e8f2a53c3d6e 
>   indra/llvfs/CMakeLists.txt e8f2a53c3d6e 
>   indra/llvfs/lldiriterator.cpp e8f2a53c3d6e 
>   indra/llvfs/tests/lldiriterator_test.cpp PRE-CREATION 
>   indra/newview/lllogchat.cpp e8f2a53c3d6e 
> 
> Diff: http://codereview.secondlife.com/r/357/diff
> 
> 
> Testing
> ---
> 
> Added a unit test for lldiriterator which checks for construction failures on 
> examples of the most common group names which were causing the crashes.
> 
> 
> Thanks,
> 
> Squire
> 
>

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

[opensource-dev] Review Request: Check for null ptr to hopefully prevent crash in LLToolPie

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/359/
---

Review request for Viewer, Oz Linden and Vadim ProductEngine.


Summary
---

checking for null gAgentAvatarp should hopefully prevent the crash.  I can't 
reproduce it so I don't know for sure.


This addresses bug EXP-825.
http://jira.secondlife.com/browse/EXP-825


Diffs
-

  indra/newview/lltoolpie.cpp e8f2a53c3d6e 

Diff: http://codereview.secondlife.com/r/359/diff


Testing
---


Thanks,

Alain

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

[opensource-dev] Review Request: Adds support for packaging and extracting zip formatted archives

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/494/
---

Review request for Viewer, Oz Linden, Jenn, and Nat Goodspeed.


Summary
---

These changes are to support packaging archives as zip files.  This is useful 
when you are using autobuild to construct an archive that isn't really intended 
to be used as an autobuild installable.  Internally autobuild can handle 
tarballs on any platform, but if you are making a package for windows and not 
installing with autobuild, it is better to use zip which is natively supported. 
 Changes here do the following:

1. enable the --archive-format option for autobuild package to choose the 
format of the package.
2. add support for reading an 'archive' entry per platform with the new 
'format' element to select the archive format with the configuration file.
3. add new edit command 'autobuild edit archive' to add archive configurations.
4. add support for installing zip formatted archives.


Diffs
-

  autobuild/autobuild_tool_edit.py ac90b03614ea 
  autobuild/autobuild_tool_package.py ac90b03614ea 
  autobuild/common.py ac90b03614ea 
  autobuild/configfile.py ac90b03614ea 
  autobuild/tests/test_common.py ac90b03614ea 
  autobuild/tests/test_package.py ac90b03614ea 

Diff: http://codereview.secondlife.com/r/494/diff


Testing
---


Thanks,

Alain

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges

Re: [opensource-dev] Review Request: Adds support for packaging and extracting zip formatted archives

[Alain Linden]

---
This is an automatically generated e-mail. To reply, visit:
http://codereview.secondlife.com/r/494/
---

(Updated Oct. 14, 2011, 9:32 a.m.)


Review request for Viewer, Oz Linden, Jenn, and Nat Goodspeed.


Changes
---

Included a few bug fixes.


Summary
---

These changes are to support packaging archives as zip files.  This is useful 
when you are using autobuild to construct an archive that isn't really intended 
to be used as an autobuild installable.  Internally autobuild can handle 
tarballs on any platform, but if you are making a package for windows and not 
installing with autobuild, it is better to use zip which is natively supported. 
 Changes here do the following:

1. enable the --archive-format option for autobuild package to choose the 
format of the package.
2. add support for reading an 'archive' entry per platform with the new 
'format' element to select the archive format with the configuration file.
3. add new edit command 'autobuild edit archive' to add archive configurations.
4. add support for installing zip formatted archives.


Diffs (updated)
-

  autobuild/autobuild_tool_edit.py ac90b03614ea 
  autobuild/autobuild_tool_package.py ac90b03614ea 
  autobuild/common.py ac90b03614ea 
  autobuild/configfile.py ac90b03614ea 
  autobuild/tests/test_common.py ac90b03614ea 
  autobuild/tests/test_package.py ac90b03614ea 

Diff: http://codereview.secondlife.com/r/494/diff


Testing
---


Thanks,

Alain

___
Policies and (un)subscribe information available here:
http://wiki.secondlife.com/wiki/OpenSource-Dev
Please read the policies before posting to keep unmoderated posting privileges