]> git.ipfire.org Git - thirdparty/openembedded/openembedded-core-contrib.git/log
thirdparty/openembedded/openembedded-core-contrib.git
12 years agoAdded distroy signal to information dialogs
Andrei Dinu [Wed, 29 May 2013 08:42:22 +0000 (11:42 +0300)] 
Added distroy signal to information dialogs

Added distroy signal to information dialogs
so that when pressing the close button,
the dialog would close.

[HOB #4568]

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agolib/bb/cache.py: Change debugging note to a debug message
Mark Hatle [Wed, 29 May 2013 04:01:18 +0000 (23:01 -0500)] 
lib/bb/cache.py: Change debugging note to a debug message

Some apparently debugging was left in in a previous commit.  This caused
bitbake to return a list of bbappends when things changed from the cache.

Make this a proper debug message.

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoprserv: Unbreak after bb.server changes
Richard Purdie [Thu, 30 May 2013 17:05:14 +0000 (17:05 +0000)] 
prserv: Unbreak after bb.server changes

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocookerdata: Fix tinfoil by adding missinge extra_caches default
Richard Purdie [Thu, 30 May 2013 12:37:12 +0000 (12:37 +0000)] 
cookerdata: Fix tinfoil by adding missinge extra_caches default

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: xmlrpc remote server
Alexandru DAMIAN [Tue, 28 May 2013 16:52:02 +0000 (16:52 +0000)] 
bitbake: xmlrpc remote server

Added code in XMLRPC server that creates a stub local server
for a client-only connection and is able to connect to
a remote server, and receive events from the remote server.

Added the option to start a client with a remote server in
bitbake.

Original code by Bogdan Marinescu <bogdan.a.marinescu@intel.com>

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: xmlrpc transport has identification token
Alexandru DAMIAN [Tue, 28 May 2013 13:55:09 +0000 (13:55 +0000)] 
bitbake: xmlrpc transport has identification token

In order to be able to identify different clients over a
stateless XMLRPC connection, we add a custom header named
Bitbake-token, which identifies each client.

We refactor the rest of the code to use the new transport.

Based on a patch by Bogdan Marinescu <bogdan.a.marinescu@intel.com>

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: move start server code in a separate function
Alexandru DAMIAN [Tue, 28 May 2013 13:32:48 +0000 (13:32 +0000)] 
bitbake: move start server code in a separate function

This is a code sanitization targeted at making further
server-related changes easier (launch a server
separately or creating a mockup-server) to do.

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: move extra cache collection out of cooker
Alexandru DAMIAN [Tue, 28 May 2013 12:49:55 +0000 (12:49 +0000)] 
bitbake: move extra cache collection out of cooker

The collection of the extra caching data should not
be performed by the cooker, but supplied to it.
This patch will also streamline the code for launching servers
without a UI attached.

Based on a patch by Bogdan Marinescu <bogdan.a.marinescu@intel.com>

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: retrieve the entire ui module in launcher
Alexandru DAMIAN [Tue, 28 May 2013 12:31:45 +0000 (12:31 +0000)] 
bitbake: retrieve the entire ui module in launcher

Slight change in bitbake as to retrieve the entire UI module
for further processing instead of just the main function.

Based on a patch by Bogdan Marinescu <bogdan.a.marinescu@intel.com>

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2: Fix missing function call
Richard Purdie [Thu, 30 May 2013 08:50:41 +0000 (09:50 +0100)] 
fetch2: Fix missing function call

Fix the issue:

File: '/srv/home/pokybuild/yocto-autobuilder-new/yocto-slave/build-appliance/build/bitbake/lib/bb/fetch2/__init__.py', lineno: 813, function: try_mirror_url
     0809:    except bb.fetch2.BBFetchException as e:
     0810:        if isinstance(e, ChecksumError):
     0811:            logger.warn("Mirror checksum failure for url %s (original url: %s)\nCleaning and trying again." % (newuri, origud.url))
     0812:            logger.warn(str(e))
 *** 0813:            self.rename_bad_checksum(ud, e.checksum)
     0814:        elif isinstance(e, NoChecksumError):
     0815:            raise
     0816:        else:
     0817:            logger.debug(1, "Mirror fetch failure for url %s (original url: %s)" % (newuri, origud.url))
Exception: NameError: global name 'self' is not defined

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoserver/bitbake: Remove launchUI method
Richard Purdie [Tue, 28 May 2013 13:32:53 +0000 (14:32 +0100)] 
server/bitbake: Remove launchUI method

With the removal of the none server type, we can remove the launchUI
method and simplify the code slightly.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoserver: Remove none server
Richard Purdie [Tue, 28 May 2013 13:28:30 +0000 (14:28 +0100)] 
server: Remove none server

The process server backend has been serving well as the default for a long
time now and the UI model is much better thought out that it used to be. With
the move to make bitbake a memory resident process, the none server is now
looking rather pointless and complicates the code needlessly. Lets therefore
now remove it.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker/cookerdata: Fix up event handling for hob
Richard Purdie [Fri, 24 May 2013 12:54:59 +0000 (12:54 +0000)] 
cooker/cookerdata: Fix up event handling for hob

The init function call was clobbering data hob relies on injecting, fix
this so the data injected by hob isn't overwritten.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobuild: Avoid empty promises
Richard Purdie [Fri, 24 May 2013 10:28:38 +0000 (11:28 +0100)] 
build: Avoid empty promises

The "see xxx for further information" is misleading since it is just
the same information. Clarify just to mention the that this is the location
of the logfile without any empty promise.

[YOCTO #4343]

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake-layers: fix show-appends and flatten for recent cooker changes
Paul Eggleton [Fri, 24 May 2013 10:12:50 +0000 (11:12 +0100)] 
bitbake-layers: fix show-appends and flatten for recent cooker changes

Fixes the show-appends and flatten subcommands for recent refactoring in
cooker.

[YOCTO #4536]
[YOCTO #4535]

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker/cookerdata: Improve configuration object handling
Richard Purdie [Fri, 24 May 2013 09:27:25 +0000 (09:27 +0000)] 
cooker/cookerdata: Improve configuration object handling

Originally it seemed like a good idea to keep the parameters around. Having
seen this in real life use, its incorrect, we should pull all the data we need
into the cooker's configuguration and then use this to build the datastore.

Being able to just build the datastore from the parameters seemed like a good
idea but having a dummy cooker configuration object is now looking like
the better option.

This also fixes failures in hob since the parseFiles command can call
into cooker directly now and reset the configuration prefiles and postfiles
at will, rather than the indirect calls before which were breaking the datastore
(e.g. BBPATH wasn't set).

The cleanup this allows in tinfoil illustrates how this change makes more sense.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2: Fix missing assignment
Richard Purdie [Thu, 23 May 2013 10:51:04 +0000 (11:51 +0100)] 
fetch2: Fix missing assignment

The assignment to True was missing from the code, well spotted Saul!

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker/cookerdata/event: Improve class handlers management
Richard Purdie [Thu, 23 May 2013 09:50:30 +0000 (10:50 +0100)] 
cooker/cookerdata/event: Improve class handlers management

Similarly to the execution context changes, establish better lifetime
management API of the class event handlers.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker/cookerdata/utils: Improve context management
Richard Purdie [Thu, 23 May 2013 09:49:57 +0000 (10:49 +0100)] 
cooker/cookerdata/utils: Improve context management

The current execution context management for bitbake is ugly and the
use of a global variable is nasty. Fixing that is hard, however we
can improve things to start to establish an API for accessing
and changing that context.

This patch also adds in an explicit reset of the context when we reparse
the configuration data which starts to improve the lifecycle of the data
in setups like hob.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoevent: Drop Handled/NotHandled status return values
Richard Purdie [Thu, 23 May 2013 09:49:06 +0000 (10:49 +0100)] 
event: Drop Handled/NotHandled status return values

These have long since been deprecated, lets remove them.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker.py: Remove global scope class event handlers
Richard Purdie [Thu, 23 May 2013 09:48:26 +0000 (10:48 +0100)] 
cooker.py: Remove global scope class event handlers

Its confusing when an event handler in for example a bbclass receives
events for a recipe which doesn't use the class. This is due to the
class event handlers having confusing scope. Worse, the issue is not
deterministic and the events a handler will see depends on the parse
order.

To avoid these issues, remove the class event handler global scope
and only have class handlers recieve events for recipes using the
class.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agomethodpool: Retire it, remove global method scope
Richard Purdie [Thu, 23 May 2013 09:47:10 +0000 (10:47 +0100)] 
methodpool: Retire it, remove global method scope

Having a global method scope confuses users and with the introduction
of parallel parsing, its not even possible to correctly detect conflicting
functions. Rather than try and fix that, its simpler to retire the global
method scope and restrict functions to those locations they're defined
within. This is more what users actually expect too.

If we remove the global function scope, the need for methodpool is reduced
to the point we may as well retire it. There is some small loss of caching
of parsed functions but timing measurements so the impact to be neglibile
in the overall parsing time.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agomethodpool: Conflicting methodnames should be a fatal error
Richard Purdie [Thu, 23 May 2013 09:47:45 +0000 (10:47 +0100)] 
methodpool: Conflicting methodnames should be a fatal error

When this error occurs, the build should stop, not continue uninterrupted.

[YOCTO #4460]

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: Table sorting criteria should not change after deselecting recipes/packages
Cristiana Voicu [Wed, 22 May 2013 09:52:35 +0000 (12:52 +0300)] 
hob: Table sorting criteria should not change after deselecting recipes/packages

After a recipe/package is deselected, the table is sorted as before the
fadeout animation.

[YOCTO #4453]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: apply a fade out animation when deselecting from tables
Cristiana Voicu [Wed, 22 May 2013 09:52:34 +0000 (12:52 +0300)] 
hob: apply a fade out animation when deselecting from tables

-applied a fade out mechanism for deselctiong on all the tabs of
recipes and packages list page
-set cursor on the package/recipe when it is included(important
when the its place is changing; like when the tree si order by
included tab)

[YOCTO #4342]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocookerdata: rename _parse to parse_config_file
Paul Eggleton [Wed, 22 May 2013 14:25:13 +0000 (15:25 +0100)] 
cookerdata: rename _parse to parse_config_file

We use this externally in the OE layer index update script, so it
shouldn't really be named as an internal function.

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotinfoil: fix for changes to cooker config structure
Paul Eggleton [Wed, 22 May 2013 14:25:12 +0000 (15:25 +0100)] 
tinfoil: fix for changes to cooker config structure

Fix the code here for recent changes to the initialisation of
configuration objects for cooker.

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake-layers: fix for move of calc_bbfile_priority within cooker
Paul Eggleton [Wed, 22 May 2013 14:25:11 +0000 (15:25 +0100)] 
bitbake-layers: fix for move of calc_bbfile_priority within cooker

calc_bbfile_priority is now in CookerCollectFiles which can be accessed
on the collection attribute of a cooker instance.

Fixes [YOCTO #4513].

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocommand: Fix function module reference after code rearrangement
Richard Purdie [Wed, 22 May 2013 14:55:14 +0000 (14:55 +0000)] 
command: Fix function module reference after code rearrangement

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agorunqueue.py: Ensure export flag is set for fakeroot environment variables
Richard Purdie [Mon, 18 Feb 2013 13:18:31 +0000 (13:18 +0000)] 
runqueue.py: Ensure export flag is set for fakeroot environment variables

This means the variables show up in the shell execution "run" files since
its useful to know what the fakeroot environment is and how to set it up
manually.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake/cookerdata: Explicitly specify cooker configuration options
Richard Purdie [Mon, 20 May 2013 22:01:48 +0000 (23:01 +0100)] 
bitbake/cookerdata: Explicitly specify cooker configuration options

As the code stands today its hard to know which configuration variables
are used by which parts of the system. Some are used by the UIs, some
by bin/bitbake itself, some by cooker.

This patch changes the configuration to just contain the variables cooker
uses, and changes bin/bitbake to access the variables it needs directly
which hopefully lets us start to untangle this mess.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Split configuration parsing code into cookerdata
Richard Purdie [Mon, 20 May 2013 22:00:31 +0000 (23:00 +0100)] 
cooker: Split configuration parsing code into cookerdata

In order to have a memory resident bitbake and to allow task execution, we need
to be able to rebuild the base configuration without a cooker. This moves the
code into its own class so it can be built independently.

The interface is less than ideal here but I didn't want to add parsing methods
a subclassed DataSmart, at least until we've experimented further with this code
and are certain that makes sense. At the very least, the methods are ugly and need
cleaning up.  Spliting the code out seems to be the right thing to do though and
should unblock various activities on BitBake so I believe this code is a step in
the right direction.

Based on a patch from Alexandru Damian <alexandru.damian@intel.com>

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Move commandline parsing back into the UI/cookerdata
Richard Purdie [Mon, 20 May 2013 21:54:41 +0000 (22:54 +0100)] 
cooker: Move commandline parsing back into the UI/cookerdata

Building up a set of actions for the server is tricky since we depend upon the
commandline but fall back to values from the datastore. We should be able to build
a datastore without a commandline and vice versa. Ultimately the UI should send
the commands to the server.

This patch amounts to code rearranging, moving the heavy lifting to the UI, though
a helper in the configuration option. This will need further cleanup/tweaking but
this should be the only update needed to the UIs. The code now queries the server
for any missing data should it need to.

This code allows various knowledge of configuration variables to move to the UI side
only, partcularly pkgs_to_build but also all the command specifiers. It should also
be possible to move cmd eventually, I'm just unsure if any callers call the commands
expecting this to default to something sane right now.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: Create cookerdata splitting config from cooker and bin/bitbake
Richard Purdie [Mon, 20 May 2013 21:54:30 +0000 (22:54 +0100)] 
bitbake: Create cookerdata splitting config from cooker and bin/bitbake

Currently the UI and server configuration is one big incestuous mess. To
start to untangle this we creater cookerdata, a new module which contains
various confiuration modules and the code for building the base datastore.

To start with we add a ConfigParameters() class which contains information
about both the commandline configuration and the original environment.

The CookerConfiguration class is created to contain the cooker.configuration
options. This means we can transfer new paramters to the server over something
like XMLRPC and then build a new configuration from these on the server.

Based on a patch from Alexandru Damian <alexandru.damian@intel.com>

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobin/bitbake: Improve exception handling
Richard Purdie [Mon, 20 May 2013 22:00:10 +0000 (23:00 +0100)] 
bin/bitbake: Improve exception handling

Due to the internal event processing, this excepting handler usually raises an
Empty error, masking the underlying failure. Ensure the original exception is
raised.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2: Shorten long srcrevs
Richard Purdie [Sun, 19 May 2013 10:21:55 +0000 (13:21 +0300)] 
fetch2: Shorten long srcrevs

The long srcrevs are mainly used or the workdir construction as well as
the package version. The long entries are hashes generated by the git fetcher
and other scms using a similar revision mechanism.

We need these to change when the package changes however collisions are
unlikely to happen within the domains we care about. The long revisions
have generated negative user feedback due to the use in path and file
names.

This patch therefore truncates the revisions to 10 characters maximum.

This should be safe in the contexts where these revisions are used as
the chances of spatially close collisions is very low (distant
collisions are not a major issue in the way we use these).

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2/git: Clean up sortable_revision
Richard Purdie [Sun, 19 May 2013 10:17:58 +0000 (13:17 +0300)] 
fetch2/git: Clean up sortable_revision

Now we no longer try and provide increasing values from the fetcher,
we can simplify the function structure for the sortable_revision
pieces and move the AUTOINC handling directly into the function
which needs it, simplifying the code.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agodata: Ensure dependencies of subfunctions are accounted for
Richard Purdie [Sun, 19 May 2013 10:16:54 +0000 (13:16 +0300)] 
data: Ensure dependencies of subfunctions are accounted for

Currently we account of the top level function's vardeps but not
those of any subfunction. This would imply we'd have to manually
write the dependencies of all parent functions which would be crazy.

This patch adds the dependencies to fix the issue.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotest/fetch: Allow the conditional network tests to work under python 2.6
Richard Purdie [Fri, 17 May 2013 09:41:04 +0000 (12:41 +0300)] 
test/fetch: Allow the conditional network tests to work under python 2.6

The skipIf decorator is unavailable under python 2.6 and finding a syntax
works there is hard. This patch does allow network tests under 2.6.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotest/fetch: Split the network tests from the rest of the fetcher tests
Richard Purdie [Fri, 17 May 2013 08:46:47 +0000 (11:46 +0300)] 
test/fetch: Split the network tests from the rest of the fetcher tests

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2/utils: Clean up imports
Richard Purdie [Fri, 17 May 2013 08:40:01 +0000 (11:40 +0300)] 
fetch2/utils: Clean up imports

Move various random imports to the start of the modules as cleanup
and avoid an import issue with bb.process on python 2.6.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agotests/fetch: Allow to work on python 2.6
Richard Purdie [Fri, 17 May 2013 08:38:57 +0000 (11:38 +0300)] 
tests/fetch: Allow to work on python 2.6

Currently the fetch tests fail on python 2.6 since check_output
doesn't exist on this version of python. Use bb.process instead
to avoid this problem.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobin/bitbake: Fix queue import for pyhton 2 and 3
Richard Purdie [Tue, 14 May 2013 12:04:14 +0000 (15:04 +0300)] 
bin/bitbake: Fix queue import for pyhton 2 and 3

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: fix for XMLRPC mode
Bogdan Marinescu [Wed, 16 Jan 2013 12:36:55 +0000 (12:36 +0000)] 
hob: fix for XMLRPC mode

Contains a small fix to the builder.py code
(a missing "self" in a variable assignment).
With this changes, "hob" can run with the XMLRPC backend:

$ hob -t xmlrpc

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker/bitbake-layers/tinfoil: Fix recipecache typo
Richard Purdie [Mon, 13 May 2013 18:53:26 +0000 (21:53 +0300)] 
cooker/bitbake-layers/tinfoil: Fix recipecache typo

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake-layers/tinfoil: Catch up with status -> recpiecache rename
Richard Purdie [Sun, 12 May 2013 17:00:39 +0000 (18:00 +0100)] 
bitbake-layers/tinfoil: Catch up with status -> recpiecache rename

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Rename status variable to recipecache
Richard Purdie [Fri, 10 May 2013 17:35:47 +0000 (17:35 +0000)] 
cooker: Rename status variable to recipecache

"status" is a relic of very early versions of bitbake, rename it to match what it now
means.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Rename confusing buildDepgraph task to handlePrefProviders
Richard Purdie [Fri, 10 May 2013 15:26:03 +0000 (15:26 +0000)] 
cooker: Rename confusing buildDepgraph task to handlePrefProviders

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Drop pointless variables
Richard Purdie [Fri, 10 May 2013 15:24:01 +0000 (15:24 +0000)] 
cooker: Drop pointless variables

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker.py: Drop confusing updateCache calls
Richard Purdie [Fri, 10 May 2013 15:09:36 +0000 (15:09 +0000)] 
cooker.py: Drop confusing updateCache calls

updateCache() gets called by command.py when needed and needs to be iterated
over. The calls in cooker.py are therefore just plain wrong/confusing now.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Remove pointless repeated import
Richard Purdie [Fri, 10 May 2013 14:54:46 +0000 (14:54 +0000)] 
cooker: Remove pointless repeated import

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Fix get_bbfiles()
Richard Purdie [Fri, 10 May 2013 14:52:47 +0000 (14:52 +0000)] 
cooker: Fix get_bbfiles()

Its clear looking at the code the path argument has never been used and doesn't
work. Tweak the function to be a little more readable.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Separate out collections handling code into its own class
Richard Purdie [Fri, 10 May 2013 14:35:39 +0000 (14:35 +0000)] 
cooker: Separate out collections handling code into its own class

The Cooker class is too large and needs to be split up into different
functional units. Splitting out the collections code into its own class
seems like a good place to start to try and disentangle things.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Drop unused variables from collect_bbfiles()
Richard Purdie [Fri, 10 May 2013 15:16:26 +0000 (16:16 +0100)] 
cooker: Drop unused variables from collect_bbfiles()

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agopysh: Say what kind of token isn't implemented
Olof Johansson [Tue, 7 May 2013 09:22:37 +0000 (11:22 +0200)] 
pysh: Say what kind of token isn't implemented

When the shell lexer finds an unrecognized dollar token, the error
message should contain what kind of token it is having problems with.

Signed-off-by: Olof Johansson <olof.johansson@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: assure consistency for image size label
Cristiana Voicu [Tue, 7 May 2013 08:07:56 +0000 (11:07 +0300)] 
hob: assure consistency for image size label

After the image was build, update the image size in the packagelistpage.

[YOCTO #4388]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: clicking on "cancel" should reset table sorting to default
Cristiana Voicu [Tue, 7 May 2013 08:07:55 +0000 (11:07 +0300)] 
hob: clicking on "cancel" should reset table sorting to default

Changed the "cancel" callback function for both recipes and packages
list page.

[YOCTO #4333]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: "Brought in by" sorting should take into account the (+#) information
Cristiana Voicu [Tue, 7 May 2013 08:07:54 +0000 (11:07 +0300)] 
hob: "Brought in by" sorting should take into account the (+#) information

[YOCTO #4332]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: sort alphabetically results in groups of recipes/packages
Cristiana Voicu [Tue, 7 May 2013 08:07:53 +0000 (11:07 +0300)] 
hob: sort alphabetically results in groups of recipes/packages

When you search a string, there are 2 groups: recipes/packages starting
with that string, and the rest. These 2 groups are now sorted.

[YOCTO #4111]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoUpdated information in tooltips
Andrei Dinu [Tue, 7 May 2013 08:06:34 +0000 (11:06 +0300)] 
Updated information in tooltips

Updated information for "Additional free space" and
"Image basic size" to match the description in the
manual.

[HOB #4354]
[HOB #4355]

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agolib: print is a function in moderm python, covert remaining users
Richard Purdie [Thu, 9 May 2013 21:19:04 +0000 (21:19 +0000)] 
lib: print is a function in moderm python, covert remaining users

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agolib: Use modern expcetion syntax
Richard Purdie [Thu, 9 May 2013 21:18:20 +0000 (21:18 +0000)] 
lib: Use modern expcetion syntax

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoprserv: Drop StandardError usage
Richard Purdie [Thu, 9 May 2013 21:17:18 +0000 (21:17 +0000)] 
prserv: Drop StandardError usage

StandardError doesn't exist in python 3, use Exception instead.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agosiggen: Add missing error message parameter
Richard Purdie [Thu, 9 May 2013 21:16:09 +0000 (21:16 +0000)] 
siggen: Add missing error message parameter

There is clearly a missing parameter from this error message, add it.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoproviders: Remove pointless lambda sort function
Richard Purdie [Thu, 9 May 2013 21:14:50 +0000 (21:14 +0000)] 
providers: Remove pointless lambda sort function

This lambda function is equivalent to the default sort used by sorted,
so we can simply remove this. The syntax isn't compatible with python 3.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker: Remove deprecated operator
Richard Purdie [Thu, 9 May 2013 21:14:06 +0000 (21:14 +0000)] 
cooker: Remove deprecated operator

The <> operator is deprecated, replace with !=.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agolib: Clean up various file access syntax
Richard Purdie [Thu, 9 May 2013 21:06:45 +0000 (21:06 +0000)] 
lib: Clean up various file access syntax

Python 3 is stricter about how files are accessed. Specficially:

 * Use open(), not file()
 * Use binary mode for binary files (when checksumming)
 * Use with statements to ensure files get closed
 * Add missing file close statements

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agogit: remove symling before updating mirror tarball
Martin Jansa [Tue, 7 May 2013 16:39:20 +0000 (18:39 +0200)] 
git: remove symling before updating mirror tarball

* with read-only PREMIRROR (e.g. mounted over NFS or CIFS
  and referenced as file:///mnt/premirror) we cannot use
  BB_GENERATE_MIRROR_TARBALLS because all git2_abc.git.tar.gz
  files later became just symlinks to read-only location in PREMIRROR
  (it works fine on first build and for new components, because
  at that time there isn't tarball on PREMIRROR yet).

  ERROR: Fetcher failure: Fetch command failed with exit code 141, output:
  tar (child): /build/downloads/git2_abc.git.tar.gz: Cannot open: Read-only file system
  tar (child): Error is not recoverable: exiting now

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobin/bitbake: add a check for /dev/shm existing
Paul Eggleton [Thu, 9 May 2013 10:22:26 +0000 (11:22 +0100)] 
bin/bitbake: add a check for /dev/shm existing

Python multiprocessing needs /dev/shm, and if it doesn't exist (e.g. in
a chroot environment) or is not writable, you'll get
"OSError: [Errno 38] Function not implemented" or
"OSError: [Errno 13] Permission denied" with a traceback, which doesn't
really help to explain what is wrong.

Implements [YOCTO #4407].

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoutils: Improve better_compile error message
Richard Purdie [Wed, 8 May 2013 17:16:00 +0000 (18:16 +0100)] 
utils: Improve better_compile error message

Similarly to the better_exec improvements, improve the compile failure
messages to be more user readable.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoutils: Improve better_exec traceback handling
Richard Purdie [Wed, 8 May 2013 17:14:53 +0000 (18:14 +0100)] 
utils: Improve better_exec traceback handling

The current bitbake tracebacks are hard to read/confusing and sometimes
incomplete. This patch attempts to do better by:

* Moving the note about the exact exception to the end to make things
  read in sequence
* Merged the initial stack trace to become part of the code dump
* Added handling for "/xxxx" file paths since we can load these files
  and include the data as part of the trace
* Dropped the ERROR: prefix to every line, allowing the error messages to
  be spacially accosicated in the UIs
* Moved the "From:" line to the top of each code block and ensured its present
  consistently

With the complexity now in this funciton, I've added try/except wrapping around
it to ensure we catch exceptions in the exception handler too.

Example before:

"""
ERROR: Error executing a python function in /media/build1/poky/meta/recipes-core/eglibc/eglibc-initial_2.17.bb:
TypeError: 'filter' object is not subscriptable

ERROR: The stack trace of python calls that resulted in this exception/failure was:
ERROR:   File "do_populate_lic", line 13, in <module>
ERROR:
ERROR:   File "do_populate_lic", line 6, in do_populate_lic
ERROR:
ERROR:   File "license.bbclass", line 99, in find_license_files
ERROR:
ERROR:   File "/media/build1/poky/meta/lib/oe/license.py", line 38, in visit_string
ERROR:     if pos > 0 and license_pattern.match(elements[pos-1]):
ERROR:
ERROR: The code that was being executed was:
ERROR:      0009:    destdir = os.path.join(d.getVar('LICSSTATEDIR', True), d.getVar('PN', True))
ERROR:      0010:    copy_license_files(lic_files_paths, destdir)
ERROR:      0011:
ERROR:      0012:
ERROR:  *** 0013:do_populate_lic(d)
ERROR:      0014:
ERROR: [From file: 'do_populate_lic', lineno: 13, function: <module>]
ERROR:      0002:def do_populate_lic(d):
ERROR:      0003:    """
ERROR:      0004:    Populate LICENSE_DIRECTORY with licenses.
ERROR:      0005:    """
ERROR:  *** 0006:    lic_files_paths = find_license_files(d)
ERROR:      0007:
ERROR:      0008:    # The base directory we wrangle licenses to
ERROR:      0009:    destdir = os.path.join(d.getVar('LICSSTATEDIR', True), d.getVar('PN', True))
ERROR:      0010:    copy_license_files(lic_files_paths, destdir)
ERROR: [From file: 'do_populate_lic', lineno: 6, function: do_populate_lic]
ERROR:      0095:        lic_files_paths.append((os.path.basename(path), srclicfile))
ERROR:      0096:
ERROR:      0097:    v = FindVisitor()
ERROR:      0098:    try:
ERROR:  *** 0099:        v.visit_string(license_types)
ERROR:      0100:    except oe.license.InvalidLicense as exc:
ERROR:      0101:        bb.fatal('%s: %s' % (d.getVar('PF', True), exc))
ERROR:      0102:    except SyntaxError:
ERROR:      0103:        bb.warn("%s: Failed to parse it's LICENSE field." % (d.getVar('PF', True)))
ERROR: [From file: 'license.bbclass', lineno: 99, function: find_license_files]
ERROR: Function failed: do_populate_lic
ERROR: Logfile of failure stored in: /media/build1/poky/build/tmp/work/i586-poky-linux/eglibc-initial/2.17-r3/temp/log.do_populate_lic.17442
"""

Example after:

"""
ERROR: Error executing a python function in /media/build1/poky/meta/recipes-core/eglibc/eglibc-initial_2.17.bb:

The stack trace of python calls that resulted in this exception/failure was:
File: 'do_populate_lic', lineno: 13, function: <module>
     0009:    destdir = os.path.join(d.getVar('LICSSTATEDIR', True), d.getVar('PN', True))
     0010:    copy_license_files(lic_files_paths, destdir)
     0011:
     0012:
 *** 0013:do_populate_lic(d)
     0014:
File: 'do_populate_lic', lineno: 6, function: do_populate_lic
     0002:def do_populate_lic(d):
     0003:    """
     0004:    Populate LICENSE_DIRECTORY with licenses.
     0005:    """
 *** 0006:    lic_files_paths = find_license_files(d)
     0007:
     0008:    # The base directory we wrangle licenses to
     0009:    destdir = os.path.join(d.getVar('LICSSTATEDIR', True), d.getVar('PN', True))
     0010:    copy_license_files(lic_files_paths, destdir)
File: 'license.bbclass', lineno: 99, function: find_license_files
     0095:        lic_files_paths.append((os.path.basename(path), srclicfile))
     0096:
     0097:    v = FindVisitor()
     0098:    try:
 *** 0099:        v.visit_string(license_types)
     0100:    except oe.license.InvalidLicense as exc:
     0101:        bb.fatal('%s: %s' % (d.getVar('PF', True), exc))
     0102:    except SyntaxError:
     0103:        bb.warn("%s: Failed to parse it's LICENSE field." % (d.getVar('PF', True)))
File: '/media/build1/poky/meta/lib/oe/license.py', lineno: 38, function: visit_string
     0034:        new_elements = []
     0035:        elements = filter(lambda x: x.strip(), license_operator.split(licensestr))
     0036:        for pos, element in enumerate(elements):
     0037:            if license_pattern.match(element):
 *** 0038:                if pos > 0 and license_pattern.match(elements[pos-1]):
     0039:                    new_elements.append('&')
     0040:                element = '"' + element + '"'
     0041:            elif not license_operator.match(element):
     0042:                raise InvalidLicense(element)
Exception: TypeError: 'filter' object is not subscriptable

ERROR: Function failed: do_populate_lic
ERROR: Logfile of failure stored in: /media/build1/poky/build/tmp/work/i586-poky-linux/eglibc-initial/2.17-r3/temp/log.do_populate_lic.3275
ERROR: Task 9 (/media/build1/poky/meta/recipes-core/eglibc/eglibc-initial_2.17.bb, do_populate_lic) failed with exit code '1
"""

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobin/bitbake: Ensure early errors are shown to the user
Richard Purdie [Tue, 7 May 2013 08:23:47 +0000 (08:23 +0000)] 
bin/bitbake: Ensure early errors are shown to the user

Currently if errors occur early in the init process, the errors may
not be shown to the user. This change ensures that if a failure does
occur, the messages are flushed from the queue and shown to the user.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2/svn: Do not save authentication
Vanveerdeghem, Simon [Fri, 12 Apr 2013 06:35:38 +0000 (08:35 +0200)] 
fetch2/svn: Do not save authentication

Signed-off-by: Simon Vanveerdeghem <simon.vanveerdeghem@barco.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake: remove stack trace depth limitation
Bogdan Marinescu [Tue, 23 Apr 2013 13:43:45 +0000 (16:43 +0300)] 
bitbake: remove stack trace depth limitation

In case of an error in bitbake, the stack trace is limited to 5
items. This is an endless source of confusion and it makes bugs
reports impractical, since a full stack trace can't be included
in the bug report. This patch simply removes the depth limit.

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agodata_smart.py: add some internal bitbake variables to configuration hash
Laurentiu Palcu [Fri, 3 May 2013 14:37:17 +0000 (17:37 +0300)] 
data_smart.py: add some internal bitbake variables to configuration hash

Take __BBTASKS, __BBHANDLERS and __BBANONFUNCS into account when
computing the configuration hash.

[YOCTO #4447]

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoprserv/cooker: Handle PRService errors cleanly
Richard Purdie [Thu, 2 May 2013 14:46:11 +0000 (15:46 +0100)] 
prserv/cooker: Handle PRService errors cleanly

Current if the PR Service fails to start, bitbake carries on regardless or
hangs with no error message. This adds an exception and then handles it correctly
so the UIs correctly handle the error and exit cleanly.

[YOCTO #4010]

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agogoggle: fix invalid error check
Martin Donnelly [Thu, 25 Apr 2013 16:41:06 +0000 (17:41 +0100)] 
goggle: fix invalid error check

runCommand was returning 'error' but checking for 'err'
resulting in an exception.

Signed-off-by: Martin Donnelly <martin.donnelly@ge.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agodata: fix performance regression
Paul Eggleton [Mon, 15 Apr 2013 14:27:34 +0000 (15:27 +0100)] 
data: fix performance regression

BitBake commit 7c568132c54a21161de28907159f902462f1e2bb resulted in a
fairly serious performance regression during parsing, almost doubling
the time taken to do a full parse and almost certainly impacting
performance during building. The expandKeys function is called
frequently, and if we avoid using keys() and instead just use the normal
variable lookup mechanism, performance is restored.

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agofetch2: rename file with bad checksum instead of removing it completely
Martin Jansa [Fri, 12 Apr 2013 14:08:10 +0000 (16:08 +0200)] 
fetch2: rename file with bad checksum instead of removing it completely

* this can be useful when someone wan't to compare old file with
  bad checksum and new one

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: Refine sorting mechanisms in Hob
Cristiana Voicu [Thu, 11 Apr 2013 12:07:59 +0000 (15:07 +0300)] 
hob: Refine sorting mechanisms in Hob

-refine sorting functions for each column from recipe list page
and package list page
-sort correctly size column from packages list page
-set default sroting order and secondary sorting criteria
-make included on included recipes/packages no sortable

[YOCTO #2346 & #4194]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agomonitordisk.py: disable inode checking for btrfs
Robert Yang [Mon, 15 Apr 2013 08:47:36 +0000 (16:47 +0800)] 
monitordisk.py: disable inode checking for btrfs

The btrfs doesn't have static inode, so disable the inode check for it,
the previouse patch has set it:

minInode = None

But this is incorrect, the minInode is just a temporary variable, it
should be:

self.devDict[k][2] = None

[YOCTO #3609]

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoUpdate to version 1.19.0 for master
Richard Purdie [Wed, 10 Apr 2013 15:10:29 +0000 (16:10 +0100)] 
Update to version 1.19.0 for master

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoUpdate to version 1.18.0
Richard Purdie [Wed, 10 Apr 2013 15:08:53 +0000 (16:08 +0100)] 
Update to version 1.18.0

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agodata.py: Add a warning when expandKeys overwrites an existing key
Mark Hatle [Tue, 19 Mar 2013 20:28:51 +0000 (15:28 -0500)] 
data.py: Add a warning when expandKeys overwrites an existing key

When two variables are defined as:

${var} = "bar"
foo = "foobar"

The value of 'foo' when ${var} == foo becomes indeterminate.  We
want to warn a user when this situation has been encountered so they
can take corrective actions.

In the above example usually foo == bar, unless multilibs are enabled.
Then ml-foo = "ml-foobar".

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob:Only display scrolled list of images if needed
Ioana Grigoropol [Tue, 9 Apr 2013 10:25:48 +0000 (13:25 +0300)] 
hob:Only display scrolled list of images if needed

- do not set the Images window to always display a scroll bar if it is not needed

[Yocto #4171]
Signed-off-by: Ioana Grigoropol <ioanax.grigoropol@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake:hob: use a socks proxy mechanism for git
Cristian Iorga [Mon, 8 Apr 2013 16:06:33 +0000 (19:06 +0300)] 
bitbake:hob: use a socks proxy mechanism for git

Instead of a custom git proxy mechanism, Hob now
uses a SOCKS proxy in order to work with external
repos via the oe-git-proxy helper script.

Fixes [YOCTO #4187]

Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Cristian Iorga <ubik3000@gmail.com>
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agopropertydialog adjustments for package.bbclass
Andrei Dinu [Wed, 3 Apr 2013 11:58:47 +0000 (14:58 +0300)] 
propertydialog adjustments for package.bbclass

After moving the code from packageinfo.bbclass to
package.bbclass, minor adjustments were made to the
parsing of the package items.

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: giving focus to the search field loses the table sorting
Cristiana Voicu [Mon, 1 Apr 2013 10:36:25 +0000 (13:36 +0300)] 
hob: giving focus to the search field loses the table sorting

Giving focus to the search text field should not impact the table
sorting.

[YOCTO #4113]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: add tooltip on "clear search" button
Cristiana Voicu [Fri, 29 Mar 2013 13:40:18 +0000 (15:40 +0200)] 
hob: add tooltip on "clear search" button

[YOCTO #4116]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoRemoved popup when including a package
Andrei Dinu [Fri, 29 Mar 2013 13:26:37 +0000 (15:26 +0200)] 
Removed popup when including a package

Fixed the functionality which made an information
dialog pop up when including any package.

[HOB #4138]

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobb.tests.fetch: Opt-out for unittests that require network
Olof Johansson [Tue, 26 Mar 2013 08:28:38 +0000 (09:28 +0100)] 
bb.tests.fetch: Opt-out for unittests that require network

With this change, you can opt-out to skip unit tests that require an
internet connection. To opt-out, you'll set the environment variable
BB_SKIP_NETTESTS to 'yes'.

Signed-off-by: Olof Johansson <olof.johansson@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohob: Search strings and results should be persistent
Cristiana Voicu [Wed, 27 Mar 2013 15:20:07 +0000 (17:20 +0200)] 
hob: Search strings and results should be persistent

Now, the search results stay until I clear the search field,
so that I can manipulate the search results.

[YOCTO #4112 & #4117]
Signed-off-by: Cristiana Voicu <cristiana.voicu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agopackageselectionpage.py : added information to hob
Andrei Dinu [Thu, 28 Mar 2013 08:23:22 +0000 (10:23 +0200)] 
packageselectionpage.py : added information to hob

In order to have information for each package in hob,
a new item is added to the dictionary, represeting the
files that are brought in by each package.

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agohoblistmodel.py : passing the package information to hob
Andrei Dinu [Thu, 28 Mar 2013 08:23:21 +0000 (10:23 +0200)] 
hoblistmodel.py : passing the package information to hob

Added a new column to the model and also populating
it with the information brought in from the
packageinfo.bbclass.

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agopropertydialog.py : added 'Package files' functionality
Andrei Dinu [Thu, 28 Mar 2013 08:23:20 +0000 (10:23 +0200)] 
propertydialog.py : added 'Package files' functionality

Extended the packages page information with the
listing of the files brought in by every package.

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocooker.py : added variables related to cache_extra
Andrei Dinu [Thu, 28 Mar 2013 08:23:18 +0000 (10:23 +0200)] 
cooker.py : added variables related to cache_extra

So that the information added to cache_extra could
be accesed by hob, new variables were added in
the cooker.py.

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agocache_extra.py : added package information
Andrei Dinu [Thu, 28 Mar 2013 08:23:17 +0000 (10:23 +0200)] 
cache_extra.py : added package information

Added a new variable to cache_extra so that
the files brought in by a package can be
displayed in hob.

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agobitbake/fetch: Add git submodules fetcher
Richard Purdie [Fri, 22 Mar 2013 14:18:18 +0000 (14:18 +0000)] 
bitbake/fetch: Add git submodules fetcher

This adds very basic git submodule support to the fetcher. It can be
used by replacing a git:// url prefix with a gitsm:// prefix, otherwise
behaviour is the same as the git fetcher. Whilst this code should be
functional, its not as efficient as the usual git fetcher due to the
need to checkout the tree to fetch/update the submodule information. git
doesn't support submodule operations on the bare clones the standard git
fetcher uses which is also problematic.

This code does however give a starting point to people wanting to use
submodules.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agodoc/bitbake.1: Update with missing parameters
Richard Purdie [Mon, 25 Mar 2013 15:54:55 +0000 (15:54 +0000)] 
doc/bitbake.1: Update with missing parameters

Patch from Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com> to
add missing parameters to bitbake man page. Added hob to ui list too.

[YCOTO #4049]

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
12 years agoAdjust spacing on information windows
Andrei Dinu [Fri, 22 Mar 2013 15:20:10 +0000 (17:20 +0200)] 
Adjust spacing on information windows

[Hob #4070]

- tweaked the spacing on info windows

- fixed text bolding in simple settings dialog

Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>