Add new raptor error category to raptor parser (tool_didnt_return_version)

Set project name and codeline from subproject only if these properties are not found in project's job_props.ant.xml

Merge

Fix again releaseables dir not existing as erroronmissing dir is not supported

If a package does not have a definition in the sources.csv then it won't get built, so don't try to incorporate it into the system definition.

Avoid build failure if releaseable dir is not present

Made error messages more useful.

Should fail the build if we can't generate the system model.

Fixed paths and drives for executing script.

Added invokation to build.xml and adjusted path to backup locations to match my development environment.

Added tool for collating a build definition from the set of package defintions indicated by the sources.csv Location of tree of package defintions will need to be adjusted...

Change package baseline to PDK 2.0.d

Added 3 raptor error categories, adjusted severity of all categories

Fix regexp to take account of new codeline naming convention

Allow subprojects within a project repo and model file relative to the subproject dir

Generate error and summary file even if build reports no error

Restructured package build targets and sbs calls

Reverted category-property to build.family

Enable build tagging on Diamonds (default: 'test')

Fix to account of log file names for test builds

Added transformation of model file from version 3.0.0 to a version usable by Raptor

change made in smoketest.xml

change made in ats3_specialise_test_drop.pl script.

Add my computer following ATS installation instructions found at: https://sites.google.com/a/symbian.org/build_and_test/Home/ats-automated-test-system/ats-howtos,

changing smoketest.xml

catchup merge

Changed 'publish' to 'sf-publish' to only publish the zips, not all logs. Changed behaviour so that 'sf.spec.package.src.enable'=false stops the source repos from getting packaged, but not the RnD ones (which are extracted again in platform builds)

source syncing ftl - Added 'sf.spec.sourcesync.archive' preference to do 'hg archive' instead of 'hg sync'. If set and the repo sync is s'posed to be synced by tag it still does the slower clone+update behaviour.

platform build - added 'sf.os.compile.iterate' variable to iterate though the main targets rather than all at once to work around raptor problems with single model.

Split sf-run-analysis into bit-sized targets, which can run in parallel to reduce the post-build time on a platform build.