Page MenuHomeHEPForge
Files F9501580

No OneTemporary
Actions

View Options

This document is not UTF8. It was detected as ISO-8859-1 (Latin 1) and converted to UTF8 for display.
diff --git a/ChangeLog b/ChangeLog
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3959 +1,3961 @@
2013-06-10 Andy Buckley <andy.buckley@cern.ch>
+ * Add HepData URL for Inspire ID lookup to the rivet script.
+
* Fix analyses' info files which accidentally listed the Inspire
ID under the SpiresID metadata key.
2013-06-07 Andy Buckley <andy.buckley@cern.ch>
* Updating mk-analysis-html script to produce MathJax output
* Adding a version of Analysis::removeAnalysisObject() which takes
an AO pointer as its argument.
* bin/rivet: Adding pandoc-based conversion of TeX summary and
description strings to plain text on the terminal output.
* Add MathJax to rivet-mkhtml output, set up so the .info entries should render ok.
* Mark the OPAL 1993 analysis as UNVALIDATED: from the H++
benchmark runs it looks nothing like the data, and there are some
outstanding ambiguities.
2013-06-06 Andy Buckley <andy.buckley@cern.ch>
* Releasing 2.0.0b2 beta version.
2013-06-05 Andy Buckley <andy.buckley@cern.ch>
* Renaming Analysis::add() etc. to very explicit
addAnalysisObject(), sorting out shared_pointer polymorphism
issues via the Boost dynamic_pointer_cast, and adding a full set
of getHisto1D(), etc. explicitly named and typed accessors,
including ones with HepData dataset/axis ID signatures.
* Adding histo booking from an explicit reference Scatter2D (and
more placeholders for 2D histos / 3D scatters) and rewriting
existing autobooking to use this.
* Converting inappropriate uses of size_t to unsigned int in Analysis.
* Moving Analysis::addPlot to Analysis::add() (or reg()?) and
adding get() and remove() (or unreg()?)
* Fixing attempted abstraction of import fallbacks in rivet.util.import_ET().
* Removing broken attempt at histoDir() caching which led to all
histograms being registered under the same analysis name.
2013-06-04 Andy Buckley <andy.buckley@cern.ch>
* Updating the Cython version requirement to 0.18
* Adding Analysis::integrate() functions and tidying the Analysis.hh file a bit.
2013-06-03 Andy Buckley <andy.buckley@cern.ch>
* Adding explicit protection against using inf/nan scalefactors in
ATLAS_2011_S9131140 and H1_2000_S4129130.
* Making Analysis::scale noisly complain about invalid
scalefactors.
2013-05-31 Andy Buckley <andy.buckley@cern.ch>
* Reducing the TeX main memory to ~500MB. Turns out that it *can*
be too large with new versions of TeXLive!
2013-05-30 Andy Buckley <andy.buckley@cern.ch>
* Reverting bookScatter2D behaviour to never look at ref data, and
updating a few affected analyses. This should fix some bugs with
doubled datapoints introduced by the previous behaviour+addPoint.
* Adding a couple of minor Utils.hh and MathUtils.hh features.
2013-05-29 Andy Buckley <andy.buckley@cern.ch>
* Removing Constraints.hh header.
* Minor bugfixes and improvements in Scatter2D booking and MC_JetAnalysis.
2013-05-28 Andy Buckley <andy.buckley@cern.ch>
* Removing defunct HistoFormat.hh and HistoHandler.{hh,cc}
2013-05-27 Andy Buckley <andy.buckley@cern.ch>
* Removing includes of Logging.hh, RivetYODA.hh, and
ParticleIdUtils.hh from analyses (and adding an include of
ParticleIdUtils.hh to Analysis.hh)
* Removing now-unused .fhh files.
* Removing lots of unnecessary .fhh includes from core classes:
everything still compiling ok. A good opportunity to tidy this up
before the release.
* Moving the rivet-completion script from the data dir to bin (the
completion is for scripts in bin, and this makes development
easier).
* Updating bash completion scripts for YODA format and
compare-histos -> rivet-cmphistos.
2013-05-23 Andy Buckley <andy.buckley@cern.ch>
* Adding Doxy comments and a couple of useful convenience functions to Utils.hh.
* Final tweaks to ATLAS ttbar jet veto analysis (checked logic with Kiran Joshi).
2013-05-15 Andy Buckley <andy.buckley@cern.ch>
* Many 1.0 -> weight bugfixes in ATLAS_2011_I945498.
* yaml-cpp v3 support re-introduced in .info parsing.
* Lots of analysis clean-ups for YODA TODO issues.
2013-05-13 Andy Buckley <andy.buckley@cern.ch>
* Analysis histo booking improvements for Scatter2D, placeholders
for 2D histos, and general tidying.
2013-05-12 Andy Buckley <andy.buckley@cern.ch>
* Adding configure-time differentiation between yaml-cpp API versions 3 and 5.
2013-05-07 Andy Buckley <andy.buckley@cern.ch>
* Converting info file reading to use the yaml-cpp 0.5.x API.
2013-04-12 Andy Buckley <andy.buckley@cern.ch>
* Tagging as 2.0.0b1
2013-04-04 Andy Buckley <andy.buckley@cern.ch>
* Removing bundling of yaml-cpp: it needs to be installed by the
user / bootstrap script from now on.
2013-04-03 Andy Buckley <andy.buckley@cern.ch>
* Removing svn:external m4 directory, and converting Boost
detection to use better boost.m4 macros.
2013-03-22 Andy Buckley <andy.buckley@cern.ch>
* Moving PID consts to the PID namespace and corresponding code
updates and opportunistic clean-ups.
* Adding Particle::fromDecay() method.
2013-03-09 Andy Buckley <andy.buckley@cern.ch>
* Version bump to 2.0.0b1 in anticipation of first beta release.
* Adding many more 'popular' particle ID code named-consts and
aliases, and updating the RapScheme enum with ETA -> ETARAP, and
fixing affected analyses (plus other opportunistic tidying / minor
bug-fixing).
* Fixing a symbol misnaming in ATLAS_2012_I1119557.
2013-03-07 Andy Buckley <andy.buckley@cern.ch>
* Renaming existing uses of ParticleVector to the new 'Particles' type.
* Updating util classes, projections, and analyses to deal with
the HepMC return value changes.
* Adding new Particle(const GenParticle*) constructor.
* Converting Particle::genParticle() to return a const pointer
rather than a reference, for the same reason as below (+
consistency within Rivet and with the HepMC pointer-centric coding
design).
* Converting Event to use a different implementation of original
and modified GenParticles, and to manage the memory in a more
future-proof way. Event::genParticle() now returns a const pointer
rather than a reference, to signal that the user is leaving the
happy pastures of 'normal' Rivet behind.
* Adding a Particles typedef by analogy to Jets, and in preference
to the cumbersome ParticleVector.
* bin/: Lots of tidying/pruning of messy/defunct scripts.
* Creating spiresbib, util, and plotinfo rivet python module
submodules: this eliminates lighthisto and the standalone
spiresbib modules. Util contains convenience functions for Python
version testing, clean ElementTree import, and process renaming,
for primary use by the rivet-* scripts.
* Removing defunct scripts that have been replaced/obsoleted by YODA.
2013-03-06 Andy Buckley <andy.buckley@cern.ch>
* Fixing doc build so that the reference histos and titles are
~correctly documented. We may want to truncate some of the lists!
2013-03-06 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ATLAS_2012_I1125575 analysis
* Converted rivet-mkhtml to yoda
* Introduced rivet-cmphistos as yoda based replacement for compare-histos
2013-03-05 Andy Buckley <andy.buckley@cern.ch>
* Replacing all AIDA ref data with YODA versions.
* Fixing the histograms entries in the documentation to be
tolerant to plotinfo loading failures.
* Making the findDatafile() function primarily find YODA data
files, then fall back to AIDA. The ref data loader will use the
appropriate YODA format reader.
2013-02-05 David Grellscheid <David.Grellscheid@durham.ac.uk>
* include/Rivet/Math/MathUtils.hh: added BWspace bin edge method
to give equal-area Breit-Wigner bins
2013-02-01 Andy Buckley <andy.buckley@cern.ch>
* Adding an element to the PhiMapping enum and a new mapAngle(angle, mapping) function.
* Fixes to Vector3::azimuthalAngle and Vector3::polarAngle calculation (using the mapAngle functions).
2013-01-25 Frank Siegert <frank.siegert@cern.ch>
* Split MC_*JETS analyses into three separate bits:
MC_*INC (inclusive properties)
MC_*JETS (jet properties)
MC_*KTSPLITTINGS (kT splitting scales).
2013-01-22 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix TeX variable in the rivetenv scripts, especially for csh
2012-12-21 Andy Buckley <andy.buckley@cern.ch>
* Version 1.8.2 release!
2012-12-20 Andy Buckley <andy.buckley@cern.ch>
* Adding ATLAS_2012_I1119557 analysis (from Roman Lysak and Lily Asquith).
2012-12-18 Andy Buckley <andy.buckley@cern.ch>
* Adding TOTEM_2012_002 analysis, from Sercan Sen.
2012-12-18 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2011_I954992 analysis
2012-12-17 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2012_I1193338 analysis
* Fixed xi cut in ATLAS_2011_I894867
2012-12-17 Andy Buckley <andy.buckley@cern.ch>
* Adding analysis descriptions to the HTML analysis page ToC.
2012-12-14 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2012_PAS_FWD_11_003 analysis
* Added LHCB_2012_I1119400 analysis
2012-12-12 Andy Buckley <andy.buckley@cern.ch>
* Correction to jet acceptance in CMS_2011_S9120041, from Sercan Sen: thanks!
2012-12-12 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2012_PAS_QCD_11_010 analysis
2012-12-07 Andy Buckley <andy.buckley@cern.ch>
* Version number bump to 1.8.2 -- release approaching.
* Rewrite of ALICE_2012_I1181770 analysis to make it a bit more sane and acceptable.
* Adding a note on FourVector and FourMomentum that operator- and
operator-= invert both the space and time components: use of -=
can result in a vector with negative energy.
* Adding particlesByRapidity and particlesByAbsRapidity to FinalState.
2012-12-07 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ALICE_2012_I1181770 analysis
* Bump version to 1.8.2
2012-12-06 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ATLAS_2012_I1188891 analysis
* Added ATLAS_2012_I1118269 analysis
* Added CMS_2012_I1184941 analysis
* Added LHCB_2010_I867355 analysis
* Added TGraphErrors support to root2flat
2012-11-27 Andy Buckley <andy.buckley@cern.ch>
* Converting CMS_2012_I1102908 analysis to use YODA.
* Adding XLabel and YLabel setting in histo/profile/scatter booking.
2012-11-27 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix make-plots png creation for SL5
2012-11-23 Peter Richardson <peter.richardson@durham.ac.uk>
* Added ATLAS_2012_CONF_2012_153 4-lepton SUSY search
2012-11-17 Andy Buckley <andy.buckley@cern.ch>
* Adding MC_PHOTONS by Steve Lloyd and AB, for testing general
unisolated photon properties, especially those associated with
charged leptons (e and mu).
2012-11-16 Andy Buckley <andy.buckley@cern.ch>
* Adding MC_PRINTEVENT, a convenient (but verbose!) analysis for
printing out event details to stdout.
2012-11-15 Andy Buckley <andy.buckley@cern.ch>
* Removing the long-unused/defunct autopackage system.
2012-11-15 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added LHCF_2012_I1115479 analysis
* Added ATLAS_2011_I894867 analysis
2012-11-14 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2012_I1102908 analysis
2012-11-14 Andy Buckley <andy.buckley@cern.ch>
* Converting the argument order of logspace, clarifying the
arguments, updating affected code, and removing Analysis::logBinEdges.
* Merging updates from the AIDA maintenance branch up to r4002
(latest revision for next merges is r4009).
2012-11-11 Andy Buckley <andy.buckley@cern.ch>
* include/Math/: Various numerical fixes to Vector3::angle and
changing the 4 vector mass treatment to permit spacelike
virtualities (in some cases even the fuzzy isZero assert check was
being violated). The angle check allows a clean-up of some
workaround code in MC_VH2BB.
2012-10-15 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2012_I1107658 analysis
2012-10-11 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CDF_2012_NOTE10874 analysis
2012-10-04 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ATLAS_2012_I1183818 analysis
2012-07-17 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Cleanup and multiple fixes in CMS_2011_S9120041
* Bugfixed in ALEPH_2004_S5765862 and ATLAS_2010_CONF_2010_049
(thanks to Anil Pratap)
2012-08-09 Andy Buckley <andy.buckley@cern.ch>
* Fixing aida2root command-line help message and converting to TH*
rather than TGraph by default.
2012-07-24 Andy Buckley <andy.buckley@cern.ch>
* Improvements/migrations to rivet-mkhtml, rivet-mkanalysis, and
rivet-buildplugin.
2012-07-17 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Add CMS_2012_I1087342
2012-07-12 Andy Buckley <andy.buckley@cern.ch>
* Fix rivet-mkanalysis a bit for YODA compatibility.
2012-07-05 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Version 1.8.1!
2012-07-05 Holger Schulz <holger.schulz@physik.hu-berlin.de>
* Add ATLAS_2011_I945498
2012-07-03 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Bugfix for transverse mass (thanks to Gavin Hesketh)
2012-06-29 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Merge YODA branch into trunk. YODA is alive!!!!!!
2012-06-26 Holger Schulz <holger.schulz@physik.hu-berlin.de>
* Add ATLAS_2012_I1091481
2012-06-20 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added D0_2011_I895662: 3-jet mass
2012-04-24 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* fixed a few bugs in rivet-rmgaps
* Added new TOTEM dN/deta analysis
2012-03-19 Andy Buckley <andy.buckley@cern.ch>
* Version 1.8.0!
* src/Projections/UnstableFinalState.cc: Fix compiler warning.
* Version bump for testing: 1.8.0beta1.
* src/Core/AnalysisInfo.cc: Add printout of YAML parser exception error messages to aid debugging.
* bin/Makefile.am: Attempt to fix rivet-nopy build on SLC5.
* src/Analyses/LHCB_2010_S8758301.cc: Add two missing entries to the PDGID -> lifetime map.
* src/Projections/UnstableFinalState.cc: Extend list of vetoed particles to include reggeons.
2012-03-16 Andy Buckley <andy.buckley@cern.ch>
* Version change to 1.8.0beta0 -- nearly ready for long-awaited release!
* pyext/setup.py.in: Adding handling for the YAML library: fix for
Genser build from Anton Karneyeu.
* src/Analyses/LHCB_2011_I917009.cc: Hiding lifetime-lookup error
message if the offending particle is not a hadron.
* include/Rivet/Math/MathHeader.hh: Using unnamespaced std::isnan
and std::isinf as standard.
2012-03-16 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Improve default plot behaviour for 2D histograms
2012-03-15 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Make ATLAS_2012_I1084540 less verbose, and general code
cleanup of that analysis.
* New-style plugin hook in ATLAS_2011_I926145,
ATLAS_2011_I944826 and ATLAS_2012_I1084540
* Fix compiler warnings in ATLAS_2011_I944826 and CMS_2011_S8973270
* CMS_2011_S8941262: Weights are double, not int.
* disable inRange() tests in test/testMath.cc until we have a proper
fix for the compiler warnings we see on SL5.
2012-03-07 Andy Buckley <andy.buckley@cern.ch>
* Marking ATLAS_2011_I919017 as VALIDATED (this should have
happened a long time ago) and adding more references.
2012-02-28 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* lighthisto.py: Caching for re.compile(). This speeds up aida2flat
and flat2aida by more than an order of magnitude.
2012-02-27 Andy Buckley <andy.buckley@cern.ch>
* doc/mk-analysis-html: Adding more LaTeX/text -> HTML conversion
replacements, including better <,> handling.
2012-02-26 Andy Buckley <andy.buckley@cern.ch>
* Adding CMS_2011_S8973270, CMS_2011_S8941262, CMS_2011_S9215166,
CMS_QCD_10_024, from CMS.
* Adding LHCB_2011_I917009 analysis, from Alex Grecu.
* src/Core/Analysis.cc, include/Rivet/Analysis.hh: Add a numeric-arg version of histoPath().
2012-02-24 Holger Schulz <holger.schulz@physik.hu-berlin.de>
* Adding ATLAS Ks/Lambda analysis.
2012-02-20 Andy Buckley <andy.buckley@cern.ch>
* src/Analyses/ATLAS_2011_I925932.cc: Using new overflow-aware
normalize() in place of counters and scale(..., 1/count)
2012-02-14 Andy Buckley <andy.buckley@cern.ch>
* Splitting MC_GENERIC to put the PDF and PID plotting into
MC_PDFS and MC_IDENTIFIED respectively.
* Renaming MC_LEADINGJETS to MC_LEADJETUE.
2012-02-14 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* DELPHI_1996_S3430090 and ALEPH_1996_S3486095:
fix rapidity vs {Thrust,Sphericity}-axis.
2012-02-14 Andy Buckley <andy.buckley@cern.ch>
* bin/compare-histos: Don't attempt to remove bins from MC histos
where they aren't found in the ref file, if the ref file is not
expt data, or if the new --no-rmgapbins arg is given.
* bin/rivet: Remove the conversion of requested analysis names to
upper-case: mixed-case analysis names will now work.
2012-02-14 Frank Siegert <frank.siegert@cern.ch>
* Bugfixes and improvements for MC_TTBAR:
- Avoid assert failure with logspace starting at 0.0
- Ignore charged lepton in jet finding (otherwise jet multi is always
+1).
- Add some dR/deta/dphi distributions as noted in TODO
- Change pT plots to logspace as well (to avoid low-stat high pT bins)
2012-02-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* rivet-mkhtml -c option now has the semantics of a .plot
file. The contents are appended to the dat output by
compare-histos.
2012-02-09 David Grellscheid <david.grellscheid@durham.ac.uk>
* Fixed broken UnstableFS behaviour
2012-01-25 Frank Siegert <frank.siegert@cern.ch>
* Improvements in make-plots:
- Add PlotTickLabels and RatioPlotTickLabels options (cf.
make-plots.txt)
- Make ErrorBars and ErrorBands non-exclusive (and change
their order, such that Bars are on top of Bands)
2012-01-25 Holger Schulz <holger.schulz@physik.hu-berlin.de>
* Add ATLAS diffractive gap analysis
2012-01-23 Andy Buckley <andy.buckley@cern.ch>
* bin/rivet: When using --list-analyses, the analysis summary is
now printed out when log level is <= INFO, rather than < INFO.
The effect on command line behaviour is that useful identifying
info is now printed by default when using --list-analyses, rather
than requiring --list-analyses -v. To get the old behaviour,
e.g. if using the output of rivet --list-analyses for scripting,
now use --list-analyses -q.
2012-01-22 Andy Buckley <andy.buckley@cern.ch>
* Tidying lighthisto, including fixing the order in which +- error
values are passed to the Bin constructor in fromFlatHisto.
2012-01-16 Frank Siegert <frank.siegert@cern.ch>
* Bugfix in ATLAS_2012_I1083318: Include non-signal neutrinos in
jet clustering.
* Add first version of ATLAS_2012_I1083318 (W+jets). Still
UNVALIDATED until final happiness with validation plots arises and
data is in Hepdata.
* Bugfix in ATLAS_2010_S8919674: Really use neutrino with highest
pT for Etmiss. Doesn't seem to make very much difference, but is
more correct in principle.
2012-01-16 Peter Richardson <peter.richardson@durham.ac.uk>
* Fixes to ATLAS_20111_S9225137 to include reference data
2012-01-13 Holger Schulz <holger.schulz@physik.hu-berlin.de>
* Add ATLAS inclusive lepton analysis
2012-01-12 Hendrik Hoeth <hendrik.hoeth@durham.ac.uk>
* Font selection support in rivet-mkhtml
2012-01-11 Peter Richardson <peter.richardson@durham.ac.uk>
* Added pi0 to list of particles.
2012-01-11 Andy Buckley <andy.buckley@cern.ch>
* Removing references to Boost random numbers.
2011-12-30 Andy Buckley <andy.buckley@cern.ch>
* Adding a placeholder rivet-which script (not currently
installed).
* Tweaking to avoid a very time-consuming debug printout in
compare-histos with the -v flag, and modifying the Rivet env vars
in rivet-mkhtml before calling compare-histos to eliminate
problems induced by relative paths (i.e. "." does not mean the
same thing when the directory is changed within the script).
2011-12-12 Andy Buckley <andy.buckley@cern.ch>
* Adding a command line completion function for rivet-mkhtml.
2011-12-12 Frank Siegert <frank.siegert@cern.ch>
* Fix for factor of 2.0 in normalisation of CMS_2011_S9086218
* Add --ignore-missing option to rivet-mkhtml to ignore non-existing
AIDA files.
2011-12-06 Andy Buckley <andy.buckley@cern.ch>
* Include underflow and overflow bins in the normalisation when
calling Analysis::normalise(h)
2011-11-23 Andy Buckley <andy.buckley@cern.ch>
* Bumping version to 1.8.0alpha0 since the Jet interface changes
are quite a major break with backward compatibility (although the
vast majority of analyses should be unaffected).
* Removing crufty legacy stuff from the Jet class -- there is
never any ambiguity between whether Particle or FourMomentum
objects are the constituents now, and the jet 4-momentum is set
explicitly by the jet alg so that e.g. there is no mismatch if the
FastJet pt recombination scheme is used.
* Adding default do-nothing implementations of Analysis::init()
and Analysis::finalize(), since it is possible for analysis
implementations to not need to do anything in these methods, and
forcing analysis authors to write do-nothing boilerplate code is
not "the Rivet way"!
2011-11-19 Andy Buckley <andy.buckley@cern.ch>
* Adding variant constructors to FastJets with a more natural
Plugin* argument, and decrufting the constructor implementations a
bit.
* bin/rivet: Adding a more helpful error message if the rivet
module can't be loaded, grouping the option parser options,
removing the -A option (this just doesn't seem useful anymore),
and providing a --pwd option as a shortcut to append "." to the
search path.
2011-11-18 Andy Buckley <andy.buckley@cern.ch>
* Adding a guide to compiling a new analysis template to the
output message of rivet-mkanalysis.
2011-11-11 Andy Buckley <andy.buckley@cern.ch>
* Version 1.7.0 release!
* Protecting the OPAL 2004 analysis against NaNs in the
hemispheres projection -- I can't track the origin of these and
suspect some occasional memory corruption.
2011-11-09 Andy Buckley <andy@insectnation.org>
* Renaming source files for EXAMPLE and
PDG_HADRON_MULTIPLICITIES(_RATIOS) analyses to match the analysis
names.
* Cosmetic fixes in ATLAS_2011_S9212183 SUSY analysis.
* Adding new ATLAS W pT analysis from Elena Yatsenko (slightly adapted).
2011-10-20 Frank Siegert <frank.siegert@cern.ch>
* Extend API of W/ZFinder to allow for specification of input final
state in which to search for leptons/photons.
2011-10-19 Andy Buckley <andy@insectnation.org>
* Adding new version of LHCB_2010_S8758301, based on submission
from Alex Grecu. There is some slightly dodgy-looking GenParticle*
fiddling going on, but apparently it's necessary (and hopefully robust).
2011-10-17 Andy Buckley <andy@insectnation.org>
* bin/rivet-nopy linker line tweak to make compilation work with
GCC 4.6 (-lHepMC has to be explicitly added for some reason).
2011-10-13 Frank Siegert <frank.siegert@cern.ch>
* Add four CMS QCD analyses kindly provided by CMS.
2011-10-12 Andy Buckley <andy@insectnation.org>
* Adding a separate test program for non-matrix/vector math
functions, and adding a new set of int/float literal arg tests for
the inRange functions in it.
* Adding a jet multiplicity plot for jets with pT > 30 GeV to
MC_TTBAR.
2011-10-11 Andy Buckley <andy@insectnation.org>
* Removing SVertex.
2011-10-11 James Monk <jmonk@cern.ch>
* root2flat was missing the first bin (plus spurious last bin)
* My version of bash does not understand the pipe syntax |& in rivet-buildplugin
2011-09-30 James Monk <jmonk@cern.ch>
* Fix bug in ATLAS_2010_S8817804 that misidentified the akt4 jets
as akt6
2011-09-29 Andy Buckley <andy@insectnation.org>
* Converting FinalStateHCM to a slightly more general
DISFinalState.
2011-09-26 Andy Buckley <andy@insectnation.org>
* Adding a default libname argument to rivet-buildplugin. If the
first argument doesn't have a .so library suffix, then use
RivetAnalysis.so as the default.
2011-09-19 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* make-plots: Fixing regex for \physicscoor. Adding "FrameColor"
option.
2011-09-17 Andy Buckley <andy@insectnation.org>
* Improving interactive metadata printout, by not printing
headings for missing info.
* Bumping the release number to 1.7.0alpha0, since with these
SPIRES/Inspire changes and the MissingMomentum API change we need
more than a minor release.
* Updating the mkanalysis, BibTeX-grabbing and other places that
care about analysis SPIRES IDs to also be able to handle the new
Inspire system record IDs. The missing link is getting to HepData
from an Inspire code...
* Using the .info file rather than an in-code declaration to
specify that an analysis needs cross-section information.
* Adding Inspire support to the AnalysisInfo and Analysis
interfaces. Maybe we can find a way to combine the two,
e.g. return the SPIRES code prefixed with an "S" if no Inspire ID
is available...
2011-09-17 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ALICE_2011_S8909580 (strange particle production at 900 GeV)
* Feed-down correction in ALICE_2011_S8945144
2011-09-16 Andy Buckley <andy@insectnation.org>
* Adding ATLAS track jet analysis, modified from the version
provided by Seth Zenz: ATLAS_2011_I919017. Note that this analysis
is currently using the Inspire ID rather than the Spires one:
we're clearly going to have to update the API to handle Inspire
codes, so might as well start now...
2011-09-14 Andy Buckley <andy@insectnation.org>
* Adding the ATLAS Z pT measurement at 7 TeV (ATLAS_2011_S9131140)
and an MC analysis for VH->bb events (MC_VH2BB).
2011-09-12 Andy Buckley <andy@insectnation.org>
* Removing uses of getLog, cout, cerr, and endl from all standard
analyses and projections, except in a very few special cases.
2011-09-10 Andy Buckley <andy@insectnation.org>
* Changing the behaviour and interface of the MissingMomentum
projection to calculate vector ET correctly. This was previously
calculated according to the common definition of -E*sin(theta) of
the summed visible 4-momentum in the event, but that is incorrect
because the timelike term grows monotonically. Instead, transverse
2-vectors of size ET need to be constructed for each visible
particle, and vector-summed in the transverse plane.
The rewrite of this behaviour made it opportune to make an API
improvement: the previous method names scalarET/vectorET() have
been renamed to scalar/vectorEt() to better match the Rivet
FourMomentum::Et() method, and MissingMomentum::vectorEt() now
returns a Vector3 rather than a double so that the transverse
missing Et direction is also available.
Only one data analysis has been affected by this change in
behaviour: the D0_2004_S5992206 dijet delta(phi) analysis. It's
expected that this change will not be very significant, as it is
a *veto* on significant missing ET to reduce non-QCD
contributions. MC studies using this analysis ~always run with QCD
events only, so these contributions should be small. The analysis
efficiency may have been greatly improved, as fewer events will
now fail the missing ET veto cut.
* Add sorting of the ParticleVector returned by the ChargedLeptons
projection.
* configure.ac: Adding a check to make sure that no-one tries to
install into --prefix=$PWD.
2011-09-04 Andy Buckley <andy@insectnation.org>
* lighthisto fixes from Christian Roehr.
2011-08-26 Andy Buckley <andy@insectnation.org>
* Removing deprecated features: the setBeams(...) method on
Analysis, the MaxRapidity constant, the split(...) function, the
default init() method from AnalysisHandler and its test, and the
deprecated TotalVisibleMomentum and PVertex projections.
2011-08-23 Andy Buckley <andy@insectnation.org>
* Adding a new DECLARE_RIVET_PLUGIN wrapper macro to hide the
details of the plugin hook system from analysis authors. Migration
of all analyses and the rivet-mkanalysis script to use this as the
standard plugin hook syntax.
* Also call the --cflags option on root-config when using the
--root option with rivet-biuldplugin (thanks to Richard Corke for
the report)
2011-08-23 Frank Siegert <frank.siegert@cern.ch>
* Added ATLAS_2011_S9126244
* Added ATLAS_2011_S9128077
2011-08-23 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ALICE_2011_S8945144
* Remove obsolete setBeams() from the analyses
* Update CMS_2011_S8957746 reference data to the official numbers
* Use Inspire rather than Spires.
2011-08-19 Frank Siegert <frank.siegert@cern.ch>
* More NLO parton level generator friendliness: Don't crash or fail when
there are no beam particles.
* Add --ignore-beams option to skip compatibility check.
2011-08-09 David Mallows <dave.mallows@gmail.com>
* Fix aida2flat to ignore empty dataPointSet
2011-08-07 Andy Buckley <andy@insectnation.org>
* Adding TEXINPUTS and LATEXINPUTS prepend definitions to the
variables provided by rivetenv.(c)sh. A manual setting of these
variables that didn't include the Rivet TEXMFHOME path was
breaking make-plots on lxplus, presumably since the system LaTeX
packages are so old there.
2011-08-02 Frank Siegert <frank.siegert@cern.ch>
Version 1.6.0 release!
2011-08-01 Frank Siegert <frank.siegert@cern.ch>
* Overhaul of the WFinder and ZFinder projections, including a change
of interface. This solves potential problems with leptons which are not
W/Z constituents being excluded from the RemainingFinalState.
2011-07-29 Andy Buckley <andy@insectnation.org>
* Version 1.5.2 release!
* New version of aida2root from James Monk.
2011-07-29 Frank Siegert <frank.siegert@cern.ch>
* Fix implementation of --config file option in make-plots.
2011-07-27 David Mallows <dave.mallows@gmail.com>
* Updated MC_TTBAR.plot to reflect updated analysis.
2011-07-25 Andy Buckley <andy@insectnation.org>
* Adding a useTransverseMass flag method and implementation to
InvMassFinalState, and using it in the WFinder, after feedback
from Gavin Hesketh. This was the neatest way I could do it :S Some
other tidying up happened along the way.
* Adding transverse mass massT and massT2 methods and functions
for FourMomentum.
2011-07-22 Frank Siegert <frank.siegert@cern.ch>
* Added ATLAS_2011_S9120807
* Add two more observables to MC_DIPHOTON and make its isolation cut
more LHC-like
* Add linear photon pT histo to MC_PHOTONJETS
2011-07-20 Andy Buckley <andy@insectnation.org>
* Making MC_TTBAR work with semileptonic ttbar events and generally
tidying the code.
2011-07-19 Andy Buckley <andy@insectnation.org>
* Version bump to 1.5.2.b01 in preparation for a release in the
very near future.
2011-07-18 David Mallows <dave.mallows@gmail.com>
* Replaced MC_TTBAR: Added t,tbar reconstruction. Not yet working.
2011-07-18 Andy Buckley <andy@insectnation.org>
* bin/rivet-buildplugin.in: Pass the AM_CXXFLAGS
variable (including the warning flags) to the C++ compiler when
building user analysis plugins.
* include/LWH/DataPointSet.h: Fix accidental setting of errorMinus
= scalefactor * error_Plus_. Thanks to Anton Karneyeu for the bug
report!
2011-07-18 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added CMS_2011_S8884919 (charged hadron multiplicity in NSD
events corrected to pT>=0).
* Added CMS_2010_S8656010 and CMS_2010_S8547297 (charged
hadron pT and eta in NSD events)
* Added CMS_2011_S8968497 (chi_dijet)
* Added CMS_2011_S8978280 (strangeness)
2011-07-13 Andy Buckley <andy@insectnation.org>
* Rivet PDF manual updates, to not spread disinformation about
bootstrapping a Genser repo.
2011-07-12 Andy Buckley <andy@insectnation.org>
* bin/make-plots: Protect property reading against unstripped \r
characters from DOS newlines.
* bin/rivet-mkhtml: Add a -M unmatch regex flag (note that these
are matching the analysis path rather than individual histos on
this script), and speed up the initial analysis identification and
selection by avoiding loops of regex comparisons for repeats of
strings which have already been analysed.
* bin/compare-histos: remove the completely (?) unused histogram
list, and add -m and -M regex flags, as for aida2flat and
flat2aida.
2011-06-30 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* fix fromFlat() in lighthistos: It would ignore histogram paths
before.
* flat2aida: preserve histogram order from .dat files
2011-06-27 Andy Buckley <andy@insectnation.org>
* pyext/setup.py.in: Use CXXFLAGS and LDFLAGS safely in the Python
extension build, and improve the use of build/src directory
arguments.
2011-06-23 Andy Buckley <andy@insectnation.org>
* Adding a tentative rivet-updateanalyses script, based on
lhapdf-getdata, which will download new analyses as requested. We
could change our analysis-providing behaviour a bit to allow this
sort of delivery mechanism to be used as the normal way of getting
analysis updates without us having to make a whole new Rivet
release. It is nice to be able to identify analyses with releases,
though, for tracking whether bugs have been addressed.
2011-06-10 Frank Siegert <frank.siegert@cern.ch>
* Bugfixes in WFinder.
2011-06-10 Andy Buckley <andy@insectnation.org>
* Adding \physicsxcoor and \physicsycoor treatment to make-plots.
2011-06-06 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Allow for negative cross-sections. NLO tools need this.
* make-plots: For RatioPlotMode=deviation also consider the MC
uncertainties, not just data.
2011-06-04 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Add support for goodness-of-fit calculations to make-plots.
The results are shown in the legend, and one histogram can
be selected to determine the color of the plot margin. See
the documentation for more details.
2011-06-04 Andy Buckley <andy@insectnation.org>
* Adding auto conversion of Histogram2D to DataPointSets in the
AnalysisHandler _normalizeTree method.
2011-06-03 Andy Buckley <andy@insectnation.org>
* Adding a file-weight feature to the Run object, which will
optionally rescale the weights in the provided HepMC files. This
should be useful for e.g. running on multiple differently-weighted
AlpGen HepMC files/streams. The new functionality is used by the
rivet command via an optional weight appended to the filename with
a colon delimiter, e.g. "rivet fifo1.hepmc fifo2.hepmc:2.31"
2011-06-01 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Add BeamThrust projection
2011-05-31 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix LIBS for fastjet-3.0
* Add basic infrastructure for Taylor plots in make-plots
* Fix OPAL_2004_S6132243: They are using charged+neutral.
* Release 1.5.1
2011-05-22 Andy Buckley <andy@insectnation.org>
* Adding plots of stable and decayed PID multiplicities to
MC_GENERIC (useful for sanity-checking generator setups).
* Removing actually-unused ProjectionApplier.fhh forward
declaration header.
2011-05-20 Andy Buckley <andy@insectnation.org>
* Removing import of ipython shell from rivet-rescale, having just
seen it throw a multi-coloured warning message on a student's
lxplus Rivet session!
* Adding support for the compare-histos --no-ratio flag when using
rivet-mkhtml. Adding --rel-ratio, --linear, etc. is an exercise
for the enthusiast ;-)
2011-05-10 Andy Buckley <andy@insectnation.org>
* Internal minor changes to the ProjectionHandler and
ProjectionApplier interfaces, in particular changing the
ProjectionHandler::create() function to be called getInstance and
to return a reference rather than a pointer. The reference change
is to make way for an improved singleton implementation, which
cannot yet be used due to a bug in projection memory
management. The code of the improved singleton is available, but
commented out, in ProjectionManager.hh to allow for easier
migration and to avoid branching.
2011-05-08 Andy Buckley <andy@insectnation.org>
* Extending flat2aida to be able to read from and write to
stdin/out as for aida2flat, and also eliminating the internal
histo parsing representation in favour of the one in
lighthisto. lighthisto's fromFlat also needed a bit of an
overhaul: it has been extended to parse each histo's chunk of
text (including BEGIN and END lines) in fromFlatHisto, and for
fromFlat to parse a collection of histos from a file, in keeping
with the behaviour of fromDPS/fromAIDA. Merging into Professor is
now needed.
* Extending aida2flat to have a better usage message, to accept
input from stdin for command chaining via pipes, and to be a bit
more sensibly internally structured (although it also now has to
hold all histos in memory before writing out -- that shouldn't be
a problem for anything other than truly huge histo files).
2011-05-04 Andy Buckley <andy@insectnation.org>
* compare-histos: If using --mc-errs style, prefer dotted and
dashdotted line styles to dashed, since dashes are often too long
to be distinguishable from solid lines. Even better might be to
always use a solid line for MC errs style, and to add more colours.
* rivet-mkhtml: use a no-mc-errors drawing style by default,
to match the behaviour of compare-histos, which it calls. The
--no-mc-errs option has been replaced with an --mc-errs option.
2011-05-04 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Ignore duplicate files in compare-histos.
2011-04-25 Andy Buckley <andy@insectnation.org>
* Adding some hadron-specific N and sumET vs. |eta| plots to MC_GENERIC.
* Re-adding an explicit attempt to get the beam particles, since
HepMC's IO_HERWIG seems to not always set them even though it's
meant to.
2011-04-19 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added ATLAS_2011_S9002537 W asymmetry analysis
2011-04-14 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* deltaR, deltaPhi, deltaEta now available in all combinations of
FourVector, FourMomentum, Vector3, doubles. They also accept jets
and particles as arguments now.
2011-04-13 David Grellscheid <david.grellscheid@durham.ac.uk>
* added ATLAS 8983313: 0-lepton BSM
2011-04-01 Andy Buckley <andy@insectnation.org>
* bin/rivet-mkanalysis: Don't try to download SPIRES or HepData
info if it's not a standard analysis (i.e. if the SPIRES ID is not
known), and make the default .info file validly parseable by YAML,
which was an unfortunate gotcha for anyone writing a first
analysis.
2011-03-31 Andy Buckley <andy@insectnation.org>
* bin/compare-histos: Write more appropriate ratio plot labels
when not comparing to data, and use the default make-plots labels
when comparing to data.
* bin/rivet-mkhtml: Adding a timestamp to the generated pages, and
a -t/--title option to allow setting the main HTML page title on
the command line: otherwise it becomes impossible to tell these
pages apart when you have a lot of them, except by URL!
2011-03-24 Andy Buckley <andy@insectnation.org>
* bin/aida2flat: Adding a -M option to *exclude* histograms whose
paths match a regex. Writing a negative lookahead regex with
positive matching was far too awkward!
2011-03-18 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Core/AnalysisHandler.cc (AnalysisHandler::removeAnalysis):
Fixed strange shared pointer assignment that caused seg-fault.
2011-03-13 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* filling of functions works now in a more intuitive way (I hope).
2011-03-09 Andy Buckley <andy@insectnation.org>
* Version 1.5.0 release!
2011-03-08 Andy Buckley <andy@insectnation.org>
* Adding some extra checks for external packages in make-plots.
2011-03-07 Andy Buckley <andy@insectnation.org>
* Changing the accuracy of the beam energy checking to 1%, to make
the UI a bit more forgiving. It's still best to specify exactly the right
energy of course!
2011-03-01 Andy Buckley <andy@insectnation.org>
* Adding --no-plottitle to compare-histos (+ completion).
* Fixing segfaults in UA1_1990_S2044935 and UA5_1982_S875503.
* Bump ABI version numbers for 1.5.0 release.
* Use AnalysisInfo for storage of the NeedsCrossSection analysis flag.
* Allow field setting in AnalysisInfo.
2011-02-27 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Support LineStyle=dashdotted in make-plots
* New command line option --style for compare-histos. Options are
"default", "bw" and "talk".
* cleaner uninstall
2011-02-26 Andy Buckley <andy@insectnation.org>
* Changing internal storage and return type of Particle::pdgId()
to PdgId, and adding Particle::energy().
* Renaming Analysis::energies() as Analysis::requiredEnergies().
* Adding beam energies into beam consistency checking:
Analysis::isCompatible methods now also require the beam energies
to be provided.
* Removing long-deprecated AnalysisHandler::init() constructor and
AnalysisHandler::removeIncompatibleAnalyses() methods.
2011-02-25 Andy Buckley <andy@insectnation.org>
* Adding --disable-obsolete, which takes its value from the value
of --disable-preliminary by default.
* Replacing RivetUnvalidated and RivetPreliminary plugin libraries
with optionally-configured analysis contents in the
experiment-specific plugin libraries. This avoids issues with
making libraries rebuild consistently when sources were reassigned
between libraries.
2011-02-24 Andy Buckley <andy@insectnation.org>
* Changing analysis plugin registration to fall back through
available paths rather than have RIVET_ANALYSIS_PATH totally
override the built-in paths. The first analysis hook of a given
name to be found is now the one that's used: any duplicates found
will be warned about but unused. getAnalysisLibPaths() now returns
*all* the search paths, in keeping with the new search behaviour.
2011-02-22 Andy Buckley <andy@insectnation.org>
* Moving the definition of the MSG_* macros into the Logging.hh
header. They can't be used everywhere, though, as they depend on
the existence of a this->getLog() method in the call scope. This
move makes them available in e.g. AnalysisHandler and other bits
of framework other than projections and analyses.
* Adding a gentle print-out from the Rivet AnalysisHandler if
preliminary analyses are being used, and strengthening the current
warning if unvalidated analyses are used.
* Adding documentation about the validation "process" and
the (un)validated and preliminary analysis statuses.
* Adding the new RivetPreliminary analysis library, and the
corresponding --disable-preliminary configure flag. Analyses in
this library are subject to change names, histograms, reference
data values, etc. between releases: make sure you check any
dependences on these analyses when upgrading Rivet.
* Change the Python script ref data search behaviours to use Rivet
ref data by default where available, rather than requiring a -R
option. Where relevant, -R is still a valid option, to avoid
breaking legacy scripts, and there is a new --no-rivet-refs option
to turn the default searching *off*.
* Add the prepending and appending optional arguments to the path
searching functions. This will make it easier to combine the
search functions with user-supplied paths in Python scripts.
* Make make-plots killable!
* Adding Rivet version to top of run printout.
* Adding Run::crossSection() and printing out the cross-section in
pb at the end of a Rivet run.
2011-02-22 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Make lighthisto.py aware of 2D histograms
* Adding published versions of the CDF_2008 leading jets and DY
analyses, and marking the preliminary ones as "OBSOLETE".
2011-02-21 Andy Buckley <andy@insectnation.org>
* Adding PDF documentation for path searching and .info/.plot
files, and tidying overfull lines.
* Removing unneeded const declarations from various return by
value path and internal binning functions. Should not affect ABI
compatibility but will force recompilation of external packages
using the RivetPaths.hh and Utils.hh headers.
* Adding findAnalysis*File(fname) functions, to be used by Rivet
scripts and external programs to find files known to Rivet
according to Rivet's (newly standard) lookup rule.
* Changing search path function behaviour to always return *all*
search directories rather than overriding the built-in locations
if the environment variables are set.
2011-02-20 Andy Buckley <andy@insectnation.org>
* Adding the ATLAS 2011 transverse jet shapes analysis.
2011-02-18 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Support for transparency in make-plots
2011-02-18 Frank Siegert <frank.siegert@cern.ch>
* Added ATLAS prompt photon analysis ATLAS_2010_S8914702
2011-02-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Simple NOOP constructor for Thrust projection
* Add CMS event shape analysis. Data read off the plots. We
will get the final numbers when the paper is accepted by
the journal.
2011-02-10 Frank Siegert <frank.siegert@cern.ch>
* Add final version of ATLAS dijet azimuthal decorrelation
2011-02-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* remove ATLAS conf note analyses for which we have final data
* reshuffle histograms in ATLAS minbias analysis to match Hepdata
* small pT-cut fix in ATLAS track based UE analysis
2011-01-31 Andy Buckley <andy@insectnation.org>
* Doc tweaks and adding cmp-by-|p| functions for Jets, to match
those added by Hendrik for Particles.
* Don't sum photons around muons in the D0 2010 Z pT analysis.
2011-01-27 Andy Buckley <andy@insectnation.org>
* Adding ATLAS 2010 min bias and underlying event analyses and data.
2011-01-23 Andy Buckley <andy@insectnation.org>
* Make make-plots write out PDF rather than PS by default.
2011-01-12 Andy Buckley <andy@insectnation.org>
* Fix several rendering and comparison bugs in rivet-mkhtml.
* Allow make-plots to write into an existing directory, at the
user's own risk.
* Make rivet-mkhtml produce PDF-based output rather than PS by
default (most people want PDF these days). Can we do the same
change of default for make-plots?
* Add getAnalysisPlotPaths() function, and use it in compare-histos
* Use proper .info file search path function internally in AnalysisInfo::make.
2011-01-11 Andy Buckley <andy@insectnation.org>
* Clean up ATLAS dijet analysis.
2010-12-30 Andy Buckley <andy@insectnation.org>
* Adding a run timeout option, and small bug-fixes to the event
timeout handling, and making first event timeout work nicely with
the run timeout. Run timeout is intended to be used in conjunction
with timed batch token expiry, of the type that likes to make 0
byte AIDA files on LCG when Grid proxies time out.
2010-12-21 Andy Buckley <andy@insectnation.org>
* Fix the cuts in the CDF 1994 colour coherence analysis.
2010-12-19 Andy Buckley <andy@insectnation.org>
* Fixing CDF midpoint cone jet algorithm default construction to
have an overlap threshold of 0.5 rather than 0.75. This was
recommended by the FastJet manual, and noticed while adding the
ATLAS and CMS cones.
* Adding ATLAS and CMS old iterative cones as "official" FastJets
constructor options (they could always have been used by explicit
instantiation and attachment of a Fastjet plugin object).
* Removing defunct and unused ClosestJetShape projection.
2010-12-16 Andy Buckley <andy@insectnation.org>
* bin/compare-histos, pyext/lighthisto.py: Take ref paths from
rivet module API rather than getting the environment by hand.
* pyext/lighthisto.py: Only read .plot info from the first
matching file (speed-up compare-histos).
2010-12-14 Andy Buckley <andy@insectnation.org>
* Augmenting the physics vector functionality to make FourMomentum
support maths operators with the correct return type (FourMomentum
rather than FourVector).
2010-12-11 Andy Buckley <andy@insectnation.org>
* Adding a --event-timeout option to control the event timeout,
adding it to the completion script, and making sure that the init
time check is turned OFF once successful!
* Adding an 3600 second timeout for initialising an event file. If
it takes longer than (or anywhere close to) this long, chances are
that the event source is inactive for some reason (perhaps
accidentally unspecified and stdin is not active, or the event
generator has died at the other end of the pipe. The reason for
not making it something shorter is that e.g. Herwig++ or Sherpa
can have long initialisation times to set up the MPI handler or to
run the matrix element integration. An timeout after an hour is
still better than a batch job which runs for two days before you
realise that you forgot to generate any events!
2010-12-10 Andy Buckley <andy@insectnation.org>
* Fixing unbooked-histo segfault in UA1_1990_S2044935 at 63 GeV.
2010-12-08 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixes in ATLAS_2010_CONF_083, declaring it validated
* Added ATLAS_2010_CONF_046, only two plots are implemented.
The paper will be out soon, and we don't need the other plots
right now. Data is read off the plots in the note.
* New option "SmoothLine" for HISTOGRAM sections in make-plots
* Changed CustomTicks to CustomMajorTicks and added CustomMinorTicks
in make-plots.
2010-12-07 Andy Buckley <andy@insectnation.org>
* Update the documentation to explain this latest bump to path
lookup behaviours.
* Various improvements to existing path lookups. In particular,
the analysis lib path locations are added to the info and ref
paths to avoid having to set three variables when you have all
three file types in the same personal plugin directory.
* Adding setAnalysisLibPaths and addAnalysisLibPath
functions. rivet --analysis-path{,-append} now use these and work
correctly. Hurrah!
* Add --show-analyses as an alias for --show-analysis, following a
comment at the ATLAS tutorial.
2010-12-07 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Change LegendXPos behaviour in make-plots. Now the top left
corner of the legend is used as anchor point.
2010-12-03 Andy Buckley <andy@insectnation.org>
* 1.4.0 release.
* Add bin-skipping to compare-histos to avoid one use of
rivet-rmgaps (it's still needed for non-plotting post-processing
like Professor).
2010-12-03 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix normalisation issues in UA5 and ALEPH analyses
2010-11-27 Andy Buckley <andy@insectnation.org>
* MathUtils.hh: Adding fuzzyGtrEquals and fuzzyLessEquals, and
tidying up the math utils collection a bit.
* CDF 1994 colour coherence analysis overhauled and
correction/norm factors fixed. Moved to VALIDATED status.
* Adding programmable completion for aida2flat and flat2aida.
* Improvements to programmable completion using the neat _filedir
completion shell function which I just discovered.
2010-11-26 Andy Buckley <andy@insectnation.org>
* Tweak to floating point inRange to use fuzzyEquals for CLOSED
interval equality comparisons.
* Some BibTeX generation improvements, and fixing the ATLAS dijet
BibTeX key.
* Resolution upgrade in PNG make-plots output.
* CDF_2005_S6217184.cc, CDF_2008_S7782535.cc: Updates to use the
new per-jet JetAlg interface (and some other fixes).
* JetAlg.cc: Changed the interface on request to return per-jet
rather than per-event jet shapes, with an extra jet index argument.
* MathUtils.hh: Adding index_between(...) function, which is handy
for working out which bin a value falls in, given a set of bin edges.
2010-11-25 Andy Buckley <andy@insectnation.org>
* Cmp.hh: Adding ASC/DESC (and ANTISORTED) as preferred
non-EQUIVALENT enum value synonyms over misleading
SORTED/UNSORTED.
* Change of rapidity scheme enum name to RapScheme
* Reworking JetShape a bit further: constructor args now avoid
inconsistencies (it was previously possible to define incompatible
range-ends and interval). Internal binning implementation also
reworked to use a vector of bin edges: the bin details are
available via the interface. The general jet pT cuts can be
applied via the JetShape constructor.
* MathUtils.hh: Adding linspace and logspace utility
functions. Useful for defining binnings.
* Adding more general cuts on jet pT and (pseudo)rapidity.
2010-11-11 Andy Buckley <andy@insectnation.org>
* Adding special handling of FourMomentum::mass() for computed
zero-mass vectors for which mass2 can go (very slightly) negative
due to numerical precision.
2010-11-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Adding ATLAS-CONF-2010-083 conference note. Data is read from plots.
When I run Pythia 6 the bins close to pi/2 are higher than in the note,
so I call this "unvalidated". But then ... the note doesn't specify
a tune or even just a version for the generators in the plots. Not even
if they used Pythia 6 or Pythia 8. Probably 6, since they mention AGILe.
2010-11-10 Andy Buckley <andy@insectnation.org>
* Adding a JetAlg::useInvisibles(bool) mechanism to allow ATLAS
jet studies to include neutrinos. Anyone who chooses to use this
mechanism had better be careful to remove hard neutrinos manually
in the provided FinalState object.
2010-11-09 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Adding ATLAS-CONF-2010-049 conference note. Data is read from plots.
Fragmentation functions look good, but I can't reproduce the MC lines
(or even the relative differences between them) in the jet cross-section
plots. So consider those unvalidated for now. Oh, and it seems ATLAS
screwed up the error bands in their ratio plots, too. They are
upside-down.
2010-11-07 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Adding ATLAS-CONF-2010-081 conference note. Data is read from plots.
2010-11-06 Andy Buckley <andy@insectnation.org>
* Deprecating the old JetShape projection and renaming to
ClosestJetShape: the algorithm has a tenuous relationship with
that actually used in the CDF (and ATLAS) jet shape analyses. CDF
analyses to be migrated to the new JetShape projection... and some
of that projection's features, design elements, etc. to be
finished off: we may as well take this opportunity to clear up
what was one of our nastiest pieces of code.
2010-11-05 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Adding ATLAS-CONF-2010-031 conference note. Data is read from plots.
2010-10-29 Andy Buckley <andy@insectnation.org>
* Making rivet-buildplugin use the same C++ compiler and CXXFLAGS
variable as used for the Rivet system build.
* Fixing NeutralFinalState projection to, erm, actually select
neutral particles (by Hendrik).
* Allow passing a general FinalState reference to the JetShape
projection, rather than requiring a VetoedFS.
2010-10-07 Andy Buckley <andy@insectnation.org>
* Adding a --with-root flag to rivet-buildplugin to add
root-config --libs flags to the plugin build command.
2010-09-24 Andy Buckley <andy@insectnation.org>
* Releasing as Rivet 1.3.0.
* Bundling underscore.sty to fix problems with running make-plots
on dat files generated by compare-histos from AIDA files with
underscores in their names.
2010-09-16 Andy Buckley <andy@insectnation.org>
* Fix error in N_effective definition for weighted profile errors.
2010-08-18 Andy Buckley <andy@insectnation.org>
* Adding MC_GENERIC analysis. NB. Frank Siegert also added MC_HJETS.
2010-08-03 Andy Buckley <andy@insectnation.org>
* Fixing compare-histos treatment of what is now a ref file, and
speeding things up... again. What a mess!
2010-08-02 Andy Buckley <andy@insectnation.org>
* Adding rivet-nopy: a super-simple Rivet C++ command line
interface which avoids Python to make profiling and debugging
easier.
* Adding graceful exception handling to the AnalysisHandler event
loop methods.
* Changing compare-histos behaviour to always show plots for which
there is at least one MC histo. The default behaviour should now
be the correct one in 99% of use cases.
2010-07-30 Andy Buckley <andy@insectnation.org>
* Merging in a fix for shared_ptrs not being compared for
insertion into a set based on raw pointer value.
2010-07-16 Andy Buckley <andy@insectnation.org>
* Adding an explicit library dependency declaration on libHepMC,
and hence removing the -lHepMC from the rivet-config --libs
output.
2010-07-14 Andy Buckley <andy@insectnation.org>
* Adding a manual section on use of Rivet (and AGILe) as
libraries, and how to use the -config scripts to aid compilation.
* FastJets projection now allows setting of a jet area definition,
plus a hacky mapping for getting the area-enabled cluster
sequence. Requested by Pavel Starovoitov & Paolo Francavilla.
* Lots of script updates in last two weeks!
2010-06-30 Andy Buckley <andy@insectnation.org>
* Minimising amount of Log class mapped into SWIG.
* Making Python ext build checks fail with error rather than
warning if it has been requested (or, rather, not explicitly
disabled).
2010-06-28 Andy Buckley <andy@insectnation.org>
* Converting rivet Python module to be a package, with the dlopen
flag setting etc. done around the SWIG generated core wrapper
module (rivet.rivetwrap).
* Requiring Python >= 2.4.0 in rivet scripts (and adding a Python
version checker function to rivet module)
* Adding --epspng option to make-plots (and converting to use subprocess.Popen).
2010-06-27 Andy Buckley <andy@insectnation.org>
* Converting JADE_OPAL analysis to use the fastjet
exclusive_ymerge_*max* function, rather than just
exclusive_ymerge: everything looks good now. It seems that fastjet
>= 2.4.2 is needed for this to work properly.
2010-06-24 Andy Buckley <andy@insectnation.org>
* Making rivet-buildplugin look in its own bin directory when
trying to find rivet-config.
2010-06-23 Andy Buckley <andy@insectnation.org>
* Adding protection and warning about numerical precision issues
in jet mass calculation/histogramming to the MC_JetAnalysis
analysis.
* Numerical precision improvement in calculation of
Vector4::mass2.
* Adding relative scale ratio plot flag to compare-histos
* Extended command completion to rivet-config, compare-histos, and
make-plots.
* Providing protected log messaging macros,
MSG_{TRACE,DEBUG,INFO,WARNING,ERROR} cf. Athena.
* Adding environment-aware functions for Rivet search path list access.
2010-06-21 Andy Buckley <andy@insectnation.org>
* Using .info file beam ID and energy info in HTML and LaTeX documentation.
* Using .info file beam ID and energy info in command-line printout.
* Fixing a couple of references to temporary variables in the
analysis beam info, which had been introduced during refactoring:
have reinstated reference-type returns as the more efficient
solution. This should not affect API compatibility.
* Making SWIG configure-time check include testing for
incompatibilities with the C++ compiler (re. the recurring _const_
char* literals issue).
* Various tweaks to scripts: make-plots and compare-histos
processes are now renamed (on Linux), rivet-config is avoided when
computing the Rivet version,and RIVET_REF_PATH is also set using
the rivet --analysis-path* flags. compare-histos now uses multiple
ref data paths for .aida file globbing.
* Hendrik changed VetoedFinalState comparison to always return
UNDEFINED if vetoing on the results of other FS projections is
being used. This is the only simple way to avoid problems
emanating from the remainingFinalState thing.
2010-06-19 Andy Buckley <andy@insectnation.org>
* Adding --analysis-path and --analysis-path-append command-line
flags to the rivet script, as a "persistent" way to set or extend
the RIVET_ANALYSIS_PATH variable.
* Changing -Q/-V script verbosity arguments to more standard
-q/-v, after Hendrik moaned about it ;)
* Small fix to TinyXML operator precendence: removes a warning,
and I think fixes a small bug.
* Adding plotinfo entries for new jet rapidity and jet mass plots
in MC_JetAnalysis derivatives.
* Moving MC_JetAnalysis base class into a new
libRivetAnalysisTools library, with analysis base class and helper
headers to be stored in the reinstated Rivet/Analyses include
directory.
2010-06-08 Andy Buckley <andy@insectnation.org>
* Removing check for CEDARSTD #define guard, since we no longer
compile against AGILe and don't have to be careful about
duplication.
* Moving crappy closest approach and decay significance functions
from Utils into SVertex, which is the only place they have ever
been used (and is itself almost entirely pointless).
* Overhauling particle ID <-> name system to clear up ambiguities
between enums, ints, particles and beams. There are no more enums,
although the names are still available as const static ints, and
names are now obtained via a singleton class which wraps an STL
map for name/ID lookups in both directions.
2010-05-18 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing factor-of-2 bug in the error calculation when scaling
histograms.
* Fixing D0_2001_S4674421 analysis.
2010-05-11 Andy Buckley <andy@insectnation.org>
* Replacing TotalVisibleMomentum with MissingMomentum in analyses
and WFinder. Using vector ET rather than scalar ET in some places.
2010-05-07 Andy Buckley <andy@insectnation.org>
* Revamping the AnalysisHandler constructor and data writing, with
some LWH/AIDA mangling to bypass the stupid AIDA idea of having to
specify the sole output file and format when making the data
tree. Preferred AnalysisHandler constructor now takes only one arg
-- the runname -- and there is a new AH.writeData(outfile) method
to replace AH.commitData(). Doing this now to begin migration to
more flexible histogramming in the long term.
2010-04-21 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing LaTeX problems (make-plots) on ancient machines, like lxplus.
2010-04-29 Andy Buckley <andy@insectnation.org>
* Fixing (I hope!) the treatment of weighted profile bin errors in LWH.
2010-04-21 Andy Buckley <andy@insectnation.org>
* Removing defunct and unused KtJets and Configuration classes.
* Hiding some internal details from Doxygen.
* Add @brief Doxygen comments to all analyses, projections and
core classes which were missing them.
2010-04-21 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* remove obsolete reference to InitialQuarks from DELPHI_2002
* fix normalisation in CDF_2000_S4155203
2010-04-20 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* bin/make-plots: real support for 2-dim histograms plotted as
colormaps, updated the documentation accordingly.
* fix misleading help comment in configure.ac
2010-04-08 Andy Buckley <andy@insectnation.org>
* bin/root2flat: Adding this little helper script, minimally
modified from one which Holger Schulz made for internal use in
ATLAS.
2010-04-05 Andy Buckley <andy@insectnation.org>
* Using spiresbib in rivet-mkanalysis: analysis templates made
with rivet-mkanalysis will now contain a SPIRES-dumped BibTeX key
and entry if possible!
* Adding BibKey and BibTeX entries to analysis metadata files, and
updating doc build to use them rather than the time-consuming
SPIRES screen-scraping. Added SPIRES BibTeX dumps to all analysis
metadata using new (uninstalled & unpackaged) doc/get-spires-data
script hack.
* Updating metadata files to add Energies, Beams and PtCuts
entries to all of them.
* Adding ToDo, NeedsCrossSection, and better treatment of Beams
and Energies entries in metadata files and in AnalysisInfo and
Analysis interfaces.
2010-04-03 Andy Buckley <andy@insectnation.org>
* Frank Siegert: Update of rivet-mkhtml to conform to improved
compare-histos.
* Frank Siegert: LWH output in precision-8 scientific notation, to
solve a binning precision problem... the first time weve noticed a
problem!
* Improved treatment of data/reference datasets and labels in
compare-histos.
* Rewrite of rivet-mkanalysis in Python to make way for neat
additions.
* Improving SWIG tests, since once again the user's biuld system
must include SWIG (no test to check that it's a 'good SWIG', since
the meaning of that depends on which compiler is being used and we
hope that the user system is consistent... evidence from Finkified
Macs and bloody SLC5 notwithstanding).
2010-03-23 Andy Buckley <andy@insectnation.org>
* Tag as patch release 1.2.1.
2010-03-22 Andy Buckley <andy@insectnation.org>
* General tidying of return arguments and intentionally unused
parameters to keep -Wextra happy (some complaints remain from
TinyXML, FastJet, and HepMC).
* Some extra bug fixes: in FastJets projection with explicit
plugin argument, removing muon veto cut on FoxWolframMoments.
* Adding UNUSED macro to help with places where compiler warnings
can't be helped.
* Turning on -Wextra warnings, and fixing some violations.
2010-03-21 Andy Buckley <andy@insectnation.org>
* Adding MissingMomentum projection, as replacement for ~all uses
of now-deprecated TotalVisibleMomentum projection.
* Fixing bug with TotalVisibleMomentum projection usage in MC_SUSY
analysis.
* Frank Siegert fixed major bug in pTmin param passing to FastJets
projection. D'oh: requires patch release.
2010-03-02 Andy Buckley <andy@insectnation.org>
* Tagging for 1.2.0 release... at last!
2010-03-01 Andy Buckley <andy@insectnation.org>
* Updates to manual, manual generation scripts, analysis info etc.
* Add HepData URL to metadata print-out with rivet --show-analysis
* Fix average Et plot in UA1 analysis to only apply to the tracker
acceptance (but to include neutral particle contributions,
i.e. the region of the calorimeter in the tracker acceptance).
* Use Et rather than pT in filling the scalar Et measure in
TotalVisibleMomentum.
* Fixes to UA1 normalisation (which is rather funny in the paper).
2010-02-26 Andy Buckley <andy@insectnation.org>
* Update WFinder to not place cuts and other restrictions on the
neutrino.
2010-02-11 Andy Buckley <andy@insectnation.org>
* Change analysis loader behaviour to use ONLY RIVET_ANALYSIS_PATH
locations if set, otherwise use ONLY the standard Rivet analysis
install path. Should only impact users of personal plugin
analyses, who should now explicitly set RIVET_ANALYSIS_PATH to
load their analysis... and who can now create personal versions of
standard analyses without getting an error message about duplicate
loading.
2010-01-15 Andy Buckley <andy@insectnation.org>
* Add tests for "stable" heavy flavour hadrons in jets (rather
than just testing for c/b hadrons in the ancestor lists of stable
jet constituents)
2009-12-23 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New option "RatioPlotMode=deviation" in make-plots.
2009-12-14 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New option "MainPlot" in make-plots. For people who only want
the ratio plot and nothing else.
* New option "ConnectGaps" in make-plots. Set to 1 if you
want to connect gaps in histograms with a line when ErrorBars=0.
Works both in PLOT and in HISTOGRAM sections.
* Eliminated global variables for coordinates in make-plots and
enabled multithreading.
2009-12-14 Andy Buckley <andy@insectnation.org>
* AnalysisHandler::execute now calls AnalysisHandler::init(event)
if it has not yet been initialised.
* Adding more beam configuration features to Beam and
AnalysisHandler: the setRunBeams(...) methods on the latter now
allows a beam configuration for the run to be specified without
using the Run class.
2009-12-11 Andy Buckley <andy@insectnation.org>
* Removing use of PVertex from few remaining analyses. Still used
by SVertex, which is itself hardly used and could maybe be
removed...
2009-12-10 Andy Buckley <andy@insectnation.org>
* Updating JADE_OPAL to do the histo booking in init(), since
sqrtS() is now available at that stage.
* Renaming and slightly re-engineering all MC_*_* analyses to not
be collider-specific (now the Analysis::sqrtS/beams()) methods
mean that histograms can be dynamically binned.
* Creating RivetUnvalidated.so plugin library for unvalidated
analyses. Unvalidated analyses now need to be explicitly enabled
with a --enable-unvalidated flag on the configure script.
* Various min bias analyses updated and validated.
2009-12-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Propagate SPECIAL and HISTOGRAM sections from .plot files
through compare-histos
* STAR_2006_S6860818: <pT> vs particle mass, validate analysis
2009-12-04 Andy Buckley <andy@insectnation.org>
* Use scaling rather than normalising in DELPHI_1996: this is
generally desirable, since normalizing to 1 for 1/sig dsig/dx
observables isn't correct if any events fall outside the histo
bounds.
* Many fixes to OPAL_2004.
* Improved Hemispheres interface to remove unnecessary consts on
returned doubles, and to also return non-squared versions
of (scaled) hemisphere masses.
* Add "make pyclean" make target at the top level to make it
easier for developers to clean their Python module build when the
API is extended.
* Identify use of unvalidated analyses with a warning message at
runtime.
* Providing Analysis::sqrtS() and Analysis::beams(), and making
sure they're available by the time the init methods are called.
2009-12-02 Andy Buckley <andy@insectnation.org>
* Adding passing of first event sqrt(s) and beams to analysis handler.
* Restructuring running to only use one HepMC input file (no-one
was using multiple ones, right?) and to break down the Run class
to cleanly separate the init and event loop phases. End of file is
now neater.
2009-12-01 Andy Buckley <andy@insectnation.org>
* Adding parsing of beam types and pairs of energies from YAML.
2009-12-01 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing trigger efficiency in CDF_2009_S8233977
2009-11-30 Andy Buckley <andy@insectnation.org>
* Using shared pointers to make I/O object memory management
neater and less error-prone.
* Adding crossSectionPerEvent() method [==
crossSection()/sumOfWeights()] to Analysis. Useful for histogram
scaling since numerator of sumW_passed/sumW_total (to calculate
pass-cuts xsec) is cancelled by dividing histo by sumW_passed.
* Clean-up of Particle class and provision of inline PID::
functions which take a Particle as an argument to avoid having to
explicitly call the Particle::pdgId() method.
2009-11-30 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing division by zero in Profile1D bin errors for
bins with just a single entry.
2009-11-24 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* First working version of STAR_2006_S6860818
2009-11-23 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Adding missing CDF_2001_S4751469 plots to uemerge
* New "ShowZero" option in make-plots
* Improving lots of plot defaults
* Fixing typos / non-existing bins in CDF_2001_S4751469 and
CDF_2004_S5839831 reference data
2009-11-19 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing our compare() for doubles.
2009-11-17 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Zeroth version of STAR_2006_S6860818 analysis (identified
strange particles). Not working yet for unstable particles.
2009-11-11 Andy Buckley <andy@insectnation.org>
* Adding separate jet-oriented and photon-oriented observables to
MC PHOTONJETUE analysis.
* Bug fix in MC leading jets analysis, and general tidying of
leading jet analyses to insert units, etc. (should not affect any
current results)
2009-11-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing last issues in STAR_2006_S6500200 and setting it to
VALIDATED.
* Noramlise STAR_2006_S6870392 to cross-section
2009-11-09 Andy Buckley <andy@insectnation.org>
* Overhaul of jet caching and ParticleBase interface.
* Adding lists of analyses' histograms (obtained by scanning the
plot info files) to the LaTeX documentation.
2009-11-07 Andy Buckley <andy@insectnation.org>
* Adding checking system to ensure that Projections aren't
registered before the init phase of analyses.
* Now that the ProjHandler isn't full of defunct pointers (which
tend to coincidentally point to *new* Projection pointers rather
than undefined memory, hence it wasn't noticed until recently!),
use of a duplicate projection name is now banned with a helpful
message at runtime.
* (Huge) overhaul of ProjectionHandler system to use shared_ptr:
projections are now deleted much more efficiently, naturally
cleaning themselves out of the central repository as they go out
of scope.
2009-11-06 Andy Buckley <andy@insectnation.org>
* Adding Cmp<double> specialisation, using fuzzyEquals().
2009-11-05 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing histogram division code.
2009-11-04 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New analysis STAR_2006_S6500200 (pion and proton pT spectra in
pp collisions at 200 GeV). It is still unclear if they used a cut
in rapidity or pseudorapidity, thus the analysis is declared
"UNDER DEVELOPMENT" and "DO NOT USE".
* Fixing compare() in NeutralFinalState and MergedFinalState
2009-11-04 Andy Buckley <andy@insectnation.org>
* Adding consistence checking on beam ID and sqrt(s) vs. those
from first event.
2009-11-03 Andy Buckley <andy@insectnation.org>
* Adding more assertion checks to linear algebra testing.
2009-11-02 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixing normalisation issue with stacked histograms in
make-plots.
2009-10-30 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* CDF_2009_S8233977: Updating data and axes labels to match final
paper. Normalise to cross-section instead of data.
2009-10-23 Andy Buckley <andy@insectnation.org>
* Fixing Cheese-3 plot in CDF 2004... at last!
2009-10-23 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix muon veto in CDF_1994_S2952106, CDF_2005_S6217184,
CDF_2008_S7782535, and D0_2004_S5992206
2009-10-19 Andy Buckley <andy@insectnation.org>
* Adding analysis info files for MC SUSY and PHOTONJETUE analyses.
* Adding MC UE analysis in photon+jet events.
2009-10-19 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Adding new NeutralFinalState projection. Note that this final
state takes E_T instead of p_T as argument (makes more sense for
neutral particles). The compare() method does not yet work as
expected (E_T comparison still missing).
* Adding new MergedFinalState projection. This merges two final
states, removing duplicate particles. Duplicates are identified by
looking at the genParticle(), so users need to take care of any
manually added particles themselves.
* Fixing most open issues with the STAR_2009_UE_HELEN analysis.
There is only one question left, regarding the away region.
* Set the default split-merge value for SISCone in our FastJets
projection to the recommended (but not Fastjet-default!) value of
0.75.
2009-10-17 Andy Buckley <andy@insectnation.org>
* Adding parsing of units in cross-sections passed to the "-x"
flag, i.e. "-x 101 mb" is parsed internally into 1.01e11 pb.
2009-10-16 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Disabling DELPHI_2003_WUD_03_11 in the Makefiles, since I don't
trust the data.
* Getting STAR_2009_UE_HELEN to work.
2009-10-04 Andy Buckley <andy@insectnation.org>
* Adding triggers and other tidying to (still unvalidated)
UA1_1990 analysis.
* Fixing definition of UA5 trigger to not be intrinscally
different for pp and ppbar: this is corrected for (although it
takes some readng to work this out) in the 1982 paper, which I
think is the only one to compare the two modes.
* Moving projection setup and registration into init() method for
remaining analyses.
* Adding trigger implementations as projections for CDF Runs 0 &
1, and for UA5.
2009-10-01 Andy Buckley <andy.buckley@cern.ch>
* Moving projection setup and registration into init() method for
analyses from ALEPH, CDF and the MC_ group.
* Adding generic SUSY validation analysis, based on plots used in
ATLAS Herwig++ validation.
* Adding sorted particle accessors to FinalState (cf. JetAlg).
2009-09-29 Andy Buckley <andy@insectnation.org>
* Adding optional use of args as regex match expressions with
-l/--list-analyses.
2009-09-03 Andy Buckley <andy.buckley@cern.ch>
* Passing GSL include path to compiler, since its absence was
breaking builds on systems with no GSL installation in a standard
location (such as SLC5, for some mysterious reason!)
* Removing lib extension passing to compiler from the configure
script, because Macs and Linux now both use .so extensions for the
plugin analysis modules.
2009-09-02 Andy Buckley <andy@insectnation.org>
* Adding analysis info file path search with RIVET_DATA_PATH
variable (and using this to fix doc build.)
* Improvements to AnalysisLoader path search.
* Moving analysis sources back into single directory, after a
proletarian uprising ;)
2009-09-01 Andy Buckley <andy@insectnation.org>
* Adding WFinder and WAnalysis, based on Z proj and analysis, with
some tidying of the Z code.
* ClusteredPhotons now uses an IdentifiedFS to pick the photons to
be looped over, and only clusters photons around *charged* signal
particles.
2009-08-31 Andy Buckley <andy@insectnation.org>
* Splitting analyses by directory, to make it easier to disable
building of particular analysis group plugin libs.
* Removing/merging headers for all analyses except for the special
MC_JetAnalysis base class.
* Exit with an error message if addProjection is used twice from
the same parent with distinct projections.
2009-08-28 Andy Buckley <andy@insectnation.org>
* Changed naming convention for analysis plugin libraries, since
the loader has changed so much: they must now *start* with the
word "Rivet" (i.e. no lib prefix).
* Split standard plugin analyses into several plugin libraries:
these will eventually move into separate subdirs for extra build
convenience.
* Started merging analysis headers into the source files, now that
we can (the plugin hooks previously forbade this).
* Replacement of analysis loader system with a new one based on
ideas from ThePEG, which uses dlopen-time instantiation of
templated global variables to reduce boilerplate plugin hooks to
one line in analyses.
2009-07-14 Frank Siegert <frank.siegert@durham.ac.uk>
* Replacing in-source histo-booking metadata with .plot files.
2009-07-14 Andy Buckley <andy@insectnation.org>
* Making Python wrapper files copy into place based on bundled
versions for each active HepMC interface (2.3, 2.4 & 2.5), using a
new HepMC version detector test in configure.
* Adding YAML metadata files and parser, removing same metadata
from the analysis classes' source headers.
2009-07-07 Andy Buckley <andy@insectnation.org>
* Adding Jet::hadronicEnergy()
* Adding VisibleFinalState and automatically using it in JetAlg
projections.
* Adding YAML parser for new metadata (and eventually ref data)
files.
2009-07-02 Andy Buckley <andy@insectnation.org>
* Adding Jet::neutralEnergy() (and Jet::totalEnergy() for
convenience/symmetry).
2009-06-25 Andy Buckley <andy@insectnation.org>
* Tidying and small efficiency improvements in CDF_2008_S7541902
W+jets analysis (remove unneeded second stage of jet storing,
sorting the jets twice, using foreach, etc.).
2009-06-24 Andy Buckley <andy@insectnation.org>
* Fixing Jet's containsBottom and containsCharm methods, since B
hadrons are not necessarily to be found in the final
state. Discovered at the same time that HepMC::GenParticle defines
a massively unhelpful copy constructor that actually loses the
tree information; it would be better to hide it entirely!
* Adding RivetHepMC.hh, which defines container-type accessors to
HepMC particles and vertices, making it possible to use Boost
foreach and hence avoiding the usual huge boilerplate for-loops.
2009-06-11 Andy Buckley <andy@insectnation.org>
* Adding --disable-pdfmanual option, to make the bootstrap a bit
more robust.
* Re-enabling D0IL in FastJets: adding 10^-10 to the pTmin removes
the numerical instability!
* Fixing CDF_2004 min/max cone analysis to use calo jets for the
leading jet Et binning. Thanks to Markus Warsinsky
for (re)discovering this bug: I was sure it had been fixed. I'm
optimistic that this will fix the main distributions, although
Swiss Cheese "minus 3" is still likely to be broken. Early tests
look okay, but it'll take more stats before we can remove the "do
not trust" sign.
2009-06-10 Andy Buckley <andy@insectnation.org>
* Providing "calc" methods so that Thrust and Sphericity
projections can be used as calculators without having to use the
projecting/caching system.
2009-06-09 Andy Buckley <andy@insectnation.org>
* 1.1.3 release!
* More doc building and SWIG robustness tweaks.
2009-06-07 Andy Buckley <andy@insectnation.org>
* Make doc build from metadata work even before the library is
installed.
2009-06-07 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix phi rotation in CDF_2008_LEADINGJETS.
2009-06-07 Andy Buckley <andy@insectnation.org>
* Disabling D0 IL midpoint cone (using CDF modpoint instead),
since there seems to be a crashing bug in FastJet's
implementation: we can't release that way, since ~no D0 analyses
will run.
2009-06-03 Andy Buckley <andy@insectnation.org>
* Putting SWIG-generated source files under SVN control to make
life easier for people who we advise to check out the SVN head
version, but who don't have a sufficiently modern copy of SWIG to
* Adding the --disable-analyses option, for people who just want
to use Rivet as a framework for their own analyses.
* Enabling HepMC cross-section reading, now that HepMC 2.5.0 has
been released.
2009-05-23 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Using gsl-config to locate libgsl
* Fix the paths for linking such that our own libraries are found
before any system libraries, e.g. for the case that there is an
outdated fastjet version installed on the system while we want to
use our own up-to-date version.
* Change dmerge to ymerge in the e+e- analyses using JADE or
DURHAM from fastjet. That's what it is called in fastjet-2.4 now.
2009-05-18 Andy Buckley <andy@insectnation.org>
* Adding use of gsl-config in configure script.
2009-05-16 Andy Buckley <andy@insectnation.org>
* Removing argument to vetoEvent macro, since no weight
subtraction is now needed. It's now just an annotated return, with
built-in debug log message.
* Adding an "open" FinalState, which is only calculated once per
even, then used by all other FSes, avoiding the loop over
non-status 1 particles.
2009-05-15 Andy Buckley <andy@insectnation.org>
* Removing incorrect setting of DPS x-errs in CDF_2008 jet shape
analysis: the DPS autobooking already gets this bit right.
* Using Jet rather than FastJet::PseudoJet where possible, as it
means that the phi ranges match up nicely between Particle and the
Jet object. The FastJet objects are only really needed if you want
to do detailed things like look at split/merge scales for
e.g. diff jet rates or "y-split" analyses.
* Tidying and debugging CDF jet shape analyses and jet shape
plugin... ongoing, but I think I've found at least one real bug,
plus a lot of stuff that can be done a lot more nicely.
* Fully removing deprecated math functions and updating affected
analyses.
2009-05-14 Andy Buckley <andy@insectnation.org>
* Removing redundant rotation in DISKinematics... this was a
legacy of Peter using theta rather than pi-theta in his rotation.
* Adding convenience phi, rho, eta, theta, and perp,perp2 methods
to the 3 and 4 vector classes.
2009-05-12 Andy Buckley <andy@insectnation.org>
* Adding event auto-rotation for events with one proton... more
complete approach?
2009-05-09 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Renaming CDF_2008_NOTE_9337 to CDF_2009_S8233977.
* Numerous small bug fixes in ALEPH_1996_S3486095.
* Adding data for one of the Rick-Field-style STAR UE analyses.
2009-05-08 Andy Buckley <andy@insectnation.org>
* Adding rivet-mkanalysis script, to make generating new analysis
source templates easier.
2009-05-07 Andy Buckley <andy@insectnation.org>
* Adding null vector check to Vector3::azimuthalAngle().
* Fixing definition of HCM/Breit frames in DISKinematics, and
adding asserts to check that the transformation is doing what it
should.
2009-05-05 Andy Buckley <andy@insectnation.org>
* Removing eta and Et cuts from CDF 2000 Z pT analysis, based on
our reading of the paper, and converting most of the analysis to a
call of the ZFinder projection.
2009-05-05 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Support non-default seed_threshold in CDF cone jet algorithms.
* New analyses STAR_2006_S6870392 and STAR_2008_S7993412. In
STAR_2008_S7993412 only the first distribution is filled at the
moment. STAR_2006_S6870392 is normalised to data instead of the
Monte Carlo cross-section, since we don't have that available in
the HepMC stream yet.
2009-05-05 Andy Buckley <andy@insectnation.org>
* Changing Event wrapper to copy whole GenEvents rather than
pointers, use std units if supported in HepMC, and run a
placeholder function for event auto-orientation.
2009-04-28 Andy Buckley <andy@insectnation.org>
* Removing inclusion of IsolationTools header by analyses that
aren't actually using the isolation tools... which is all of
them. Leaving the isolation tools in place for now, as there might
still be use cases for them and there's quite a lot of code there
that deserves a second chance to be used!
2009-04-24 Andy Buckley <andy@insectnation.org>
* Deleting Rivet implementations of TrackJet and D0ILConeJets: the
code from these has now been incorporated into FastJet 2.4.0.
* Removed all mentions of the FastJet JADE patch and the HAVE_JADE
preprocessor macro.
* Bug fix to D0_2008_S6879055 to ensure that cuts compare to both
electron and positron momenta (was just comparing against
electrons, twice, probably thanks to the miracle of cut and
paste).
* Converting all D0 IL Cone jets to use FastJets. Involved tidying
D0_2004 jet azimuthal decorrelation analysis and D0_2008_S6879055
as part of migration away from using the getLorentzJets method,
and removing the D0ILConeJets header from quite a few analyses
that weren't using it at all.
* Updating CDF 2001 to use FastJets in place of TrackJet, and
adding axis labels to its plots.
* Note that ZEUS_2001_S4815815 uses the wrong jet definition: it
should be a cone but curently uses kT.
* Fixing CDF_2005_S6217184 to use correct (midpoint, R=0.7) jet
definition. That this was using a kT definition with R=1.0 was
only made obvious when the default FastJets constructor was
removed.
* Removing FastJets default constructor: since there are now
several good (IRC safe) jet definitions available, there is no
obvious safe default and analyses should have to specify which
they use.
* Moving FastJets constructors into implementation file to reduce
recompilation dependencies, and adding new plugins.
* Ensuring that axis labels actually get written to the output
data file.
2009-04-22 Andy Buckley <andy@insectnation.org>
* Adding explicit FastJet CDF jet alg overlap_threshold
constructor param values, since the default value from 2.3.x is
now removed in version 2.4.0.
* Removing use of HepMC ThreeVector::mag method (in one place
only) since this has been removed in version 2.5.0b.
* Fix to hepmc.i (included by rivet.i) to ignore new HepMC 2.5.0b
GenEvent stream operator.
2009-04-21 Andy Buckley <andy@insectnation.org>
* Dependency on FastJet now requires version 2.4.0 or later. Jade
algorithm is now native.
* Moving all analysis constructors and Projection headers from the
analysis header files into their .cc implementation files, cutting
header dependencies.
* Removing AIDA headers: now using LWH headers only, with
enhancement to use axis labels. This facility is now used by the
histo booking routines, and calling the booking function versions
which don't specify axis labels will result in a runtime warning.
2009-04-07 Andy Buckley <andy@insectnation.org>
* Adding $(DESTDIR) prefix to call to Python module "setup.py
install"
* Moving HepMC SWIG mappings into Python Rivet module for now:
seems to work-around the SL type-mapping bug.
2009-04-03 Andy Buckley <andy@insectnation.org>
* Adding MC analysis for LHC UE: higher-pT replica of Tevatron
2008 leading jets study.
* Adding CDF_1990 pseudorapidity analysis.
* Moving CDF_2001 constructor into implementation file.
* Cleaning up CDF_2008_LEADINGJETS a bit, e.g. using foreach
loops.
* Adding function interface for specifying axis labels in histo
bookings. Currently has no effect, since AIDA doesn't seem to have
a mechanism for axis labels. It really is a piece of crap.
2009-03-18 Andy Buckley <andy@insectnation.org>
* Adding docs "make upload" and other tweaks to make the doc files
fit in with the Rivet website.
* Improving LaTex docs to show email addresses in printable form
and to group analyses by collider or other metadata.
* Adding doc script to include run info in LaTeX docs, and to make
HTML docs.
* Removing WZandh projection, which wasn't generator independent
and whose sole usage was already replaced by ZFinder.
* Improvements to constructors of ZFinder and InvMassFS.
* Changing ExampleTree to use real FS-based Z finding.
2009-03-16 Andy Buckley <andy@insectnation.org>
* Allow the -H histo file spec to give a full name if wanted. If
it doesn't end in the desired extension, it will be added.
* Adding --runname option (and API elements) to choose a run name
to be prepended as a "top level directory" in histo paths. An
empty value results in no extra TLD.
2009-03-06 Andy Buckley <andy@insectnation.org>
* Adding R=0.2 photon clustering to the electrons in the CDF 2000
Z pT analysis.
2009-03-04 Andy Buckley <andy@insectnation.org>
* Fixing use of fastjet-config to not use the user's PATH
variable.
* Fixing SWIG type table for HepMC object interchange.
2009-02-20 Andy Buckley <andy@insectnation.org>
* Adding use of new metadata in command line analysis querying
with the rivet command, and in building the PDF Rivet manual.
* Adding extended metadata methods to the Analysis interface and
the Python wrapper. All standard analyses comply with this new
interface.
2009-02-19 Andy Buckley <andy@insectnation.org>
* Adding usefully-scoped config headers, a Rivet::version()
function which uses them, and installing the generated headers to
fix "external" builds against an installed copy of Rivet. The
version() function has been added to the Python wrapper.
2009-02-05 Andy Buckley <andy@insectnation.org>
* Removing ROOT dependency and linking. Woo! There's no need for
this now, because the front-end accepts no histo format switch and
we'll just use aida2root for output conversions. Simpler this way,
and it avoids about half of our compilation bug reports from 32/64
bit ROOT build confusions.
2009-02-04 Andy Buckley <andy@insectnation.org>
* Adding automatic generation of LaTeX manual entries for the
standard analyses.
2009-01-20 Andy Buckley <andy@insectnation.org>
* Removing RivetGun and TCLAP source files!
2009-01-19 Andy Buckley <andy@insectnation.org>
* Added psyco Python optimiser to rivet, make-plots and
compare-histos.
* bin/aida2root: Added "-" -> "_" mangling, following requests.
2009-01-17 Andy Buckley <andy@insectnation.org>
* 1.1.2 release.
2009-01-15 Andy Buckley <andy@insectnation.org>
* Converting Python build system to bundle SWIG output in tarball.
2009-01-14 Andy Buckley <andy@insectnation.org>
* Converting AIDA/LWH divide function to return a DPS so that bin
width factors don't get all screwed up. Analyses adapted to use
the new division operation (a DPS/DPS divide would also be
nice... but can wait for YODA).
2009-01-06 Andy Buckley <andy@insectnation.org>
* bin/make-plots: Added --png option for making PNG output files,
using 'convert' (after making a PDF --- it's a bit messy)
* bin/make-plots: Added --eps option for output filtering through
ps2eps.
2009-01-05 Andy Buckley <andy@insectnation.org>
* Python: reworking Python extension build to use distutils and
newer m4 Python macros. Probably breaks distcheck but is otherwise
more robust and platform independent (i.e. it should now work on
Macs).
2008-12-19 Andy Buckley <andy@insectnation.org>
* make-plots: Multi-threaded make-plots and cleaned up the LaTeX
building a bit (necessary to remove the implicit single global
state).
2008-12-18 Andy Buckley <andy@insectnation.org>
* make-plots: Made LaTeX run in no-stop mode.
* compare-histos: Updated to use a nicer labelling syntax on the
command line and to successfully build MC-MC plots.
2008-12-16 Andy Buckley <andy@insectnation.org>
* Made LWH bin edge comparisons safe against numerical errors.
* Added Particle comparison functions for sorting.
* Removing most bad things from ExampleTree and tidying up. Marked
WZandh projection for removal.
2008-12-03 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added the two missing observables to the CDF_2008_NOTE_9337 analysis,
i.e. track pT and sum(ET). There is a small difference between our MC
output and the MC plots of the analysis' author, we're still waiting
for the author's comments.
2008-12-02 Andy Buckley <andy@insectnation.org>
* Overloading use of a std::set in the interface, since the
version of SWIG on Sci Linux doesn't have a predefined mapping for
STL sets.
2008-12-02 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixed uemerge. The output was seriously broken by a single line
of debug information in fillAbove(). Also changed uemerge output
to exponential notation.
* Unified ref and mc histos in compare-histos. Histos with one
bin are plotted linear. Option for disabling the ratio plot.
Several fixes for labels, legends, output directories, ...
* Changed rivetgun's fallback directory for parameter files to
$PREFIX/share/AGILe, since that's where the steering files now are.
* Running aida2flat in split mode now produces make-plots compatible
dat-files for direct plotting.
2008-11-28 Andy Buckley <andy@insectnation.org>
* Replaced binreloc with an upgraded and symbol-independent copy.
2008-11-25 Andy Buckley <andy@insectnation.org>
* Added searching of $RIVET_REF_PATH for AIDA reference data
files.
2008-11-24 Andy Buckley <andy@insectnation.org>
* Removing "get"s and other obsfucated syntax from
ProjectionApplier (Projection and Analysis) interfaces.
2008-11-21 Andy Buckley <andy@insectnation.org>
* Using new "global" Jet and V4 sorting functors in
TrackJet. Looks like there was a sorting direction problem before...
* Verbose mode with --list-analyses now shows descriptions as well
as analysis names.
* Moved data/Rivet to data/refdata and moved data/RivetGun
contents to AGILe (since generator steering is no longer a Rivet
thing)
* Added unchecked ratio plots to D0 Run II jet + photon analysis.
* Added D0 inclusive photon analysis.
* Added D0 Z rapidity analysis.
* Tidied up constructor interface and projection chain
implementation of InvMassFinalState.
* Added ~complete set of Jet and FourMomentum sorting functors.
2008-11-20 Andy Buckley <andy@insectnation.org>
* Added IdentifiedFinalState.
* Moved a lot of TrackJet and Jet code into .cc files.
* Fixed a caching bug in Jet: cache flag resets should never be
conditional, since they are then sensitive to initialisation
errors.
* Added quark enum values to ParticleName.
* Rationalised JetAlg interfaces somewhat, with "size()" and
"jets()" methods in the interface.
* Added D0 W charge asymmetry and D0 inclusive jets analyses.
2008-11-18 Andy Buckley <andy@insectnation.org>
* Adding D0 inclusive Z pT shape analysis.
* Added D0 Z + jet pT and photon + jet pT spectrum analyses.
* Lots of tidying up of particle, event, particle name etc.
* Now the first event is used to detect the beam type and remove
incompatible analyses.
2008-11-17 Andy Buckley <andy@insectnation.org>
* Added bash completion for rivetgun.
* Starting to provide stand-alone call methods on projections so
they can be used without the caching infrastructure. This could
also be handy for unit testing.
* Adding functionality (sorting function and built-in sorting
schemes) to the JetAlg interface.
2008-11-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fix floating point number output format in aida2flat and flat2aida
* Added CDF_2002_S4796047: CDF Run-I charged multiplicity distribution
* Renamed CDF_2008_MINBIAS to CDF_2008_NOTE_9337, since the
note is publicly available now.
2008-11-10 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added DELPHI_2003_WUD_03_11: Delphi 4-jet angular distributions.
There is still a problem with the analysis, so don't use it yet.
But I need to commit the code because my laptop is broken ...
2008-11-06 Andy Buckley <andy@insectnation.org>
* Code review: lots of tidying up of projections and analyses.
* Fixes for compatibility with the LLVM C & C++ compiler.
* Change of Particle interface to remove "get"-prefixed method
names.
2008-11-05 Andy Buckley <andy@insectnation.org>
* Adding ability to query analysis metadata from the command line.
* Example of a plugin analyis now in plugindemo, with a make check
test to make sure that the plugin analysis is recognised by the
command line "rivet" tool.
* GCC 4.3 fix to mat-vec tests.
2008-11-04 Andy Buckley <andy@insectnation.org>
* Adding native logger control from Python interface.
2008-11-03 Andy Buckley <andy@insectnation.org>
* Adding bash_completion for rivet executable.
2008-10-31 Andy Buckley <andy@insectnation.org>
* Clean-up of histo titles and analysis code review.
* Added momentum construction functions from FastJet PseudoJets.
2008-10-28 Andy Buckley <andy@insectnation.org>
* Auto-booking of histograms with a name, rather than the HepData
ID 3-tuple is now possible.
* Fix in CDF 2001 pT spectra to get the normalisations to depend
on the pT_lead cutoff.
2008-10-23 Andy Buckley <andy@insectnation.org>
* rivet handles signals neatly, as for rivetgun, so that premature
killing of the analysis process will still result in an analysis
file.
* rivet now accepts cross-section as a command line argument and,
if it is missing and is required, will prompt the user for it
interactively.
2008-10-22 Andy Buckley <andy@insectnation.org>
* rivet (Python interface) now can list analyses, check when
adding analyses that the given names are valid, specify histo file
name, and provide sensibly graded event number logging.
2008-10-20 Andy Buckley <andy@insectnation.org>
* Corrections to CDF 2004 analysis based on correspondance with
Joey Huston. M bias dbns now use whole event within |eta| < 0.7,
and Cheese plots aren't filled at all if there are insufficient
jets (and the correct ETlead is used).
2008-10-08 Andy Buckley <andy@insectnation.org>
* Added AnalysisHandler::commitData() method, to allow the Python
interface to write out a histo file without having to know
anything about the histogramming API.
* Reduced SWIG interface file to just map a subset of Analysis and
AnalysisHandler functionality. This will be the basis for a new
command line interface.
2008-10-06 Andy Buckley <andy@insectnation.org>
* Converted FastJets plugin to use a Boost shared_pointer to the
cached ClusterSequence. The nullness of the pointer is now used to
indicate an empty tracks (and hence jets) set. Once FastJet
natively support empty CSeqs, we can rewrite this a bit more
neatly and ditch the shared_ptr.
2008-10-02 Andy Buckley <andy@insectnation.org>
* The CDF_2004 (Acosta) data file now includes the full range of
pT for the min bias data at both 630 and 1800 GeV. Previously,
only the small low-pT insert plot had been entered into HepData.
2008-09-30 Andy Buckley <andy@insectnation.org>
* Lots of updates to CDF_2004 (Acosta) UE analysis, including
sorting jets by E rather than Et, and factorising transverse cone
code into a function so that it can be called with a random
"leading jet" in min bias mode. Min bias histos are now being
trial-filled just with tracks in the transverse cones, since the
paper is very unclear on this.
* Discovered a serious caching problem in FastJets projection when
an empty tracks vector is passed from the
FinalState. Unfortunately, FastJet provides no API way to solve
the problem, so we'll have to report this upstream. For now, we're
solving this for CDF_2004 by explicitly vetoing events with no
tracks.
* Added Doxygen to the build with target "dox"
* Moved detection of whether cross-section information is needed
into the AnalysisHandler, with dynamic computation by scanning
contained analyses.
* Improved robustness of event reading to detect properly when the
input file is smaller than expected.
2008-09-29 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New analysis: CDF_2000_S4155203
2008-09-23 Andy Buckley <andy@insectnation.org>
* rivetgun can now be built and run without AGILe. Based on a
patch by Frank Siegert.
2008-09-23 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Some preliminary numbers for the CDF_2008_LEADINGJETS analysis
(only transverse region and not all observables. But all we have now.)
2008-09-17 Andy Buckley <andy@insectnation.org>
* Breaking up the mammoth "generate" function, to make Python
mapping easier, among other reasons.
* Added if-zero-return-zero checking to angle mapping functions,
to avoid problems where 1e-19 gets mapped on to 2 pi and then
fails boundary asserts.
* Added HistoHandler singleton class, which will be a central
repository for holding analyses' histogram objects to be accessed
via a user-chosen name.
2008-08-26 Andy Buckley <andy@insectnation.org>
* Allowing rivet-config to return combined flags.
2008-08-14 Andy Buckley <andy@insectnation.org>
* Fixed some g++ 4.3 compilation bugs, including "vector" not
being a valid name for a method which returns a physics vector,
since it clashes with std::vector (which we globally import). Took
the opportunity to rationalise the Jet interface a bit, since
"particle" was used to mean "FourMomentum", and "Particle" types
required a call to "getFullParticle". I removed the "gets" at the
same time, as part of our gradual migration to a coherent naming
policy.
2008-08-11 Andy Buckley <andy@insectnation.org>
* Tidying of FastJets and added new data files from HepData.
2008-08-10 James Monk <jmonk@hep.ucl.ac.uk>
* FastJets now uses user_index property of fastjet::PseudoJet to
reconstruct PID information in jet contents.
2008-08-07 Andy Buckley <andy@insectnation.org>
* Reworking of param file and command line parsing. Tab characters
are now handled by the parser, in a way equivalent to spaces.
2008-08-06 Andy Buckley <andy@insectnation.org>
* Added extra histos and filling to Acosta analysis - all HepData
histos should now be filled, depending on sqrt{s}. Also trialling
use of LaTeX math mode in titles.
2008-08-05 Andy Buckley <andy@insectnation.org>
* More data files for CDF analyses (2 x 2008, 1 x 1994), and moved
the RivetGun AtlasPythia6.params file to more standard
fpythia-atlas.params (and added to the install list).
2008-08-04 Andy Buckley <andy@insectnation.org>
* Reduced size of available options blocks in RivetGun help text
by removing "~" negating variants (which are hardly ever used in
practice) and restricting beam particles to
PROTON, ANTIPROTON,ELECTRON and POSITRON.
* Fixed Et sorting in Acosta analysis.
2008-08-01 Andy Buckley <andy@insectnation.org>
* Added AIDA headers to the install list, since
external (plugin-type) analyses need them to be present for
compilation to succeed.
2008-07-29 Andy Buckley <andy@insectnation.org>
* Fixed missing ROOT compile flags for libRivet.
* Added command line repetition to logging.
2008-07-29 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Included the missing numbers and three more observables
in the CDF_2008_NOTE_9351 analysis.
2008-07-29 Andy Buckley <andy@insectnation.org>
* Fixed wrong flags on rivet-config
2008-07-28 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Renamed CDF_2008_DRELLYAN to CDF_2008_NOTE_9351. Updated
numbers and cuts to the final version of this CDF note.
2008-07-28 Andy Buckley <andy@insectation.org>
* Fixed polar angle calcuation to use atan2.
* Added "mk" prefixes and x/setX convention to math classes.
2008-07-28 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Fixed definition of FourMomentum::pT (it had been returning pT2)
2008-07-27 Andy Buckley <andy@insectnation.org>
* Added better tests for Boost headers.
* Added testing for -ansi, -pedantic and -Wall compiler flags.
2008-07-25 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* updated DELPHI_2002_069_CONF_603 according to information
from the author
2008-07-17 Andy Buckley <andy@insectnation.org>
* Improvements to aida2flat: now can produce one output file per
histo, and there is a -g "gnuplot mode" which comments out the
YODA/make_plot headers to make the format readable by gnuplot.
* Import boost::assign namespace contents into the Rivet namespace
--- provides very useful intuitive collection initialising
functions.
2008-07-15 Andy Buckley <andy.buckley@dur.ac.uk>
* Fixed missing namespace in vector/matrix testing.
* Removed Boost headers: now a system dependency.
* Fixed polarRadius infinite loop.
2008-07-09 Andy Buckley <andy@insectnation.org>
* Fixed definitions of mapAngleMPiToPi, etc. and used them to fix
the Jet::getPhi method.
* Trialling use of "foreach" loop in CDF_2004: it works! Very nice.
2008-07-08 Andy Buckley <andy@insectnation.org>
* Removed accidental reference to an "FS" projection in
FinalStateHCM's compare method. rivetgun -A now works again.
* Added TASSO, SLD and D0_2008 reference data. The TASSO and SLD
papers aren't installed or included in the tarball since there are
currently no plans to implement these analyses.
* Added Rivet namespacing to vector, matrix etc. classes. This
required some re-writing and the opportunity was taken to move
some canonical function definitions inside the classes and to
improve the header structure of the Math area.
2008-07-07 Andy Buckley <andy@insectnation.org>
* Added Rivet namespace to Units.hh and Constants.hh.
* Added Doxygen "@brief" flags to analyses.
* Added "RIVET_" namespacing to all header guards.
* Merged Giulio Lenzi's isolation/vetoing/invmass projections and
D0 2008 analysis.
2008-06-23 Jon Butterworth <J.Butterworth@ucl.ac.uk>
* Modified FastJet to fix ysplit and split and filter.
* Modified ExampleTree to show how to call them.
2008-06-19 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added first version of the CDF_2008_DRELLYAN analysis described on
http://www-cdf.fnal.gov/physics/new/qcd/abstracts/UEinDY_08.html
There is a small difference between the analysis and this
implementation, but it's too small to be visible.
The fpythia-cdfdrellyan.params parameter file is for this analysis.
* Added first version of the CDF_2008_MINBIAS analysis described on
http://www-cdf.fnal.gov/physics/new/qcd/abstracts/minbias_08.html
The .aida file is read from the plots on the web and will change.
I'm still discussing some open questions about the analysis with
the author.
2008-06-18 Jon Butterworth <J.Butterworth@ucl.ac.uk>
* Added First versions of splitJet and filterJet methods to
fastjet.cc. Not yet tested, buyer beware.
2008-06-18 Andy Buckley <andy@insectnation.org>
* Added extra sorted Jets and Pseudojets methods to FastJets, and
added ptmin argument to the JetAlg getJets() method, requiring a
change to TrackJet.
2008-06-13 Andy Buckley <andy@insectnation.org>
* Fixed processing of "RG" parameters to ensure that invalid
iterators are never used.
2008-06-10 Andy Buckley <andy@insectnation.org>
* Updated AIDA reference files, changing "/HepData" root path to
"/REF". Still missing a couple of reference files due to upstream
problems with the HepData records.
2008-06-09 Andy Buckley <andy@insectnation.org>
* rivetgun now handles termination signals (SIGTERM, SIGINT and
SIGHUP) gracefully, finishing the event loop and finalising
histograms. This means that histograms will always get written
out, even if not all the requested events have been generated.
2008-06-04 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added DELPHI_2002_069_CONF_603 analysis
2008-05-30 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added InitialQuarks projection
* Added OPAL_1998_S3780481 analysis
2008-05-29 Andy Buckley <andy@insectnation.org>
* distcheck compatibility fixes and autotools tweaks.
2008-05-28 Andy Buckley <andy@insectnation.org>
* Converted FastJet to use Boost smart_ptr for its plugin
handling, to solve double-delete errors stemming from the heap
cloning of projections.
* Added (a subset of) Boost headers, particularly the smart
pointers.
2008-05-24 Andy Buckley <andy@insectnation.org>
* Added autopackage spec files.
* Merged these changes into the trunk.
* Added a registerClonedProjection(...) method to
ProjectionHandler: this is needed so that cloned projections will
have valid pointer entries in the ProjectHandler repository.
* Added clone() methods to all projections (need to use this,
since the templated "new PROJ(proj)" approach to cloning can't
handle object polymorphism.
2008-05-19 Andy Buckley <andy@insectnation.org>
* Moved projection-applying functions into ProjectionApplier base
class (from which Projection and Analysis both derive).
* Added Rivet-specific exceptions in place of std::runtime_error.
* Removed unused HepML reference files.
* Added error handling for requested analyses with wrong case
convention / missing name.
2008-05-15 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New analysis PDG_Hadron_Multiplicities
* flat2aida converter
2008-05-15 Andy Buckley <andy@insectnation.org>
* Removed unused mysterious Perl scripts!
* Added RivetGun.HepMC logging of HepMC event details.
2008-05-14 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New analysis DELPHI_1995_S3137023. This analysis contains
the xp spectra of Xi+- and Sigma(1385)+-.
2008-05-13 Andy Buckley <andy@insectnation.org>
* Improved logging interface: log levels are now integers (for
cross-library compatibility and level setting also applies to
existing loggers.
2008-05-09 Andy Buckley <andy@insectnation.org>
* Improvements to robustness of ROOT checks.
* Added --version flag on config scripts and rivetgun.
2008-05-06 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* New UnstableFinalState projection which selects all hadrons,
leptons and real photons including unstable particles.
* In the DELPHI_1996_S3430090 analysis the multiplicities for
pi+/pi- and p0 are filled, using the UnstableFinalState projection.
2008-05-06 Andy Buckley <andy@insectnation.org>
* FastJets projection now protects against the case where no
particles exist in the final state (where FastJet throws an
exception).
* AIDA file writing is now separated from the
AnalysisHandler::finalize method... API users can choose what to
do with the histo objects, be that writing out or further
processing.
2008-04-29 Andy Buckley <andy@insectnation.org>
* Increased default tolerances in floating point comparisons as
they were overly stringent and valid f.p. precision errors were
being treated as significant.
* Implemented remainder of Acosta UE analysis.
* Added proper getEtSum() to Jet.
* Added Et2() member and function to FourMomentum.
* Added aida2flat conversion script.
* Fixed ambiguity in TrackJet algorithm as to how the iteration
continues when tracks are merged into jets in the inner loop.
2008-04-28 Andy Buckley <andy@insectnation.org>
* Merged in major "ProjectionHandler" branch. Projections are now
all stored centrally in the singleton ProjectionHandler container,
rather than as member pointers in projections and analyses. This
also affects the applyProjection mechanism, which is now available
as a templated method on Analysis and Projection. Still a few
wrinkles need to be worked out.
* The branch changes required a comprehensive review of all
existing projections and analyses: lots of tidying up of these
classes, as well as the auxiliary code like math utils, has taken
place. Too much to list and track, unfortunately!
2008-03-28 Andy Buckley <buckley@pc54.hep.ucl.ac.uk>
* Started second CDF UE analysis ("Acosta"): histograms defined.
* Fixed anomalous factor of 2 in LWH conversion from Profile1D
to DataPointSet.
* Added pT distribution histos to CDF 2001 UE analysis.
2008-03-26 Andy Buckley <andy@insectnation.org>
* Removed charged+neutral versions of histograms and projections
from DELPHI analysis since they just duplicate the more robust
charged-only measurements and aren't really of interest for
tuning.
2008-03-10 Andy Buckley <andy@insectnation.org>
* Profile histograms now use error computation with proper
weighting, as described here:
http://en.wikipedia.org/wiki/Weighted_average
2008-02-28 Andy Buckley <andy@insectnation.org>
* Added --enable-jade flag for Professor studies with patched
FastJet.
* Minor fixes to LCG tag generator and gfilt m4 macros.
* Fixed projection slicing issues with Field UE analysis.
* Added Analysis::vetoEvent(e) function, which keeps track of the
correction to the sum of weights due to event vetoing in analysis
classes.
2008-02-26 Andy Buckley <andy@insectnation.org>
* Vector<N> and derived classes now initialise to have zeroed
components when the no-arg constructor is used.
* Added Analysis::scale() function to scale 1D
histograms. Analysis::normalize() uses it internally, and the
DELPHI (A)EEC, whose histo weights are not pure event weights, and
normalised using scale(h, 1/sumEventWeights).
2008-02-21 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Added EEC and AEEC to the DELPHI_1996_S3430090 analysis. The
normalisation of these histograms is still broken (ticket #163).
2008-02-19 Hendrik Hoeth <hendrik.hoeth@cern.ch>
* Many fixes to the DELPHI_1996_S3430090 analysis: bugfix in the
calulation of eigenvalues/eigenvectors in MatrixDiag.hh for the
sphericity, rewrite of Thrust/Major/Minor, fixed scaled momentum,
hemisphere masses, normalisation in single particle events,
final state slicing problems in the projections for Thrust,
Sphericity and Hemispheres.
2008-02-08 Andy Buckley <andy@insectnation.org>
* Applied fixes and extensions to DIS classes, based on
submissions by Dan Traynor.
2008-02-06 Andy Buckley <andy@insectnation.org>
* Made projection pointers used for cut combining into const
pointers. Required some redefinition of the Projection* comparison
operator.
* Temporarily added FinalState member to ChargedFinalState to stop
projection lifetime crash.
2008-02-01 Andy Buckley <andy@insectnation.org>
* Fixed another misplaced factor of bin width in the
Analysis::normalize() method.
2008-01-30 Andy Buckley <andy@insectnation.org>
* Fixed the conversion of IHistogram1D to DPS, both via the
explicit Analysis::normalize() method and the implicit
AnalysisHandler::treeNormalize() route. The root of the problem is
the AIDA choice of the word "height" to represent the sum of
weights in a bin: i.e. the bin width is not taken into account
either in computing bin height or error.
2008-01-22 Andy Buckley <andy@insectnation.org>
* Beam projection now uses HepMC GenEvent::beam_particles() method
to get the beam particles. This is more portable and robust for
C++ generators, and equivalent to the existing "first two" method
for Fortran generators.
2008-01-17 Andy Buckley <andy@insectnation.org>
* Added angle range fix to pseudorapidity function (thanks to
Piergiulio Lenzi).
2008-01-10 Andy Buckley <andy@insectnation.org>
* Changed autobooking plot codes to use zero-padding (gets the
order right in JAS, file browser, ROOT etc.). Also changed the
'ds' part to 'd' for consistency. HepData's AIDA output has been
correspondingly updated, as have the bundled data files.
2008-01-04 Andy Buckley <andy@insectnation.org>
* Tidied up JetShape projection a bit, including making the
constructor params const references. This seems to have sorted the
runtime segfault in the CDF_2005 analysis.
* Added caching of the analysis bin edges from the AIDA file -
each analysis object will now only read its reference file once,
which massively speeds up the rivetgun startup time for analyses
with large numbhers of autobooked histos (e.g. the
DELPHI_1996_S3430090 analysis).
2008-01-02 Andy Buckley <andy@insectnation.org>
* CDF_2001_S4751469 now uses the LossyFinalState projection, with
an 8% loss rate.
* Added LossyFinalState and HadronicFinalState, and fixed a
"polarity" bug in the charged final state projection (it was
keeping only the *uncharged* particles).
* Now using isatty(1) to determine whether or not color escapes
can be used. Also removed --color argument, since it can't have an
effect (TCLAP doesn't do position-based flag toggling).
* Made Python extension build optional (and disabled by default).
2008-01-01 Andy Buckley <andy@insectnation.org>
* Removed some unwanted DEBUG statements, and lowered the level of
some infrastructure DEBUGs to TRACE level.
* Added bash color escapes to the logger system.
2007-12-21 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* include/LWH/ManagedObject.h: Fixed infinite loop in
encodeForXML cf. ticket #135.
2007-12-20 Andy Buckley <andy@insectnation.org>
* Removed HepPID, HepPDT and Boost dependencies.
* Fixed XML entity encoding in LWH. Updated CDF_2007_S7057202
analysis to not do its own XML encoding of titles.
2007-12-19 Andy Buckley <andy@insectnation.org>
* Changed units header to set GeV = 1 (HepMC convention) and using
units in CDF UE analysis.
2007-12-15 Andy Buckley <andy@insectnation.org>
* Introduced analysis metadata methods for all analyses (and made
them part of the Analysis interface).
2007-12-11 Andy Buckley <andy@insectnation.org>
* Added JetAlg base projection for TrackJet, FastJet etc.
2007-12-06 Andy Buckley <andy@insectnation.org>
* Added checking for Boost library, and the standard Boost test
program for shared_ptr.
* Got basic Python interface running - required some tweaking
since Python and Rivet's uses of dlopen collide (another
RTLD_GLOBAL issue - see
http://muttley.hates-software.com/2006/01/25/c37456e6.html )
2007-12-05 Andy Buckley <andy@insectnation.org>
* Replaced all use of KtJets projection with FastJets
projection. KtJets projection disabled but left undeleted for
now. CLHEP and KtJet libraries removed from configure searches and
Makefile flags.
2007-12-04 Andy Buckley <andy@insectnation.org>
* Param file loading now falls back to the share/RivetGun
directory if a local file can't be found and the provided name has
no directory separators in it.
* Converted TrackJet projection to update the jet centroid with
each particle added, using pT weighting in the eta and phi
averaging.
2007-12-03 Andy Buckley <andy@insectnation.org>
* Merged all command line handling functions into one large parse
function, since only one executable now needs them. This removes a
few awkward memory leaks.
* Removed rivet executable - HepMC file reading functionality will
move into rivetgun.
* Now using HepMC IO_GenEvent format (IO_Ascii and
IO_ExtendedAscii are deprecated). Now requires HepMC >= 2.3.0.
* Added forward declarations of GSL diagonalisation routines,
eliminating need for GSL headers to be installed on build machine.
2007-11-27 Andy Buckley <andy@insectnation.org>
* Removed charge differentiation from Multiplicity projection (use
CFS proj) and updated ExampleAnalysis to produce more useful numbers.
* Introduced binreloc for runtime path determination.
* Fixed several bugs in FinalState, ChargedFinalState, TrackJet
and Field analysis.
* Completed move to new analysis naming scheme.
2007-11-26 Andy Buckley <andy@insectnation.org>
* Removed conditional HAVE_FASTJET bits: FastJet is now compulsory.
* Merging appropriate RivetGun parts into Rivet. RivetGun currently broken.
2007-11-23 Andy Buckley <andy@insectnation.org>
* Renaming analyses to Spires-ID scheme: currently of form
S<SpiresID>, to become <Expt>_<YYYY>_<SpiresID>.
2007-11-20 Andy Buckley <andy@insectnation.org>
* Merged replacement vectors, matrices and boosts into trunk.
2007-11-15 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Analysis.cc, include/Rivet/Analysis.hh: Introduced normalize
function. See ticket #126.
2007-10-31 Andy Buckley <andy@insectnation.org>
* Tagging as 1.0b2 for HERA-LHC meeting.
2007-10-25 Andy Buckley <andy@insectnation.org>
* Added AxesDefinition base interface to Sphericity and Thrust,
used by Hemispheres.
* Exposed BinaryCut class, improved its interface and fixed a few
bugs. It's now used by VetoedFinalState for momentum cuts.
* Removed extra output from autobooking AIDA reader.
* Added automatic DPS booking.
2007-10-12 Andy Buckley <andy@insectnation.org>
* Improved a few features of the build system
2007-10-09 James Monk
* Fixed dylib dlopen on Mac OS X.
2007-10-05 Andy Buckley <andy@insectnation.org>
* Added new reference files.
2007-10-03 Andy Buckley <andy@insectnation.org>
* Fixed bug in configure.ac which led to explicit CXX setting
being ignored.
* Including Logging.hh in Projection.hh, hence new transitive
dependency on Logging.hh being installed. Since this is the normal
behaviour, I don't think this is a problem.
* Fixed segfaulting bug due to use of addProjection() in
locally-scoped contained projections. This isn't a proper fix,
since the whole framework should be designed to avoid the
possibility of bugs like this.
* Added newly built HepML and AIDA reference files for current
analyses.
2007-10-02 Andy Buckley <andy@insectnation.org>
* Fixed possible null-pointer dereference in Particle copy
constructor and copy assignment: this removes one of two blocker
segfaults, the other of which is related to the copy-assignment of
the TotalVisMomentum projection in the ExampleTree analysis.
2007-10-01 Andy Buckley <andy@insectnation.org>
* Fixed portable path to Rivet share directory.
2007-09-28 Andy Buckley <andy@insectnation.org>
* Added more functionality to the rivet-config script: now has
libdir, includedir, cppflags, ldflags and ldlibs options.
2007-09-26 Andy Buckley <andy@insectnation.org>
* Added the analysis library closer function to the
AnalysisHandler finalize() method, and also moved the analysis
delete loop into AnalysisHandler::finalize() so as not to try
deleting objects whose libraries have already closed.
* Replaced the RivetPaths.cc.in method for portable paths with
something using -D defines - much simpler!
2007-09-21 Lars Sonnenschein <sonne@mail.cern.ch>
* Added HepEx0505013 analysis and JetShape projection (some fixes
by AB.)
* Added GetLorentzJets member function to D0 RunII cone jet projection
2007-09-21 Andy Buckley <andy@insectnation.org>
* Fixed lots if bugs and bad practice in HepEx0505013 (to make it
compile-able!)
* Downclassed the log messages from the Test analysis to DEBUG
level.
* Added isEmpty() method to final state projection.
* Added testing for empty final state and useful debug log
messages to sphericity projection.
2007-09-20 Andy Buckley <andy@insectnation.org>
* Added Hemispheres projection, which calculates event hemisphere
masses and broadenings.
2007-09-19 Andy Buckley <andy@insectnation.org>
* Added an explicit copy assignment operator to Particle: the
absence of one of these was responsible for the double-delete
error.
* Added a "fuzzy equals" utility function for float/double types
to Utils.hh (which already contains a variety of handy little
functions).
* Removed deprecated Beam::operator().
* Added ChargedFinalState projection and de-pointered the
contained FinalState projection in VetoedFinalState.
2007-09-18 Andy Buckley <andy@insectnation.org>
* Major bug fixes to the regularised version of the sphericity
projection (and hence the Parisi tensor projection). Don't trust
C & D param results from any previous version!
* Added extra methods to thrust and sphericity projections to get
the oblateness and the sphericity basis (currently returns dummy
axes since I can't yet work out how to get the similarity
transform eigenvectors from CLHEP)
2007-09-14 Andy Buckley <andy@insectnation.org>
* Merged in a branch of pluggable analysis mechanisms.
2007-06-25 Jon Butterworth <jmb@hep.ucl.ac.uk>
* Fixed some bugs in the root output for DataPoint.h
2007-06-25 Andy Buckley <andy@insectnation.org>
* include/Rivet/**/Makefile.am: No longer installing headers for
"internal" functionality.
* include/Rivet/Projections/*.hh: Removed the private restrictions
on copy-assignment operators.
2007-06-18 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* include/LWH/Tree.h: Fixed minor bug in listObjectNames.
* include/LWH/DataPointSet.h: Fixed setCoordinate functions so
that they resize the vector of DataPoints if it initially was
empty.
* include/LWH/DataPoint.h: Added constructor taking a vector of
measuremts.
2007-06-16 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* include/LWH/Tree.h: Implemented the listObjectNames and ls
functions.
* include/Rivet/Projections/FinalStateHCM.hh,
include/Rivet/Projections/VetoedFinalState.hh: removed
_theParticles and corresponding access function. Use base class
variable instead.
* include/Rivet/Projections/FinalState.hh: Made _theParticles
protected.
2007-06-13 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Projections/FinalStateHCM.cc,
src/Projections/DISKinematics.cc: Equality checks using
GenParticle::operator== changed to check for pointer equality.
* include/Rivet/Analysis/HepEx9506012.hh: Uses modified DISLepton
projection.
* include/Rivet/Particle.hh: Added member function to check if a
GenParticle is associated.
* include/Rivet/Projections/DISLepton.hh,
src/Projections/DISLepton.cc: Fixed bug in projection. Introduced
final state projection to limit searching for scattered
lepton. Still not properly tested.
2007-06-08 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* include/Rivet/Projections/PVertex.hh,
src/Projections/PVertex.cc: Fixed the projection to simply get the
signal_process_vertex from the GenEvent. This is the way it should
work. If the GenEvent does not have a signal_process_vertex
properly set up in this way, the problem is with the class that
fills the GenEvent.
2007-06-06 Jon Butterworth <jmb@hep.ucl.ac.uk>
* Merged TotalVisibleMomentum and CalMET
* Added pT ranges to Vetoed final state projection
2007-05-27 Jon Butterworth <jmb@hep.ucl.ac.uk>
* Fixed initialization of VetoedFinalStateProjection in ExampleTree
2007-05-27 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* include/Rivet/Projections/KtJets.*: Make sure the KtEvent is
deleted properly.
2007-05-26 Jon Butterworth <jmb@hep.ucl.ac.uk>
* Added leptons to the ExampleTree.
* Added TotalVisibleEnergy projection, and added output to ExampleTree.
2007-05-25 Jon Butterworth <jmb@hep.ucl.ac.uk>
* Added a charged lepton projection
2007-05-23 Andy Buckley <andy@insectnation.org>
* src/Analysis/HepEx0409040.cc: Changed range of the histograms to
the "pi" range rather than the "128" range.
* src/Analysis/Analysis.cc: Fixed a bug in the AIDA path building.
Histogram auto-booking now works.
2007-05-23 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Analysis/HepEx9506012.cc: Now uses the histogram booking
function in the Analysis class.
2007-05-23 Jon Butterworth <jmb@hep.ucl.ac.uk>
* Fixed bug in PRD65092002 (was failing on zero jets)
2007-05-23 Andy Buckley <andy@insectnation.org>
* Added (but haven't properly tested) a VetoedFinalState projection.
* Added normalize() method for AIDA 1D histograms.
* Added configure checking for Mac OS X version, and setting the
development target flag accordingly.
2007-05-22 Andy Buckley <andy@insectnation.org>
* Added an ostream method for AnalysisName enums.
* Converted Analyses and Projections to use projection lists, cuts
and beam constraints.
* Added beam pair combining to the BeamPair sets of Projections
by finding set meta-intersections.
* Added methods to Cuts, Analysis and Projection to make Cut
definition easier.
* Fixed default fall-through in cut handling switch statement and
now using -numeric_limits<double>::max() rather than min()
* Added more control of logging presentation via static flag
methods on Log.
2007-05-13 Andy Buckley <andy@insectnation.org>
* Added self-consistency checking mechanisms for Cuts and Beam
* Re-implemented the cut-handling part of RivetInfo as a Cuts class.
* Changed names of Analysis and Projection name() and handler()
methods to getName() and getHandler() to be more consistent with
the rest of the public method names in those classes.
2007-05-02 Andy Buckley <andy@insectnation.org>
* Added auto-booking of histogram bins from AIDA XML files. The
AIDA files are located via a C++ function which is generated from
RivetPaths.cc.in by running configure.
2007-04-18 Andy Buckley <andy@insectnation.org>
* Added a preliminary version of the Rick Field UE analysis, under
the name PRD65092002.
2007-04-19 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Analysis/HepEx0409040.cc: The reason this did not compile
under gcc-4 is that some iterators into a vector were wrongly
assued to be pointers and were initialized to 0 and later compared
to 0. I've changed this to initialize to end() of the
corresponding vector and to compare with the same end() later.
2007-04-05 Andy Buckley <andy@insectnation.org>
* Lots of name changes in anticipation of the MCNet
school. RivetHandler is now AnalysisHandler (since that's what it
does!), BeamParticle has become ParticleName, and RivetInfo has
been split into Cut and BeamConstraint portions.
* Added BeamConstraint mechanism, which can be used to determine
if an analysis is compatible with the beams being used in the
generator. The ParticleName includes an "ANY" wildcard for this
purpose.
2006-03-19 Andy Buckley <andy@insectnation.org>
* Added "rivet" executable which can read in HepMC ASCII dump
files and apply Rivet analyses on the events.
2007-02-24 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Projections/KtJets.cc: Added comparison of member variables
in compare() function
* all: Merged changes from polymorphic-projections branch into
trunk
2007-02-17 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* all: projections and analysis handlers: All projections which
uses other projctions now has a pointer rather than a copy of
those projections to allow for polymorphism. The constructors has
also been changed to require the used projections themselves,
rather than the arguments needed to construct them.
2007-02-17 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Projections/FinalState.cc,
include/Rivet/Projections/FinalState.icc (Rivet),
include/Rivet/Projections/FinalState.hh: Added cut in transverse
momentum on the particles to be included in the final state.
2007-02-06 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* include/LWH/HistogramFactory.h: Fixed divide-by-zero in divide
function. Also fixed bug in error calculation in divide
function. Introduced checkBin function to make sure two histograms
are equal even if they have variable bin widths.
* include/LWH/Histogram1D.h: In normalize(double), do not do anything
if the sum of the bins are zero to avoid dividing by zero.
2007-01-20 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* src/Test/testLWH.cc: Modified to output files using the Tree.
* configure.ac: Removed AC_CONFIG_AUX_DIR([include/Rivet/Config])
since the directory does not exist anymore.
2006-12-21 Andy Buckley <andy@insectnation.org>
* Rivet will now conditionally install the AIDA and LWH headers if
it can't find them when configure'ing.
* Started integrating Leif's LWH package to fulfill the AIDA
duties.
* Replaced multitude of CLHEP wrapper headers with a single
RivetCLHEP.h header.
2006-11-20 Andy Buckley <andy@insectnation.org>
* Introduced log4cpp logging.
* Added analysis enum, which can be used as input to an analysis
factory by Rivet users.
2006-11-02 Andy Buckley <andy@insectnation.org>
* Yet more, almost pointless, administrative moving around of
things with the intention of making the structure a bit
better-defined:
* The RivetInfo and RivetHandler classes have been
moved from src/Analysis into src as they are really the main Rivet
interface classes. The Rivet.h header has also been moved into the
"header root".
* The build of a single shared library in lib has been disabled,
with the library being built instead in src.
2006-10-14 Andy Buckley <andy@insectnation.org>
* Introduced a minimal subset of the Sherpa math tools, such as
Vector{3,4}D, Matrix, etc. The intention is to eventually cut the
dependency on CLHEP.
2006-07-28 Andy Buckley <andy@insectnation.org>
* Moving things around: all sources now in directories under src
2006-06-04 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* Analysis/Examples/HZ95108.*: Now uses CentralEtHCM. Also set GeV
units on the relevant histograms.
* Projections/CentralEtHCM.*: Making a special class just to get
out one number - the summed Et in the central rapidity bin - may
seem like an overkill. But in case some one else might nees it...
2006-06-03 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* Analysis/Examples/HZ95108.*: Added the hz95108 energy flow
analysis from HZtool.
* Projections/DISLepton.*: Since many HERA measurements do not
care if we have electron or positron beam, it is now possible to
specify lepton or anti-lepton.
* Projections/Event.*: Added member and access function for the
weight of an event (taken from the GenEvent object.weights()[0].
* Analysis/RivetHandler.*: Now depends explicitly on the AIDA
interface. An AIDA analysis factory must be specified in the
constructor, where a tree and histogram factory is automatically
created. Added access functions to the relevant AIDA objects.
* Analysis/AnalysisBase.*: Added access to the RivetHandler and
its AIDA factories.
2005-12-27 Leif Lönnblad <Leif.Lonnblad@thep.lu.se>
* configure.ac: Added -I$THEPEGPATH/include to AM_CPPFLAGS.
* Config/Rivet.h: Added some std incudes and using std::
declaration.
* Analysis/RivetInfo.*: Fixed some bugs. The RivetInfo facility
now works, although it has not been thoroughly tested.
* Analysis/Examples/TestMultiplicity.*: Re-introduced
FinalStateHCM for testing purposes but commented it away again.
* .: Made a number of changes to implement handling of RivetInfo
objects.
diff --git a/bin/rivet b/bin/rivet
--- a/bin/rivet
+++ b/bin/rivet
@@ -1,587 +1,578 @@
#! /usr/bin/env python
"""\
Run Rivet analyses on inputted events from file or Unix pipe
Examples:
%prog [options] <hepmcfile> [<hepmcfile2> ...]
my_generator -o myfifo & \ %prog [options] myfifo
agile-runmc <genname> -n 100k -o- | %prog [options]
ENVIRONMENT:
* RIVET_ANALYSIS_PATH: list of paths to be searched for plugin
analysis libraries at runtime
* RIVET_REF_PATH: list of paths to be searched for reference
data files
* RIVET_INFO_PATH: list of paths to be searched for analysis
metadata files
"""
## Try to bootstrap the Python path for rivet loading
# TODO: Move this intorivet.__init__.py?
import os, sys
PROGPATH = sys.argv[0]
PROGNAME = os.path.basename(PROGPATH)
import commands
try:
modname = sys.modules[__name__].__file__
binpath = os.path.dirname(modname)
rivetconfigpath = os.path.join(binpath, "rivet-config")
rivetpypath = commands.getoutput(rivetconfigpath + " --pythonpath")
sys.path.append(rivetpypath)
except:
pass
import rivet
rivet.util.check_python_version()
rivet.util.set_process_name("rivet")
import time, logging, signal
## Parse command line options
from optparse import OptionParser, OptionGroup
parser = OptionParser(usage=__doc__, version="rivet v%s" % rivet.version())
anagroup = OptionGroup(parser, "Analysis handling")
anagroup.add_option("-a", "--analysis", dest="ANALYSES", action="append",
default=[], metavar="ANA",
help="add an analysis to the processing list.")
-# anagroup.add_option("-A", "--all-analyses", dest="ALL_ANALYSES", action="store_true",
-# default=False, help="add all analyses to the processing list.")
anagroup.add_option("--list-analyses", dest="LIST_ANALYSES", action="store_true",
default=False, help="show the list of available analyses' names. With -v, it shows the descriptions, too")
anagroup.add_option("--list-used-analyses", action="store_true", dest="LIST_USED_ANALYSES",
default=False, help="list the analyses used by this command (after subtraction of inappropriate ones)")
anagroup.add_option("--show-analysis", "--show-analyses", dest="SHOW_ANALYSES", action="append",
default=[], help="show the details of an analysis")
anagroup.add_option("--analysis-path", dest="ANALYSIS_PATH", metavar="PATH", default=None,
help="specify the analysis search path (cf. $RIVET_ANALYSIS_PATH).")
anagroup.add_option("--analysis-path-append", dest="ANALYSIS_PATH_APPEND", metavar="PATH", default=None,
help="append to the analysis search path (cf. $RIVET_ANALYSIS_PATH).")
anagroup.add_option("--pwd", dest="ANALYSIS_PATH_PWD", action="store_true", default=False,
help="append the current directory (pwd) to the analysis search path (cf. $RIVET_ANALYSIS_PATH).")
parser.add_option_group(anagroup)
extragroup = OptionGroup(parser, "Extra run settings")
extragroup.add_option("-H", "--histo-file", dest="HISTOFILE",
default="Rivet.yoda", help="specify the output histo file path (default = %default)")
extragroup.add_option("-x", "--cross-section", dest="CROSS_SECTION",
default=None, metavar="XS",
help="specify the signal process cross-section in pb")
extragroup.add_option("-n", "--nevts", dest="MAXEVTNUM", type="int",
default=None, metavar="NUM",
help="restrict the max number of events to read.")
extragroup.add_option("--runname", dest="RUN_NAME", default=None, metavar="NAME",
help="give an optional run name, to be prepended as a 'top level directory' in histo paths")
extragroup.add_option("--ignore-beams", dest="IGNORE_BEAMS", action="store_true", default=False,
help="Ignore input event beams when checking analysis compatibility.")
parser.add_option_group(extragroup)
timinggroup = OptionGroup(parser, "Timeouts and periodic operations")
timinggroup.add_option("--event-timeout", dest="EVENT_TIMEOUT", type="int",
default=21600, metavar="NSECS",
help="max time in whole seconds to wait for an event to be generated from the specified source (default = %default)")
timinggroup.add_option("--run-timeout", dest="RUN_TIMEOUT", type="int",
default=None, metavar="NSECS",
help="max time in whole seconds to wait for the run to finish. This can be useful on batch systems such "
"as the LCG Grid where tokens expire on a fixed wall-clock and can render long Rivet runs unable to write "
"out the final histogram file (default = unlimited)")
timinggroup.add_option("--histo-interval", dest="HISTO_WRITE_INTERVAL", type=int,
default=None, help="[experimental!] specify the number of events between histogram file updates. "
"Default is to only write out at the end of the run. Note that intermediate histograms will be those "
"from the analyze step only: analysis finalizing is currently not executed until the end of the run.")
parser.add_option_group(timinggroup)
verbgroup = OptionGroup(parser, "Verbosity control")
parser.add_option("-l", dest="NATIVE_LOG_STRS", action="append",
default=[], help="set a log level in the Rivet library")
verbgroup.add_option("-v", "--verbose", action="store_const", const=logging.DEBUG, dest="LOGLEVEL",
default=logging.INFO, help="print debug (very verbose) messages")
verbgroup.add_option("-q", "--quiet", action="store_const", const=logging.WARNING, dest="LOGLEVEL",
default=logging.INFO, help="be very quiet")
parser.add_option_group(verbgroup)
opts, args = parser.parse_args()
## Configure logging
logging.basicConfig(level=opts.LOGLEVEL, format="%(message)s")
## Control native Rivet library logger
for l in opts.NATIVE_LOG_STRS:
name, level = None, None
try:
name, level = l.split("=")
except:
name = "Rivet"
level = l
## Fix name
if name != "Rivet" and not name.startswith("Rivet."):
name = "Rivet." + name
try:
## Get right error type
level = rivet.LEVELS.get(level.upper(), None)
logging.debug("Setting log level: %s %d" % (name, level))
rivet.setLogLevel(name, level)
except:
logging.warning("Couldn't process logging string '%s'" % l)
## Use pandoc (if available) to make the TeX strings in
## summary and description more readable on the text terminal
# TODO: Maybe group many strings to be processed together, to save on system call / pandoc startup?
def detex(tex):
from distutils.spawn import find_executable
if not find_executable("pandoc"):
return tex
texheader = r"""
\newcommand{\text}[1]{#1}
\newcommand{\ensuremath}[1]{#1}
\newcommand{\emph}[1]{_#1_}
\newcommand{\textrm}[1]{#1}
\newcommand{\textit}[1]{_#1_}
\newcommand{\textbf}[1]{*#1*}
\newcommand{\mathrm}[1]{#1}
\newcommand{\mathit}[1]{_#1_}
\newcommand{\mathbf}[1]{*#1*}
\newcommand{\bm}[1]{*#1*}
\newcommand{\frac}[2]{#1/#2}
\newcommand{\sqrt}[1]{sqrt(#1)}
\newcommand{\hat}[1]{#1hat}
\newcommand{\bar}[1]{#1bar}
\newcommand{\d}[1]{d#1}
\newcommand{\degree}{^\circ }
\newcommand{\infty}{oo }
\newcommand{\exp}{exp }
\newcommand{\log}{log }
\newcommand{\ln}{ln }
\newcommand{\sin}{sin }
\newcommand{\cos}{cos }
\newcommand{\tan}{tan }
\newcommand{\sinh}{sinh }
\newcommand{\cosh}{cosh }
\newcommand{\tanh}{tanh }
\newcommand{\ell}{l}
\newcommand{\varphi}{\phi}
\newcommand{\varepsilon}{\epsilon}
\newcommand{\sim}{~}
\newcommand{\lesssim}{<~ }
\newcommand{\gtrsim}{>~ }
\newcommand{\neq}{!= }
\newcommand{\ge}{>= }
\newcommand{\gg}{>> }
\newcommand{\le}{<= }
\newcommand{\ll}{<< }
\newcommand{\pm}{+- }
\newcommand{\mp}{-+ }
\newcommand{\times}{x }
\newcommand{\cdot}{. }
\newcommand{\langle}{<}
\newcommand{\rangle}{>}
\newcommand{\gets}{<- }
\newcommand{\to}{-> }
\newcommand{\leftarrow}{<- }
\newcommand{\rightarrow}{-> }
\newcommand{\leftrightarrow}{<-> }
\newcommand{\Leftarrow}{<= }
\newcommand{\Rightarrow}{=> }
\newcommand{\Leftrightarrow}{ }
\newcommand{\left}{}
\newcommand{\right}{}
\newcommand{\!}{}
\newcommand{\/}{}
\newcommand{\rm}{}
\newcommand{\it}{}
\newcommand{\,}{ }
\newcommand{\;}{ }
\newcommand{\ }{ }
\newcommand{\unit}[2]{#1 #2}
\newcommand{\bar}[1]{#1bar}
\newcommand{\pT}{pT }
\newcommand{\perp}{T}
\newcommand{\MeV}{MeV }
\newcommand{\GeV}{GeV }
\newcommand{\TeV}{TeV }
"""
import subprocess, shlex
p = subprocess.Popen(shlex.split("pandoc -f latex -t plain --no-wrap"),
stdin=subprocess.PIPE, stdout=subprocess.PIPE)
plain, err = p.communicate((texheader + tex).replace("\n", ""))
plain = plain.replace(r"\&", "&")
# TODO: Replace \gamma, \mu, \tau, \Upsilon, \rho, \psi, \pi, \eta, \Delta, \Omega, \omega -> no-\ form?
# TODO: Replace e^+-
return plain[:-1] if plain[-1] == "\n" else plain
#print detex(r"Foo \! $\int \text{bar} \d{x} \sim \; \frac{1}{3} \neq \emph{foo}$ \to \gg bar")
## Parse supplied cross-section
if opts.CROSS_SECTION is not None:
xsstr = opts.CROSS_SECTION
try:
opts.CROSS_SECTION = float(xsstr)
except:
import re
suffmatch = re.search(r"[^\d.]", xsstr)
if not suffmatch:
raise ValueError("Bad cross-section string: %s" % xsstr)
factor = base = None
suffstart = suffmatch.start()
if suffstart != -1:
base = xsstr[:suffstart]
suffix = xsstr[suffstart:].lower()
if suffix == "mb":
factor = 1e+9
elif suffix == "mub":
factor = 1e+6
elif suffix == "nb":
factor = 1e+3
elif suffix == "pb":
factor = 1
elif suffix == "fb":
factor = 1e-3
elif suffix == "ab":
factor = 1e-6
if factor is None or base is None:
raise ValueError("Bad cross-section string: %s" % xsstr)
xs = float(base) * factor
opts.CROSS_SECTION = xs
## Print the available CLI options!
#if opts.LIST_OPTIONS:
# for o in parser.option_list:
# print o.get_opt_string()
# sys.exit(0)
## Set up signal handling
RECVD_KILL_SIGNAL = None
def handleKillSignal(signum, frame):
"Declare us as having been signalled, and return to default handling behaviour"
global RECVD_KILL_SIGNAL
logging.critical("Signal handler called with signal " + str(signum))
RECVD_KILL_SIGNAL = signum
signal.signal(signum, signal.SIG_DFL)
## Signals to handle
signal.signal(signal.SIGTERM, handleKillSignal);
signal.signal(signal.SIGHUP, handleKillSignal);
signal.signal(signal.SIGINT, handleKillSignal);
signal.signal(signal.SIGUSR1, handleKillSignal);
signal.signal(signal.SIGUSR2, handleKillSignal);
try:
signal.signal(signal.SIGXCPU, handleKillSignal);
except:
pass
## Override/modify analysis search path
if opts.ANALYSIS_PATH:
rivet.setAnalysisLibPaths(opts.ANALYSIS_PATH.split(":"))
if opts.ANALYSIS_PATH_APPEND:
for ap in opts.ANALYSIS_PATH_APPEND.split(":"):
rivet.addAnalysisLibPath(ap)
if opts.ANALYSIS_PATH_PWD:
rivet.addAnalysisLibPath(".")
## List of analyses
all_analyses = rivet.AnalysisLoader.analysisNames()
if opts.LIST_ANALYSES:
## Treat args as case-insensitive regexes if present
regexes = None
if args:
import re
regexes = [re.compile(arg, re.I) for arg in args]
for aname in all_analyses:
if not regexes:
toshow = True
else:
toshow = False
for regex in regexes:
if regex.search(aname):
toshow = True
break
if toshow:
msg = aname
if opts.LOGLEVEL <= logging.INFO:
a = rivet.AnalysisLoader.getAnalysis(aname)
msg = "%-25s %s" % (aname, detex(a.summary()))
print msg
sys.exit(0)
## Show analyses' details
if len(opts.SHOW_ANALYSES) > 0:
toshow = []
for i, a in enumerate(opts.SHOW_ANALYSES):
a_up = a.upper()
if a_up in all_analyses and a_up not in toshow:
toshow.append(a_up)
else:
## Treat as a case-insensitive regex
import re
regex = re.compile(a, re.I)
for ana in all_analyses:
if regex.search(ana) and a_up not in toshow:
toshow.append(ana)
## Show the matching analyses' details
import textwrap
for i, name in enumerate(sorted(toshow)):
ana = rivet.AnalysisLoader.getAnalysis(name)
print ""
print name
print len(name) * "="
print ""
print detex(ana.summary())
print ""
print "Status: %s" % ana.status()
print ""
if ana.inspireId():
print "Inspire ID: %s" % ana.inspireId()
print "Inspire URL: http://inspire-hep.net/record/%s" % ana.inspireId()
- # TODO: Need a way to get to HepData from an Inspire record
+ print "HepData URL: http://hepdata.cedar.ac.uk/view/ins%s" % ana.inspireId()
elif ana.spiresId():
print "Spires ID: %s" % ana.spiresId()
print "Inspire URL: http://inspire-hep.net/search?p=find+key+%s" % ana.spiresId()
print "HepData URL: http://hepdata.cedar.ac.uk/view/irn%s" % ana.spiresId()
if ana.experiment():
print "Experiment: %s" % ana.experiment(),
if ana.collider():
print "(%s)" % ana.collider()
if ana.year():
print "Year of publication: %s" % ana.year()
print "Authors:"
for a in ana.authors():
print " " + a
print ""
print "Description:"
twrap = textwrap.TextWrapper(width=75, initial_indent=2*" ", subsequent_indent=2*" ")
print twrap.fill(detex(ana.description()))
print ""
if ana.requiredBeams():
def pid_to_str(pid):
if pid == 11:
return "e-"
elif pid == -11:
return "e+"
elif pid == 2212:
return "p+"
elif pid == -2212:
return "p-"
elif pid == 10000:
return "*"
else:
return str(pid)
beamstrs = []
for bp in ana.requiredBeams():
beamstrs.append(pid_to_str(bp[0]) + " " + pid_to_str(bp[1]))
print "Beams:", ", ".join(beamstrs)
if ana.requiredEnergies():
print "Beam energies:", "; ".join(["(%0.1f, %0.1f)" % (epair[0], epair[1]) for epair in ana.requiredEnergies()]), "GeV"
else:
print "Beam energies: ANY"
if ana.runInfo():
print "Run details:"
twrap = textwrap.TextWrapper(width=75, initial_indent=2*" ", subsequent_indent=4*" ")
for l in ana.runInfo().split("\n"):
print twrap.fill(l)
if ana.references():
print ""
print "References:"
for r in ana.references():
url = None
if r.startswith("arXiv:"):
code = r.split()[0].replace("arXiv:", "")
url = "http://arxiv.org/abs/" + code
elif r.startswith("doi:"):
code = r.replace("doi:", "")
url = "http://dx.doi.org/" + code
if url is not None:
r += " - " + url
print " %s" % r
if i+1 < len(toshow):
print "\n"
sys.exit(0)
## Identify HepMC files/streams
## TODO: check readability, deal with stdin
if len(args) > 0:
HEPMCFILES = args
else:
HEPMCFILES = ["-"]
## Event number logging
def logNEvt(n, starttime, maxevtnum):
- nevtloglevel = logging.DEBUG
- if n % 10 == 0:
- nevtloglevel = logging.DEBUG + 5
- if n % 100 == 0:
- nevtloglevel = logging.INFO
- if n % 200 == 0:
- nevtloglevel = logging.INFO + 5
- if n % 500 == 0:
- nevtloglevel = logging.WARNING
- if n % 1000 == 0:
- nevtloglevel = logging.WARNING + 5
if n % 10000 == 0:
nevtloglevel = logging.CRITICAL
+ elif n % 1000 == 0:
+ nevtloglevel = logging.WARNING
+ elif n % 100 == 0:
+ nevtloglevel = logging.INFO
+ else:
+ nevtloglevel = logging.DEBUG
timecurrent = time.time()
timeelapsed = timecurrent - starttime
if maxevtnum is None:
logging.log(nevtloglevel, "Event %d (%d s elapsed)" % (n, timeelapsed))
else:
timeleft = (maxevtnum-n)*timeelapsed/n
eta = time.strftime("%a %b %d %H:%M", time.localtime(timecurrent + timeleft))
- logging.log(nevtloglevel, "Event %d (%d s elapsed / %d s left) -> ETA: %s"
- % (n, timeelapsed, timeleft, eta))
+ logging.log(nevtloglevel, "Event %d (%d s elapsed / %d s left) -> ETA: %s" %
+ (n, timeelapsed, timeleft, eta))
## Set up analysis handler
RUNNAME = opts.RUN_NAME or ""
ah = rivet.AnalysisHandler(RUNNAME)
ah.setIgnoreBeams(opts.IGNORE_BEAMS)
-# if opts.ALL_ANALYSES:
-# opts.ANALYSES = all_analyses
for a in opts.ANALYSES:
#a_up = a.upper()
## Print warning message and exit if not a valid analysis name
if not a in all_analyses:
logging.warning("'%s' is not a known Rivet analysis! Do you need to set RIVET_ANALYSIS_PATH or use the --pwd switch?\n" % a)
# TODO: lay out more neatly, or even try for a "did you mean XXXX?" heuristic?
logging.warning("There are %d currently available analyses:\n" % len(all_analyses) + ", ".join(all_analyses))
sys.exit(1)
logging.debug("Adding analysis '%s'" % a)
ah.addAnalysis(a)
## Read and process events
run = rivet.Run(ah)
if opts.CROSS_SECTION is not None:
logging.info("User-supplied cross-section = %e pb" % opts.CROSS_SECTION)
run.setCrossSection(opts.CROSS_SECTION)
if opts.LIST_USED_ANALYSES is not None:
run.setListAnalyses(opts.LIST_USED_ANALYSES)
## Print platform type
import platform
logging.info("Rivet %s running on machine %s (%s)" % (rivet.version(), platform.node(), platform.machine()))
def min_nonnull(a, b):
"A version of min which considers None to always be greater than a real number"
rtn = min(a, b)
if rtn is not None:
return rtn
if a is not None:
return a
return b
## Set up an event timeout handler
class TimeoutException(Exception):
pass
if opts.EVENT_TIMEOUT or opts.RUN_TIMEOUT:
def evttimeouthandler(signum, frame):
logging.warn("It has taken more than %d secs to get an event! Is the input event stream working?" %
min_nonnull(opts.EVENT_TIMEOUT, opts.RUN_TIMEOUT))
raise TimeoutException("Event timeout")
signal.signal(signal.SIGALRM, evttimeouthandler)
## Init run based on one event
hepmcfile = HEPMCFILES[0]
## Apply a file-level weight derived from the filename
hepmcfileweight = 1.0
if ":" in hepmcfile:
hepmcfile, hepmcfileweight = hepmcfile.rsplit(":", 1)
hepmcfileweight = float(hepmcfileweight)
try:
if opts.EVENT_TIMEOUT or opts.RUN_TIMEOUT:
signal.alarm(min_nonnull(opts.EVENT_TIMEOUT, opts.RUN_TIMEOUT))
init_ok = run.init(hepmcfile, hepmcfileweight)
signal.alarm(0)
if not init_ok:
logging.error("Failed to initialise using event file '%s'... exiting" % hepmcfile)
sys.exit(2)
except TimeoutException, te:
logging.error("Timeout in initialisation from event file '%s'... exiting" % hepmcfile)
sys.exit(3)
## Event loop
evtnum = 0
starttime = time.time()
for fileidx, hepmcfile in enumerate(HEPMCFILES):
## Apply a file-level weight derived from the filename
hepmcfileweight = 1.0
if ":" in hepmcfile:
hepmcfile, hepmcfileweight = hepmcfile.rsplit(":", 1)
hepmcfileweight = float(hepmcfileweight)
## Open next HepMC file (NB. this doesn't apply to the first file: it was already used for the run init)
if fileidx > 0:
run.openFile(hepmcfile, hepmcfileweight)
if not run.readEvent():
logging.warning("Could not read events from '%s'" % hepmcfile)
continue
msg = "Reading events from '%s'" % hepmcfile
if hepmcfileweight != 1.0:
msg += " (file weight = %e)" % hepmcfileweight
logging.info(msg)
while opts.MAXEVTNUM is None or evtnum < opts.MAXEVTNUM:
evtnum += 1
logNEvt(evtnum, starttime, opts.MAXEVTNUM)
## Process this event
processed_ok = run.processEvent()
if not processed_ok:
logging.warn("Event processing failed for evt #%i!" % evtnum)
break
## Set flag to exit event loop if run timeout exceeded
if opts.RUN_TIMEOUT and (time.time() - starttime) > opts.RUN_TIMEOUT:
logging.warning("Run timeout of %d secs exceeded... exiting gracefully" % opts.RUN_TIMEOUT)
RECVD_KILL_SIGNAL = True
## Exit the loop if signalled
if RECVD_KILL_SIGNAL is not None:
break
## Read next event (with timeout handling if requested)
try:
if opts.EVENT_TIMEOUT:
signal.alarm(opts.EVENT_TIMEOUT)
read_ok = run.readEvent()
signal.alarm(0)
if not read_ok:
break
except TimeoutException, te:
logging.error("Timeout in reading event from '%s'... exiting" % hepmcfile)
sys.exit(3)
## Write a histo file snapshot if appropriate
if opts.HISTO_WRITE_INTERVAL is not None:
if evtnum % opts.HISTO_WRITE_INTERVAL == 0:
ah.writeData(opts.HISTOFILE)
logging.info("Finished event loop")
run.finalize()
## Finalize and write out data file
print "Cross-section = %e pb" % ah.crossSection()
ah.finalize()
ah.writeData(opts.HISTOFILE)

File Metadata

Mime Type
text/x-diff
Expires
Sun, Feb 23, 2:44 PM (22 h, 27 m)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
4486689
Default Alt Text
(158 KB)

Event Timeline