Merge branch 'dev'

Author: molpopgen

.travis.yml

@@ -15,6 +15,9 @@ addons:
     - clang
     - zlib1g
     - zlib1g-dev
+    - libboost-dev
+    - libboost-system-dev
+    - libboost-test-dev
 
 notifications:
   email: false

README.md

@@ -22,6 +22,10 @@ Comments are welcome.
 
 	- Kevin Thornton <krthornt@uci.edu>
 
+# User's group
+
+Please post to the [libsequence user group](https://groups.google.com/forum/#!forum/libsequence-users) for help.
+
 # Build status
 
 * master branch: [![Build Status](https://travis-ci.org/molpopgen/libsequence.svg?branch=master)](https://travis-ci.org/molpopgen/libsequence)

REVISION_HISTORY.md

@@ -30,6 +30,18 @@ However:
 
 1. Sequence/SeqRegexes.hpp -- not working.  This will not be fixed until GCC supports <regex>.  The function is now currently implemented in a non-regex manner, which is lame, but it works.
 
+## libsequence 1.8.9
+
+Major API and ABI changes!  This release is a __big__ step towards a "2.0"-style release.
+
+* Issue #8 fixed
+* Sequence::PolyTableSlice will throw std::runtime_error if input range is not properly sorted
+* War on "mutable". The use of this keyword has been removed from the library to the best extent possible.
+* The API for calculations involving codons has been modernized.  This includes Sequence::Comeron95, Sequence::RedundancyCom95, Sequence::WeightingScheme2 (and derived types), Sequence::WeightingScheme3 (and derived types), Sequence::TwoSubs, Sequence::ThreeSubs, functions in Sequence/PathwayHelper.hpp
+* Sequence::PolyTable (and derived types) have been refactored.  The fundamental idea is the same, but the API is modernized.  IMO, it is still imperfect, and can be further changed to reflect more idiomatic C++11, but that'll have to wait.
+* Private data members for classes have been hidden using the [PIMPL idiom](https://en.wikipedia.org/wiki/Opaque_pointer).  This goes a long way to future-proofing the ABI compatibility of these types against further implementation changes such as bug fixes.
+* Sequence/SummStats/classic.hpp provides a sneak previous of how summary statistics will work in the future, once the deprecated Sequence::PolySNP and Sequence::PolySIM can finally be removed
+
 ## libsequence 1.8.8
 
 * l-HAF statistic added (Sequence/SummStats/lHaf.hpp)

Sequence/Coalescent/SimTypes.hpp

@@ -173,7 +173,7 @@ namespace Sequence
 	The sample size being simulated.  The 2*nsam-1 nodes
 	in the tree are therefore indexed from 0 to 2*nsam-2
       */
-      mutable int nsam;
+      const int nsam;
       /*!
 	The current number of nodes in the tree.  By current, it 
 	is meant the current time in the simulation.  At the start

Sequence/Comeron95.hpp

@@ -113,62 +113,31 @@ long with libsequence.  If not, see <http://www.gnu.org/licenses/>.
 
 /*! \example gestimator.cc
  */
+#include <Sequence/WeightingSchemes.hpp>
 #include <Sequence/SeqEnums.hpp>
+#include <Sequence/Seq.hpp>
+#include <array>
 
 namespace Sequence
   {
-    class Seq;
-    class Sites;
-    class RedundancyCom95;
-    class WeightingScheme2;
-    class WeightingScheme3;
+    using Com95_t = std::array<double,19>;
     class Comeron95 
     {
     private:
-      bool __2wasNULL,__3wasNULL,__red_was_NULL;
-      WeightingScheme2 *weights2;
-      WeightingScheme3 *weights3;
-      int valid, maxhits, genetic_code, weighting_scheme;
-      //see Comeron '95 for a discussion of the method--will document later
-      double Qs, Bs, Qa, Ba, A2S, A4, As, A2V, A0, Aa;
-      double q0, q2S, q2V, q4, p0, p2S, p2V, p4;
-      Sites *sites;
-      const RedundancyCom95 *sitesObj;
-      void diverge (const Sequence::Seq * seq1, const Sequence::Seq * seq2,
-                    WeightingScheme2 *_weights2,
-                    WeightingScheme3 *_weights3);
-      void omega (const Sequence::Seq * seqobj1, const Sequence::Seq * seqobj2);
-      double Ka, Ks;
+      struct Com95impl;
+      std::unique_ptr<Com95impl> impl;
     public:
-      explicit Comeron95 (const Sequence::Seq * seqa,
-                          const Sequence::Seq * seqb,
-                          int max = 3, 
-			  const Sequence::RedundancyCom95 * genetic_code_redundancy = NULL,
-			  GeneticCodes code = GeneticCodes::UNIVERSAL,
-                          WeightingScheme2 *weights2 = NULL,
-                          WeightingScheme3 *weights3 = NULL);
+      explicit Comeron95( GeneticCodes code = GeneticCodes::UNIVERSAL );
       Comeron95( const Comeron95 & ) = delete;
       Comeron95 & operator=(const Comeron95 & ) = delete;
-      ~Comeron95 (void);
-      double ka (void) const;
-      double ks (void) const;
-      double ratio (void) const;
-      double P0 (void) const;
-      double P2S (void) const;
-      double P2V (void) const;
-      double P4 (void) const;
-      double Q0 (void) const;
-      double Q2S (void) const;
-      double Q2V (void) const;
-      double Q4 (void) const;
-      double as (void) const;
-      double aa (void) const;
-      double bs (void) const;
-      double ba (void) const;
-      double L0 (void) const;
-      double L2S (void) const;
-      double L2V (void) const;
-      double L4 (void) const;
+      Com95_t operator()(const Sequence::Seq & seqa,
+			 const Sequence::Seq & seqb,
+			 int maxdiffs = 3);
+      Com95_t operator()(const Sequence::Seq & seqa,
+			 const Sequence::Seq & seqb,
+			 const WeightingScheme2 *weights2,
+			 const WeightingScheme3 *weights3,
+			 int maxdiffs = 3);
   };
 }
 

Sequence/FST.hpp

@@ -32,6 +32,8 @@ long with libsequence.  If not, see <http://www.gnu.org/licenses/>.
 
 /*! \file FST.hpp
   @brief delcaration of a class (Sequence::FST) to analyze population structure
+
+  \deprecated Will be removed in libsequence 2.0
 */
 namespace Sequence
 {
@@ -40,8 +42,7 @@ namespace Sequence
   class FST 
   {
   private:
-    void doCalcs(void) const;
-    mutable std::unique_ptr<FSTimpl> impl;
+    std::unique_ptr<FSTimpl> impl;
   public:
     explicit FST(const PolyTable *data, unsigned npop, const unsigned *config=NULL,
 		 const double *weights=NULL, bool haveOutgroup = false,

Sequence/GranthamWeights.hpp

@@ -46,28 +46,14 @@ long with libsequence.  If not, see <http://www.gnu.org/licenses/>.
 namespace Sequence
   {
   class Grantham;
-  class GranthamWeights2 : public WeightingScheme2
+  struct GranthamWeights2 : public WeightingScheme2
     {
-    private:
-      Sequence::GeneticCodes code;
-      mutable double __weights[2];//logical const
-    public:
-      explicit GranthamWeights2(Sequence::GeneticCodes genetic_code = GeneticCodes::UNIVERSAL);
-      ~GranthamWeights2(void);
-      void Calculate(const std::string &codon1, const std::string &codon2) const;
-      double *weights(void) const;
+      weights2_t operator()(const std::string &codon1, const std::string &codon2,Sequence::GeneticCodes genetic_code) const;
     };
 
-  class GranthamWeights3 : public WeightingScheme3
+  struct GranthamWeights3 : public WeightingScheme3
     {
-    private:
-      GeneticCodes code;
-      mutable double __weights[6];//logical const
-    public:
-      explicit GranthamWeights3(Sequence::GeneticCodes genetic_code = GeneticCodes::UNIVERSAL);
-      ~GranthamWeights3(void);
-      void Calculate(const std::string &codon1, const std::string &codon2) const;
-      double *weights(void) const;
+      weights3_t operator()(const std::string &codon1, const std::string &codon2,Sequence::GeneticCodes genetic_code) const;
     };
 }
 #endif

Sequence/HKA.hpp

@@ -42,16 +42,16 @@ namespace Sequence
       SA is the number of polymorphic sites in species A,
       SB is the same quantity for species B
     */
-    mutable unsigned SA,SB; //# seg sites
+    const unsigned SA,SB; //# seg sites
     /*!
       D is a measure of divergence between species A and B,
       i.e. mean pairwise differences
     */
-    mutable double D;       // divergence
+    const double D;       // divergence
     /*!
       The sample sizes in species A and B
     */
-    mutable unsigned nA,nB; //sample sizes
+    const unsigned nA,nB; //sample sizes
     HKAdata();
     HKAdata(const HKAdata & d);
     HKAdata(unsigned sa,unsigned sb,double d,

Sequence/PathwayHelper.hpp

@@ -33,11 +33,13 @@ long with libsequence.  If not, see <http://www.gnu.org/licenses/>.
   or the intermedate codons that occurs between two different codons
 */
 #include <string>
-
+#include <array>
 namespace Sequence
 {
-  void Intermediates2(std::string *intermediates,const std::string &codon1, const std::string &codon2);
-  void Intermediates3(std::string *intermediates, const  std::string &codon1, const std::string &codon2);
+  using Inter2_t = std::array<std::string,2>;
+  using Inter3_t = std::array<std::string,9>;
+  Inter2_t Intermediates2(const std::string &codon1, const std::string &codon2);
+  Inter3_t Intermediates3(const std::string &codon1, const std::string &codon2);
 }
 #endif
 

Sequence/PolySIM.hpp

@@ -35,6 +35,8 @@ long with libsequence.  If not, see <http://www.gnu.org/licenses/>.
   that outgroup information is not required, as the 0,1 coding of a
   SimData object (usually) reflects ancestral and derived.
   @short Analysis of coalescent simulation data
+
+  \deprecated Will be removed in libsequence 2.0
 */
 #include <Sequence/PolySNP.hpp>