Remove clang-format key not present in some versions.

.clang-format

@@ -32,7 +32,6 @@ BraceWrapping:
   BeforeCatch:     false
   BeforeElse:      false
   IndentBraces:    false
-  SplitEmptyFunctionBody: true
 BreakBeforeBinaryOperators: All
 BreakBeforeBraces: Attach
 BreakBeforeInheritanceComma: false

stan/math/prim/scal/fun/gamma_p.hpp

@@ -48,7 +48,7 @@ namespace math {
    \f[
    \frac{\partial \, P(a, z)}{\partial z} = \frac{z^{a-1}e^{-z}}{\Gamma(a)}
    \f]
-   
+   
    * @throws domain_error if x is at pole
 
  */

test/unit/math/rev/mat/prob/neg_binomial_2_log_glm_lpmf_test.cpp

@@ -111,15 +111,15 @@ typedef std::chrono::high_resolution_clock::time_point TimeVar;
 TEST(ProbDistributionsNegBinomial2LogGLM, glm_matches_neg_binomial_2_log_speed)
 { for (size_t exponent = 7; exponent < 10; exponent++) { const int R = 10000;
     const int C = std::pow(3, exponent);
-  
+  
   Matrix<int,Dynamic,1> n(R, 1);
   for (size_t i = 0; i < R; i++) {
     n[i] = rand()%2;
   }
-  
+  
   int T1 = 0;
   int T2 = 0;
-  
+  
   for (size_t testnumber = 0; testnumber < 30; testnumber++){
     Matrix<double, Dynamic, Dynamic> xreal = Matrix<double, Dynamic,
 Dynamic>::Random(R, C); Matrix<double, Dynamic, 1> betareal = Matrix<double,
@@ -128,12 +128,12 @@ Dynamic, Dynamic>::Random(C, 1); Matrix<double, 1, 1> alphareal = Matrix<double,
 1>::Ones() * alphareal; Matrix<double, Dynamic, 1> phireal = Matrix<double,
 Dynamic, Dynamic>::Random(R, 1) + Matrix<double, Dynamic, 1>::Ones(R, 1);  // We
 want phireal to be positive.
-    
+    
     Matrix<var, Dynamic, 1> beta = betareal;
     Matrix<var, Dynamic, 1> phi = phireal;
     Matrix<var, Dynamic, 1> theta(R, 1);
 
-  
+  
     TimeVar t1 = timeNow();
     theta = (xreal * beta) + alpharealvec;
     var lp = stan::math::neg_binomial_2_log_lpmf(n, theta, phi);
@@ -145,7 +145,7 @@ want phireal to be positive.
 
     Matrix<var, Dynamic, 1> beta2 = betareal;
     Matrix<var, Dynamic, 1> phi2 = phireal;
-    
+    
     TimeVar t3 = timeNow();
     var lp2 = stan::math::neg_binomial_2_log_glm_lpmf(n, xreal, beta2,
 alphareal[0], phi2); lp2.grad(); TimeVar t4 = timeNow();
@@ -154,7 +154,7 @@ alphareal[0], phi2); lp2.grad(); TimeVar t4 = timeNow();
     T2 += duration(t4 - t3);
 
   }
-  
+  
   std::cout << "Existing Primitives:" << std::endl << T1 << std::endl  << "New
 Primitives:" << std::endl << T2 << std::endl;
 }

test/unit/math/rev/mat/prob/normal_id_glm_lpdf_test.cpp

@@ -106,15 +106,15 @@ typedef std::chrono::high_resolution_clock::time_point TimeVar;
 TEST(ProbDistributionsNormalIdGLM, glm_matches_normal_id_speed) {
   const int R = 30000;
   const int C = 1000;
-  
+  
   Matrix<int,Dynamic,1> n(R, 1);
   for (size_t i = 0; i < R; i++) {
     n[i] = rand()%2000;
   }
-  
+  
   int T1 = 0;
   int T2 = 0;
-  
+  
   for (size_t testnumber = 0; testnumber < 1; testnumber++){
     Matrix<double, Dynamic, Dynamic> xreal = Matrix<double, Dynamic,
 Dynamic>::Random(R, C); Matrix<double, Dynamic, 1> betareal = Matrix<double,
@@ -124,12 +124,12 @@ Matrix<double, Dynamic, Dynamic>::Random(R, 1)
 and 1. We add 1 to it to get positive entries. Matrix<double, 1, 1> alphareal =
 Matrix<double, 1, 1>::Random(1, 1); Matrix<double, Dynamic, 1> alpharealvec =
 Matrix<double, R, 1>::Ones() * alphareal;
-    
+    
     Matrix<var, Dynamic, 1> beta = betareal;
     Matrix<var, Dynamic, 1> sigma = sigmareal;
     Matrix<var, Dynamic, 1> theta(R, 1);
 
-  
+  
     TimeVar t1 = timeNow();
     theta = (xreal * beta) + alpharealvec;
     var lp = stan::math::normal_lpdf(n, theta, sigma);
@@ -141,7 +141,7 @@ Matrix<double, R, 1>::Ones() * alphareal;
 
     Matrix<var, Dynamic, 1> beta2 = betareal;
     Matrix<var, Dynamic, 1> sigma2 = sigmareal;
-    
+    
     TimeVar t3 = timeNow();
     var lp2 = stan::math::normal_id_glm_lpdf(n, xreal, beta2, alphareal[0],
       sigma2);
@@ -152,7 +152,7 @@ Matrix<double, R, 1>::Ones() * alphareal;
     T2 += duration(t4 - t3);
 
   }
-  
+  
   std::cout << "Existing Primitives:" << std::endl << T1 << std::endl  << "New
 Primitives:" << std::endl << T2 << std::endl;
 }

test/unit/math/rev/mat/prob/poisson_log_glm_lpmf_test.cpp

@@ -96,26 +96,26 @@ typedef std::chrono::high_resolution_clock::time_point TimeVar;
 TEST(ProbDistributionsPoissonLogGLM, glm_matches_poisson_log_speed) {
   const int R = 30000;
   const int C = 1000;
-  
+  
   Matrix<int,Dynamic,1> n(R, 1);
   for (size_t i = 0; i < R; i++) {
     n[i] = rand()%200;
   }
-  
+  
   int T1 = 0;
   int T2 = 0;
-  
+  
   for (size_t testnumber = 0; testnumber < 30; testnumber++){
     Matrix<double, Dynamic, Dynamic> xreal = Matrix<double, Dynamic,
 Dynamic>::Random(R, C); Matrix<double, Dynamic, 1> betareal = Matrix<double,
 Dynamic, Dynamic>::Random(C, 1); Matrix<double, 1, 1> alphareal = Matrix<double,
 1, 1>::Random(1, 1); Matrix<double, Dynamic, 1> alpharealvec = Matrix<double, R,
 1>::Ones() * alphareal;
-    
+    
     Matrix<var, Dynamic, 1> beta = betareal;
     Matrix<var, Dynamic, 1> theta(R, 1);
 
-  
+  
     TimeVar t1 = timeNow();
     theta = (xreal * beta) + alpharealvec;
     var lp = stan::math::poisson_log_lpmf(n, theta);
@@ -126,7 +126,7 @@ Dynamic, Dynamic>::Random(C, 1); Matrix<double, 1, 1> alphareal = Matrix<double,
     stan::math::recover_memory();
 
     Matrix<var, Dynamic, 1> beta2 = betareal;
-    
+    
     TimeVar t3 = timeNow();
     var lp2 = stan::math::poisson_log_glm_lpmf(n, xreal, beta2, alphareal[0]);
     lp2.grad();
@@ -136,7 +136,7 @@ Dynamic, Dynamic>::Random(C, 1); Matrix<double, 1, 1> alphareal = Matrix<double,
     T2 += duration(t4 - t3);
 
   }
-  
+  
   std::cout << "Existing Primitives:" << std::endl << T1 << std::endl  << "New
 Primitives:" << std::endl << T2 << std::endl;
 }