src all command sources: using the main function in each case from the namespace and also the about and help footer messages

Author: andrewdavidsmith

src/bound_pop.cpp

@@ -18,11 +18,6 @@
  * <http://www.gnu.org/licenses/>.
  */
 
-const auto about_msg = R"(
-preseq bound_pop: Estimate the size of the underlying population based on
-counts of observed species in an initial sample.
-)";
-
 #include "bound_pop.hpp"
 
 #include "common.hpp"
@@ -81,22 +76,22 @@ report_bootstrapped_moments(const std::vector<double> &bootstrap_moments,
 }
 
 // bounding n_0
-int
-bound_pop_main(int argc, char *argv[]) {  // NOLINT (*-avoid-c-arrays)
+auto
+bound_pop::main(int argc, char *argv[]) -> int {  // NOLINT (*-avoid-c-arrays)
   try {
-    bool verbose = false;
-    bool PAIRED_END = false;
-    bool HIST_INPUT = false;
-    bool VALS_INPUT = false;
-    bool QUICK_MODE = false;
+    bool verbose{false};
+    bool PAIRED_END{false};
+    bool HIST_INPUT{false};
+    bool VALS_INPUT{false};
+    bool QUICK_MODE{false};
 
     std::string input_file_name;
     std::string outfile;
     std::string histogram_outfile;
 
 #ifdef HAVE_HTSLIB
-    bool BAM_FORMAT_INPUT = false;
-    std::size_t MAX_SEGMENT_LENGTH = 5000;
+    bool BAM_FORMAT_INPUT{false};
+    std::size_t MAX_SEGMENT_LENGTH{5000};
     std::uint32_t n_threads{1};
 #endif
 
@@ -128,17 +123,17 @@ bound_pop_main(int argc, char *argv[]) {  // NOLINT (*-avoid-c-arrays)
     app.add_option("-c,--clevel", c_level, "level for confidence intervals");
     app.add_flag("-P,--pe", PAIRED_END, "input is paired end read file");
     app.add_flag("-H,--hist", HIST_INPUT,
-                   "input is a text file containing the observed histogram");
+                 "input is a text file containing the observed histogram");
     app.add_flag("-V,--vals", VALS_INPUT,
-                   "input is a text file containing only the observed duplicate counts");
+                 "input is a text file containing only the observed duplicate counts");
 #ifdef HAVE_HTSLIB
     app.add_flag("-B,--bam", BAM_FORMAT_INPUT,
-                   "input is in BAM format");
+                 "input is in BAM format");
     app.add_option("-l,--seg_len", MAX_SEGMENT_LENGTH,
                    "maximum segment length when merging paired end bam reads");
 #endif
     app.add_flag("-Q,--quick", QUICK_MODE,
-                   "quick mode, estimate without bootstrapping");
+                 "quick mode, estimate without bootstrapping");
     app.add_option("-r,--seed", seed, "seed for random number generator");
     app.add_flag("-v,--verbose", verbose, "print more info");
     // clang-format on
@@ -193,7 +188,7 @@ bound_pop_main(int argc, char *argv[]) {  // NOLINT (*-avoid-c-arrays)
     std::size_t idx = 1;
     while (idx < std::size(counts_hist) && counts_hist[idx]) {
       // idx + 1 because function calculates (x-1)!
-      measure_moments.push_back(std::exp(factorial(idx + 1) +
+      measure_moments.push_back(std::exp(log_factorial(idx + 1) +
                                          std::log(counts_hist[idx]) -
                                          std::log(counts_hist[1])));
       if (!std::isfinite(measure_moments.back())) {
@@ -324,7 +319,7 @@ bound_pop_main(int argc, char *argv[]) {  // NOLINT (*-avoid-c-arrays)
         std::vector<double> bootstrap_moments(1, 1.0);
         // moments[r] = (r + 1)! n_{r+1} / n_1
         for (std::size_t i = 0; i < 2 * max_num_points; i++)
-          bootstrap_moments.push_back(std::exp(factorial(i + 3) +
+          bootstrap_moments.push_back(std::exp(log_factorial(i + 3) +
                                                std::log(sample_hist[i + 2]) -
                                                std::log(sample_hist[1])));
 

src/c_curve.cpp

@@ -17,11 +17,6 @@
  * with this program. If not, see <http://www.gnu.org/licenses/>.
  */
 
-static constexpr auto about_msg = R"(
-Generate the full observed complexity curve for data. This does not
-extrapolate, but instead resamples from the given data.
-)";
-
 #include "c_curve.hpp"
 
 #include "common.hpp"
@@ -42,8 +37,8 @@ extrapolate, but instead resamples from the given data.
 
 // NOLINTBEGIN(*-avoid-magic-numbers,*-narrowing-conversions)
 
-int
-c_curve_main(int argc, char *argv[]) {  // NOLINT(*-avoid-c-arrays)
+auto
+c_curve::main(int argc, char *argv[]) -> int {  // NOLINT(*-avoid-c-arrays)
   try {
     bool verbose = false;
     bool PAIRED_END = false;

src/gc_extrap.cpp

@@ -18,18 +18,6 @@
  * <http://www.gnu.org/licenses/>.
  */
 
-static constexpr auto about_msg = R"(
-preseq gc_extrap: Extrapolate the size of the covered genome by mapped reads.
-)";
-
-static constexpr auto footer_msg = R"(
-This approach is described in Daley & Smith (2014). The method is the same as
-for lc_extrap: using rational function approximation to a power-series
-expansion for the number of "unobserved" bases in the initial sample. The
-gc_extrap method is adapted to deal with individual nucleotides rather than
-distinct reads.
-)";
-
 #include "gc_extrap.hpp"
 
 #include "common.hpp"
@@ -87,8 +75,8 @@ write_predicted_coverage_curve(
   }
 }
 
-int
-gc_extrap_main(int argc, char *argv[]) {  // NOLINT(*-avoid-c-arrays)
+auto
+gc_extrap::main(int argc, char *argv[]) -> int {  // NOLINT(*-avoid-c-arrays)
   try {
     static constexpr auto MIN_REQUIRED_COUNTS = 4;
 

src/lc_extrap.cpp

@@ -18,18 +18,6 @@
  * <http://www.gnu.org/licenses/>.
  */
 
-constexpr auto about_msg = R"(
-preseq lc_extrap: Estimate the complexity curve for a sequencing library.
-)";
-
-constexpr auto footer_msg = R"(
-This is the approach described in Daley & Smith (2013). The method applies
-rational function approximation via continued fractions with the original goal
-of estimating the number of distinct reads that a sequencing library would
-yield upon deeper sequencing.  This method has been used for many different
-purposes since then.
-)";
-
 #include "lc_extrap.hpp"
 
 #include "common.hpp"
@@ -52,8 +40,8 @@ purposes since then.
 
 // NOLINTBEGIN(*-avoid-magic-numbers,*-narrowing-conversions)
 
-int
-lc_extrap_main(int argc, char *argv[]) {  // NOLINT(*-avoid-c-arrays)
+auto
+lc_extrap::main(int argc, char *argv[]) -> int {  // NOLINT(*-avoid-c-arrays)
   try {
     static const std::size_t min_required_counts = 4;
     static const std::string min_required_counts_error_message =
@@ -72,16 +60,16 @@ lc_extrap_main(int argc, char *argv[]) {  // NOLINT(*-avoid-c-arrays)
     double c_level = 0.95;
     std::uint32_t seed = 408;
 
-    /* FLAGS */
-    bool verbose = false;
-    bool VALS_INPUT = false;
-    bool PAIRED_END = false;
-    bool HIST_INPUT = false;
-    bool SINGLE_ESTIMATE = false;
-    bool allow_defects = false;
+    // flags
+    bool verbose{false};
+    bool VALS_INPUT{false};
+    bool PAIRED_END{false};
+    bool HIST_INPUT{false};
+    bool SINGLE_ESTIMATE{false};
+    bool allow_defects{false};
 
 #ifdef HAVE_HTSLIB
-    bool BAM_FORMAT_INPUT = false;
+    bool BAM_FORMAT_INPUT{false};
     std::size_t MAX_SEGMENT_LENGTH = 5000;
     std::uint32_t n_threads{1};
 #endif

src/pop_size.cpp

@@ -18,18 +18,6 @@
  * <http://www.gnu.org/licenses/>.
  */
 
-static constexpr auto about_msg = R"(
-preseq pop_size: Estimate the total population size using a small sample from
-the population.
-)";
-
-static constexpr auto footer_msg = R"(
-Estimate the total population size using the approach described in Daley &
-Smith (2013), extrapolating to very long range. Default parameters assume that
-the initial sample represents at least 1e-9 of the population, which is
-sufficient for every example application we have seen.
-)";
-
 #include "pop_size.hpp"
 #include "common.hpp"
 #include "load_data_for_complexity.hpp"
@@ -60,8 +48,8 @@ using std::vector;
 
 // NOLINTBEGIN(*-avoid-magic-numbers,*-narrowing-conversions)
 
-int
-pop_size_main(int argc, char *argv[]) {  // NOLINT(*-avoid-c-arrays)
+auto
+pop_size::main(int argc, char *argv[]) -> int {  // NOLINT(*-avoid-c-arrays)
   try {
     static const std::size_t min_required_counts = 4;
     static const string min_required_counts_error_message =
@@ -94,11 +82,11 @@ pop_size_main(int argc, char *argv[]) {  // NOLINT(*-avoid-c-arrays)
     std::size_t MAX_SEGMENT_LENGTH = 5000;
     uint32_t n_threads{1};
 #endif
-    CLI::App app{rlstrip(about_msg)};
+    CLI::App app{rlstrip(pop_size::about_msg)};
     argv = app.ensure_utf8(argv);
     app.usage("\nUsage: preseq pop_size [OPTIONS]");
     if (argc >= 2)
-      app.footer(rlstrip(footer_msg));
+      app.footer(rlstrip(pop_size::footer_msg));
 
     // clang-format off
     app.add_option("-i,--input", input_file_name, "input file")