Merge pull request #456 from eseiler/infra/codechecker

[INFRA] Codechecker

.github/config/REUSE.toml

@@ -0,0 +1,14 @@
+# SPDX-FileCopyrightText: 2006-2024 Knut Reinert & Freie Universität Berlin
+# SPDX-FileCopyrightText: 2016-2024 Knut Reinert & MPI für molekulare Genetik
+# SPDX-License-Identifier: CC0-1.0
+
+version = 1
+SPDX-PackageName = "Raptor"
+SPDX-PackageDownloadLocation = "https://github.com/seqan/raptor/"
+
+[[annotations]]
+path = "cppcheck"
+precedence = "aggregate"
+SPDX-FileCopyrightText = ["2006-2024, Knut Reinert & Freie Universität Berlin", "2016-2024, Knut Reinert & MPI für molekulare Genetik"]
+SPDX-License-Identifier = "CC0-1.0"
+

.github/config/codechecker.skip

@@ -0,0 +1,6 @@
+# SPDX-FileCopyrightText: 2006-2024, Knut Reinert & Freie Universität Berlin
+# SPDX-FileCopyrightText: 2016-2024, Knut Reinert & MPI für molekulare Genetik
+# SPDX-License-Identifier: CC0-1.0
+
++${GITHUB_WORKSPACE}/*
+-*

.github/config/codechecker.yml

@@ -0,0 +1,21 @@
+# SPDX-FileCopyrightText: 2006-2024, Knut Reinert & Freie Universität Berlin
+# SPDX-FileCopyrightText: 2016-2024, Knut Reinert & MPI für molekulare Genetik
+# SPDX-License-Identifier: CC0-1.0
+
+analyze:
+  - --enable=performance
+  - --cppcheckargs=${GITHUB_WORKSPACE}/.github/config/cppcheck
+  - --enable=bugprone
+  - --disable=bugprone-easily-swappable-parameters
+  - --disable=bugprone-narrowing-conversions
+  - --disable=clang-diagnostic-implicit-int-float-conversion
+  - --disable=clang-diagnostic-float-conversion
+  - --disable=clang-diagnostic-implicit-int-conversion
+  - --skip=${GITHUB_WORKSPACE}/.github/config/codechecker.skip
+  - --clean
+
+parse:
+  - --export=html
+  - --output=./html
+  - --skip=${GITHUB_WORKSPACE}/.github/config/codechecker.skip
+  - --trim-path-prefix=${GITHUB_WORKSPACE}/

.github/config/cppcheck

@@ -0,0 +1 @@
+--check-level=exhaustive

.github/workflows/ci_codechecker.yml

@@ -0,0 +1,35 @@
+# SPDX-FileCopyrightText: 2006-2024, Knut Reinert & Freie Universität Berlin
+# SPDX-FileCopyrightText: 2016-2024, Knut Reinert & MPI für molekulare Genetik
+# SPDX-License-Identifier: CC0-1.0
+
+name: CodeChecker
+
+on:
+  schedule:
+    - cron: "0 4 * * THU"
+  workflow_dispatch:
+
+env:
+  TZ: Europe/Berlin
+
+defaults:
+  run:
+    shell: bash -Eexuo pipefail {0}
+
+jobs:
+  build:
+    name: CodeChecker
+    runs-on: ubuntu-latest
+    if: github.repository_owner == 'seqan' || github.event_name == 'workflow_dispatch'
+    container:
+      image: ghcr.io/seqan/clang-19
+      volumes:
+        - /home/runner:/home/runner
+    steps:
+      - name: Run CodeChecker
+        uses: seqan/actions/codechecker@main
+        with:
+          deploy_host: ${{ secrets.DEPLOY_HOST }}
+          deploy_user: ${{ secrets.DEPLOY_USER }}
+          deploy_ssh_key: ${{ secrets.DEPLOY_SSH_KEY }}
+          deploy_path: ${{ secrets.DEPLOY_CODECHECKER_PATH }}

include/raptor/argument_parsing/formatted_bytes.hpp

@@ -39,8 +39,8 @@ namespace raptor
         decimal >>= (iterations - 1u) * 10u;                  // Shift to next smallest unit, e.g. 800MiB
         decimal = decimal * 1000u / 1024u;                    // Account for using decimal system, i.e. 800MiB != 0.8GiB
         size_t const diff{decimal - (decimal / 100u) * 100u}; // We want to round up to 1 decimal position
-        uint32_t const round_up{diff >= 50u};
-        decimal += round_up * 100u - diff;
+        bool const round_up{diff >= 50u};
+        decimal += static_cast<size_t>(round_up * 100u) - diff;
         decimal /= 100u;
         return decimal;
     };

include/raptor/argument_parsing/formatted_index_size.hpp

@@ -19,7 +19,7 @@
 namespace raptor
 {
 
-[[nodiscard]] inline size_t index_size_in_KiB(std::filesystem::path index_path, uint8_t const parts)
+[[nodiscard]] inline size_t index_size_in_KiB(std::filesystem::path const & index_path, uint8_t const parts)
 {
     size_t index_size_in_bytes{};
     if (parts == 1u)
@@ -37,7 +37,7 @@ namespace raptor
     return index_size_in_bytes >> 10;
 }
 
-[[nodiscard]] inline std::string formatted_index_size(std::filesystem::path index_path, uint8_t const parts)
+[[nodiscard]] inline std::string formatted_index_size(std::filesystem::path const & index_path, uint8_t const parts)
 {
     return formatted_bytes(index_size_in_KiB(index_path, parts) << 10);
 }

include/raptor/argument_parsing/validators.hpp

@@ -48,7 +48,11 @@ static inline std::vector<std::string> combined_extensions{
             {
                 result.push_back(sequence_extension);
                 for (auto && compression_extension : compression_extensions)
-                    result.push_back(sequence_extension + std::string{'.'} + compression_extension);
+                {
+                    result.push_back(sequence_extension);
+                    result.back() += '.';
+                    result.back() += compression_extension;
+                }
             }
         }
         return result;

include/raptor/file_reader.hpp

@@ -18,7 +18,7 @@
 namespace raptor
 {
 
-enum class file_types
+enum class file_types : uint8_t
 {
     sequence,
     minimiser

include/raptor/prepare/cutoff.hpp

@@ -44,22 +44,25 @@ class cutoff
         {
         case cutoff_kinds::filesize_dependent:
             return impl(filename);
-        default:
-        {
-            assert(cutoff_kind == cutoff_kinds::fixed);
+        case cutoff_kinds::fixed:
             return fixed_cutoff;
-        }
+        default: // GCOVR_EXCL_LINE
+#ifndef NDEBUG
+            assert(false); // GCOVR_EXCL_LINE
+#else
+            __builtin_unreachable();
+#endif
         }
     }
 
-    static inline bool file_is_compressed(std::filesystem::path const & filepath)
+    static inline bool file_is_compressed(std::filesystem::path const & filepath) noexcept
     {
         std::filesystem::path const extension = filepath.extension();
         return extension == ".gz" || extension == ".bgzf" || extension == ".bz2";
     }
 
 private:
-    enum class cutoff_kinds
+    enum class cutoff_kinds : uint8_t
     {
         fixed,
         filesize_dependent
@@ -80,7 +83,7 @@ class cutoff
                                                                  1'073'741'824ULL,
                                                                  3'221'225'472ULL};
 
-    uint8_t impl(std::filesystem::path const & filename) const
+    uint8_t impl(std::filesystem::path const & filename) const noexcept
     {
         bool const is_compressed = file_is_compressed(filename);
         bool const is_fasta = check_for_fasta_format(filename);
@@ -103,9 +106,10 @@ class cutoff
         return cutoff;
     }
 
-    static inline bool
-    check_for_fasta_format(std::filesystem::path const & filepath,
-                           std::vector<std::string> const & valid_extensions = seqan3::format_fasta::file_extensions)
+    // NOLINTNEXTLINE(bugprone-exception-escape)
+    static inline bool check_for_fasta_format(
+        std::filesystem::path const & filepath,
+        std::vector<std::string> const & valid_extensions = seqan3::format_fasta::file_extensions) noexcept
     {
         std::string const extension = file_is_compressed(filepath) ? filepath.stem() : filepath.extension();
 

include/raptor/threshold/logspace.hpp

@@ -15,7 +15,7 @@
 namespace raptor::logspace
 {
 
-constexpr double ln_2{0.693147180559945309417232121458176568L};
+constexpr double ln_2{0.693147180559945};
 constexpr double negative_inf{-std::numeric_limits<double>::infinity()};
 
 //!\brief The log of a sum of two log terms.
@@ -31,7 +31,7 @@ constexpr double negative_inf{-std::numeric_limits<double>::infinity()};
 template <typename... types>
 [[nodiscard]] double add(double const log_x, double const log_y, types... logs) noexcept
 {
-    return add(add(log_y, log_x), logs...);
+    return add(add(log_x, log_y), logs...);
 }
 
 //!\brief The log of a difference of two log terms. (log_x - log_y)

include/raptor/threshold/threshold.hpp

@@ -30,7 +30,7 @@ class threshold
     size_t get(size_t const minimiser_count) const noexcept;
 
 private:
-    enum class threshold_kinds
+    enum class threshold_kinds : uint8_t
     {
         probabilistic,
         lemma,

src/argument_parsing/compute_bin_size.cpp

@@ -31,13 +31,13 @@ size_t kmer_count_from_minimiser_files(std::vector<std::vector<std::string>> con
     size_t max_filesize{};
     std::filesystem::path biggest_file{};
 
-    auto callback = [&callback_mutex, &biggest_file, &max_filesize](std::filesystem::path const path, size_t const size)
+    auto callback = [&callback_mutex, &biggest_file, &max_filesize](std::filesystem::path path, size_t const size)
     {
         std::lock_guard<std::mutex> guard{callback_mutex};
         if (size > max_filesize)
         {
             max_filesize = size;
-            biggest_file = path;
+            biggest_file = std::move(path);
         }
     };
 
@@ -61,7 +61,7 @@ size_t kmer_count_from_minimiser_files(std::vector<std::vector<std::string>> con
             }
         }
 
-        callback(biggest_file, max_filesize);
+        callback(std::move(biggest_file), max_filesize);
     };
 
     call_parallel_on_bins(worker, bin_path, threads);

src/raptor.cpp

@@ -48,7 +48,7 @@ int main(int argc, char ** argv)
         if (sub_parser.info.app_name == std::string_view{"Raptor-upgrade"})
             raptor::upgrade_parsing(sub_parser);
     }
-    catch (sharg::parser_error const & ext)
+    catch (std::exception const & ext)
     {
         std::cerr << "[Error] " << ext.what() << '\n';
         std::exit(-1);

src/threshold/threshold.cpp

@@ -14,7 +14,7 @@ namespace raptor::threshold
 
 threshold::threshold(threshold_parameters const & arguments)
 {
-    uint8_t const kmer_size{arguments.shape.size()};
+    size_t const kmer_size{arguments.shape.size()};
     size_t const kmers_per_window = arguments.window_size - kmer_size + 1;
 
     if (!std::isnan(arguments.percentage))
@@ -48,13 +48,18 @@ size_t threshold::get(size_t const minimiser_count) const noexcept
         return kmer_lemma;
     case threshold_kinds::percentage:
         return std::max<size_t>(1u, minimiser_count * threshold_percentage);
-    default:
+    case threshold_kinds::probabilistic:
     {
-        assert(threshold_kind == threshold_kinds::probabilistic);
         size_t const index = std::clamp(minimiser_count, minimal_number_of_minimizers, maximal_number_of_minimizers)
                            - minimal_number_of_minimizers;
         return std::max<size_t>(1u, precomp_thresholds[index] + precomp_correction[index]);
     }
+    default: // GCOVR_EXCL_LINE
+#ifndef NDEBUG
+        assert(false); // GCOVR_EXCL_LINE
+#else
+        __builtin_unreachable();
+#endif
     }
 }