Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Initial implementation of hictk fix-mcool
- Loading branch information
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,143 @@ | ||
| // Copyright (C) 2023 Roberto Rossini <roberros@uio.no> | ||
| // | ||
| // SPDX-License-Identifier: MIT | ||
|
|
||
| #include <fmt/format.h> | ||
| #include <fmt/std.h> | ||
|
|
||
| #include <CLI/CLI.hpp> | ||
| #include <cassert> | ||
| #include <cstdint> | ||
| #include <string> | ||
| #include <thread> | ||
|
|
||
| #include "hictk/tools/cli.hpp" | ||
| #include "hictk/tools/config.hpp" | ||
|
|
||
| namespace hictk::tools { | ||
|
|
||
| void Cli::make_fix_mcool_subcommand() { | ||
| auto& sc = *_cli.add_subcommand("fix-mcool", "Fix corrupted .mcool files.") | ||
| ->fallthrough() | ||
| ->preparse_callback([this]([[maybe_unused]] std::size_t i) { | ||
| assert(_config.index() == 0); | ||
| _config = FixMcoolConfig{}; | ||
| }); | ||
|
|
||
| _config = FixMcoolConfig{}; | ||
| auto& c = std::get<FixMcoolConfig>(_config); | ||
|
|
||
| // clang-format off | ||
| sc.add_option( | ||
| "input", | ||
| c.path_to_input, | ||
| "Path to a corrupted .mcool file.") | ||
| ->check(IsValidMultiresCoolerFile) | ||
| ->required(); | ||
| sc.add_option( | ||
| "output", | ||
| c.path_to_output, | ||
| "Path where to store the restored .mcool.") | ||
| ->required(); | ||
| sc.add_option( | ||
| "--tmpdir", | ||
| c.tmp_dir, | ||
| "Path to a folder where to store temporary data.") | ||
| ->capture_default_str(); | ||
| sc.add_flag( | ||
| "--skip-balancing", | ||
| c.skip_balancing, | ||
| "Do not recompute or copy balancing weights."); | ||
| sc.add_flag( | ||
| "--check-base-resolution", | ||
| c.check_base_resolution, | ||
| "Check whether the base resolution is corrupted."); | ||
| sc.add_flag( | ||
| "--in-memory", | ||
| c.in_memory, | ||
| "Store all interactions in memory while balancing (greatly improves performance).") | ||
| ->capture_default_str(); | ||
| sc.add_option( | ||
| "--chunk-size", | ||
| c.chunk_size, | ||
| "Number of interactions to process at once during balancing.\n" | ||
| "Ignored when using --in-memory.") | ||
| ->check(CLI::PositiveNumber) | ||
| ->capture_default_str(); | ||
| sc.add_option( | ||
| "-v,--verbosity", | ||
| c.verbosity, | ||
| "Set verbosity of output to the console.") | ||
| ->check(CLI::Range(1, 4)) | ||
| ->capture_default_str(); | ||
| sc.add_option( | ||
| "-t,--threads", | ||
| c.threads, | ||
| "Maximum number of parallel threads to spawn (only applies to balancing stage).") | ||
| ->check(CLI::Range(std::uint32_t(1), std::thread::hardware_concurrency())) | ||
| ->capture_default_str(); | ||
| sc.add_option( | ||
| "-l,--compression-level", | ||
| c.zstd_compression_lvl, | ||
| "Compression level used to compress temporary files using ZSTD (only applies to balancing stage).") | ||
| ->check(CLI::Range(0, 19)) | ||
| ->capture_default_str(); | ||
| sc.add_flag( | ||
| "-f,--force", | ||
| c.force, | ||
| "Overwrite existing files (if any).") | ||
| ->capture_default_str(); | ||
| // clang-format on | ||
| } | ||
|
|
||
| void Cli::validate_fix_mcool_subcommand() const { | ||
| const auto& c = std::get<FixMcoolConfig>(_config); | ||
| std::vector<std::string> errors; | ||
| std::vector<std::string> warnings{}; | ||
|
|
||
| if (!c.force && std::filesystem::exists(c.path_to_output)) { | ||
| errors.emplace_back(fmt::format( | ||
| FMT_STRING("Refusing to overwrite file {}. Pass --force to overwrite."), c.path_to_output)); | ||
| } | ||
|
|
||
| if (c.skip_balancing) { | ||
| const auto* sc = _cli.get_subcommand("fix-mcool"); | ||
| if (!sc->get_option("--tmpdir")->empty()) { | ||
| warnings.emplace_back("option --tmpdir is ignored when --skip-balancing is provided."); | ||
| } | ||
| if (!sc->get_option("--in-memory")->empty()) { | ||
| warnings.emplace_back("option --in-memory is ignored when --skip-balancing is provided."); | ||
| } | ||
| if (!sc->get_option("--compression-level")->empty()) { | ||
| warnings.emplace_back( | ||
| "option --compression-level is ignored when --skip-balancing is provided."); | ||
| } | ||
| if (!sc->get_option("--chunk-size")->empty()) { | ||
| warnings.emplace_back("option --chunk-size is ignored when --skip-balancing is provided."); | ||
| } | ||
| if (!sc->get_option("--threads")->empty()) { | ||
| warnings.emplace_back("option --threads is ignored when --skip-balancing is provided."); | ||
| } | ||
| } | ||
|
|
||
| for (const auto& w : warnings) { | ||
| SPDLOG_WARN(FMT_STRING("{}"), w); | ||
| } | ||
|
|
||
| if (!errors.empty()) { | ||
| throw std::runtime_error(fmt::format( | ||
| FMT_STRING( | ||
| "The following error(s) where encountered while validating CLI arguments:\n - {}"), | ||
| fmt::join(errors, "\n - "))); | ||
| } | ||
| } | ||
|
|
||
| void Cli::transform_args_fix_mcool_subcommand() { | ||
| auto& c = std::get<FixMcoolConfig>(_config); | ||
|
|
||
| // in spdlog, high numbers correspond to low log levels | ||
| assert(c.verbosity > 0 && c.verbosity < 5); | ||
| c.verbosity = static_cast<std::uint8_t>(spdlog::level::critical) - c.verbosity; | ||
| } | ||
|
|
||
| } // namespace hictk::tools | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.