Marco Zecchini
/
Example_RTOS
Rtos API example
mbed-os/tools/test/examples/examples_lib.py
- Committer:
- marcozecchini
- Date:
- 2019-02-23
- Revision:
- 0:9fca2b23d0ba
File content as of revision 0:9fca2b23d0ba:
""" Import and bulid a bunch of example programs This library includes functions that are shared between the examples.py and the update.py modules. """ import os from os.path import dirname, abspath, basename import os.path import sys import subprocess from shutil import rmtree from sets import Set ROOT = abspath(dirname(dirname(dirname(dirname(__file__))))) sys.path.insert(0, ROOT) from tools.build_api import get_mbed_official_release from tools.targets import TARGET_MAP from tools.export import EXPORTERS SUPPORTED_TOOLCHAINS = ["ARM", "IAR", "GCC_ARM", "ARMC6"] SUPPORTED_IDES = [exp for exp in EXPORTERS.keys() if exp != "cmsis" and exp != "zip"] def print_list(lst): """Prints to screen the contents of a list Args: lst - a list of any type, to be displayed """ if lst: for thing in lst: print("# %s" % thing) def print_category(results, index, message): summary = [example for key, summ in results.iteritems() for example in summ[index]] if all(len(s) == 0 for s in summary): return print("#") print("#" * 80) print("# %s" % message) print("#" * 80) split_summ = [s.rsplit(" ", 1) for s in summary] print_list(summary) def print_summary(results, export=False): """Prints to screen the results of compiling/exporting combinations of example programs, targets and compile toolchains/IDEs. Args: results - results of the compilation stage. See compile_repos() and export_repos() for details of the format. """ print("#"*80) print("# Examples compilation summary") print("#"*80) print_category(results, 2, "Passed example combinations") second_result = "Failed example combinations" if not export else \ "Failed export example combinations" print_category(results, 3, second_result) if export: print_category(results, 4, "Failed build combinations") print_category(results, 5, "Skipped build combinations") print("#") print("#"*80) def valid_choices(allowed_choices, all_choices): if len(allowed_choices) > 0: return [t for t in all_choices if t in allowed_choices] else: return all_choices def target_cross_toolchain(allowed_targets, allowed_toolchains, features=[]): """Generate pairs of target and toolchains Args: allowed_targets - a list of all possible targets allowed_toolchains - a list of all possible toolchains Kwargs: features - the features that must be in the features array of a target """ for target in allowed_targets: for toolchain in allowed_toolchains: if all(feature in TARGET_MAP[target].features for feature in features): yield target, toolchain def target_cross_ide(allowed_targets, allowed_ides, features=[], toolchains=[]): """Generate pairs of target and ides Args: allowed_targets - a list of all possible targets allowed_ides - a list of all possible IDEs Kwargs: features - the features that must be in the features array of a target """ for target in allowed_targets: for ide in allowed_ides: if (EXPORTERS[ide].is_target_supported(target) and (not toolchains or EXPORTERS[ide].TOOLCHAIN in toolchains) and all(feature in TARGET_MAP[target].features for feature in features)): yield target, ide def get_repo_list(example): """ Returns a list of all the repos and their types associated with the specific example in the json config file. If the key 'test-repo-source' is set to 'mbed', then it will return the mbed section as a list. Otherwise, it will return the single github repo. NOTE: This does not currently deal with multiple examples underneath a github sourced exampe repo. Args: example - Example for which the repo list is requested """ repos = [] if example['test-repo-source'] == 'mbed': for repo in example['mbed']: repos.append({ 'repo': repo, 'type': 'hg' }) else: repos.append({ 'repo': example['github'], 'type': 'git' }) return repos def source_repos(config, examples): """ Imports each of the repos and its dependencies (.lib files) associated with the specific examples name from the json config file. Note if there is already a clone of the repo then it will first be removed to ensure a clean, up to date cloning. Args: config - the json object imported from the file. """ print("\nImporting example repos....\n") for example in config['examples']: for repo_info in get_repo_list(example): name = basename(repo_info['repo']) if name in examples: if os.path.exists(name): print("'%s' example directory already exists. Deleting..." % name) rmtree(name) subprocess.call(["mbed-cli", "import", repo_info['repo']]) def clone_repos(config, examples , retry = 3): """ Clones each of the repos associated with the specific examples name from the json config file. Note if there is already a clone of the repo then it will first be removed to ensure a clean, up to date cloning. Args: config - the json object imported from the file. """ print("\nCloning example repos....\n") for example in config['examples']: for repo_info in get_repo_list(example): name = basename(repo_info['repo']) if name in examples: if os.path.exists(name): print("'%s' example directory already exists. Deleting..." % name) rmtree(name) for i in range(0, retry): if subprocess.call([repo_info['type'], "clone", repo_info['repo']]) == 0: break else: print("ERROR : unable to clone the repo {}".format(name)) def deploy_repos(config, examples): """ If the example directory exists as provided by the json config file, pull in the examples dependencies by using `mbed-cli deploy`. Args: config - the json object imported from the file. """ print("\nDeploying example repos....\n") for example in config['examples']: for repo_info in get_repo_list(example): name = basename(repo_info['repo']) if name in examples: if os.path.exists(name): os.chdir(name) subprocess.call(["mbed-cli", "deploy"]) os.chdir("..") else: print("'%s' example directory doesn't exist. Skipping..." % name) def get_num_failures(results, export=False): """ Returns the number of failed compilations from the results summary Args: results - results summary of the compilation stage. See compile_repos() for details of the format. num_failures """ num_failures = 0 for key, val in results.iteritems(): num_failures = num_failures + len(val[3]) if export: num_failures += len(val[4]) return num_failures def export_repos(config, ides, targets, examples): """Exports and builds combinations of example programs, targets and IDEs. The results are returned in a [key: value] dictionary format: Where key = The example name from the json config file value = a list containing: pass_status, successes, export failures, build_failures, and build_skips where pass_status = The overall pass status for the export of the full set of example programs comprising the example suite. IE they must build and export) True if all examples pass, false otherwise successes = list of examples that exported and built (if possible) If the exporter has no build functionality, then it is a pass if exported export_failures = list of examples that failed to export. build_failures = list of examples that failed to build build_skips = list of examples that cannot build Both successes and failures contain the example name, target and IDE Args: config - the json object imported from the file. ides - List of IDES to export to """ results = {} valid_examples = Set(examples) print("\nExporting example repos....\n") for example in config['examples']: example_names = [basename(x['repo']) for x in get_repo_list(example)] common_examples = valid_examples.intersection(Set(example_names)) if not common_examples: continue export_failures = [] build_failures = [] build_skips = [] successes = [] exported = True pass_status = True if example['export']: for repo_info in get_repo_list(example): example_project_name = basename(repo_info['repo']) os.chdir(example_project_name) # Check that the target, IDE, and features combinations are valid and return a # list of valid combinations to work through for target, ide in target_cross_ide(valid_choices(example['targets'], targets), valid_choices(example['exporters'], ides), example['features'], example['toolchains']): example_name = "{} {} {}".format(example_project_name, target, ide) def status(message): print(message + " %s" % example_name) sys.stdout.flush() status("Exporting") proc = subprocess.Popen(["mbed-cli", "export", "-i", ide, "-m", target]) proc.wait() if proc.returncode: export_failures.append(example_name) status("FAILURE exporting") else: status("SUCCESS exporting") status("Building") try: if EXPORTERS[ide].build(example_project_name, cleanup=False): status("FAILURE building") build_failures.append(example_name) else: status("SUCCESS building") successes.append(example_name) except TypeError: successes.append(example_name) build_skips.append(example_name) os.chdir("..") if len(build_failures+export_failures) > 0: pass_status= False else: exported = False results[example['name']] = [exported, pass_status, successes, export_failures, build_failures, build_skips] return results def compile_repos(config, toolchains, targets, profile, examples): """Compiles combinations of example programs, targets and compile chains. The results are returned in a [key: value] dictionary format: Where key = The example name from the json config file value = a list containing: pass_status, successes, and failures where pass_status = The overall pass status for the compilation of the full set of example programs comprising the example suite. True if all examples pass, false otherwise successes = list of passing examples. failures = list of failing examples. Both successes and failures contain the example name, target and compile chain Args: config - the json object imported from the file. toolchains - List of toolchains to compile for. results - results of the compilation stage. """ results = {} valid_examples = Set(examples) print("\nCompiling example repos....\n") for example in config['examples']: example_names = [basename(x['repo']) for x in get_repo_list(example)] common_examples = valid_examples.intersection(Set(example_names)) if not common_examples: continue failures = [] successes = [] compiled = True pass_status = True if example['compile']: for repo_info in get_repo_list(example): name = basename(repo_info['repo']) os.chdir(name) # Check that the target, toolchain and features combinations are valid and return a # list of valid combinations to work through for target, toolchain in target_cross_toolchain(valid_choices(example['targets'], targets), valid_choices(example['toolchains'], toolchains), example['features']): print("Compiling %s for %s, %s" % (name, target, toolchain)) build_command = ["mbed-cli", "compile", "-t", toolchain, "-m", target, "-v"] if profile: build_command.append("--profile") build_command.append(profile) proc = subprocess.Popen(build_command) proc.wait() example_summary = "{} {} {}".format(name, target, toolchain) if proc.returncode: failures.append(example_summary) else: successes.append(example_summary) os.chdir("..") # If there are any compilation failures for the example 'set' then the overall status is fail. if len(failures) > 0: pass_status = False else: compiled = False results[example['name']] = [compiled, pass_status, successes, failures] return results def update_mbedos_version(config, tag, examples): """ For each example repo identified in the config json object, update the version of mbed-os to that specified by the supplied GitHub tag. This function assumes that each example repo has already been cloned. Args: config - the json object imported from the file. tag - GitHub tag corresponding to a version of mbed-os to upgrade to. """ print("Updating mbed-os in examples to version %s\n" % tag) for example in config['examples']: if example['name'] not in examples: continue for repo_info in get_repo_list(example): update_dir = basename(repo_info['repo']) + "/mbed-os" print("\nChanging dir to %s\n" % update_dir) os.chdir(update_dir) subprocess.call(["mbed-cli", "update", tag, "--clean"]) os.chdir("../..")