[Git][ghc/ghc][wip/T22686] gitlab: Collect metadata about binary distributions

[Ben Gamari (@bgamari)]

Ben Gamari pushed to branch wip/T22686 at Glasgow Haskell Compiler / GHC


Commits:
083f9d8e by Ben Gamari at 2023-01-27T14:12:16-05:00
gitlab: Collect metadata about binary distributions

Fixes #22686.

- - - - -


3 changed files:

- + .gitlab/bindist_metadata.py
- .gitlab/ci.sh
- .gitlab/gen_ci.hs


Changes:

=====================================
.gitlab/bindist_metadata.py
=====================================
@@ -0,0 +1,141 @@
+#!/usr/bin/env python3
+
+import sys
+import os
+import shutil
+import re
+import ast
+from pathlib import Path
+import subprocess
+import json
+from typing import Dict, List, Set
+
+def run(args: List[str]) -> str:
+    return subprocess.check_output(args, encoding='UTF-8')
+
+def parse_hadrian_cfg(cfg: str) -> Dict[str,str]:
+    res = {}
+    for l in cfg.split('\n'):
+        if l.startswith('#'):
+            continue
+        elif '=' in l:
+            i = l.find('=')
+            k = l[:i].strip()
+            v = l[i+1:].strip()
+            res[k] = v
+
+    return res
+
+def get_ghc_info(ghc: Path) -> Dict[str,str]:
+    import ast
+    out = run([ghc, '--info'])
+    pairs = ast.literal_eval(out.strip())
+    res = {}
+    for k,v in pairs:
+        if v == 'YES':
+            v = True
+        elif v == 'NO':
+            v = False
+        res[k] = v
+
+    return res
+
+def get_configure_cmdline() -> str:
+    r = Path('config.log').read_text()
+    m = re.search(r'  $ .+', r)
+    return m
+
+def get_dpkg_packages() -> Dict[str, str]:
+    pkgs = {}
+    for l in run(['dpkg-query', '--show']).split('\n'):
+        parts = l.split()
+        if len(parts) == 2:
+            k = parts[0].strip()
+            v = parts[1].strip()
+            pkgs[k] = v
+
+    return pkgs
+
+def get_rpm_packages() -> Dict[str, str]:
+    pkgs = {}
+    for l in run(['rpm', '-qa', '--queryformat=%{NAME} %{VERSION}\n']).split('\n'):
+        parts = l.split()
+        if len(parts) == 2:
+            k = parts[0].strip()
+            v = parts[1].strip()
+            pkgs[k] = v
+
+    return pkgs
+
+def main() -> None:
+    ghc = Path('_build/stage1/bin/ghc')
+    ghc_pkg = Path('_build/stage1/bin/ghc-pkg')
+
+    metadata = {}
+
+    system_config = Path('.') / 'hadrian' / 'cfg' / 'system.config'
+    cfg = parse_hadrian_cfg(system_config.read_text())
+
+    ######
+    # GHC build configuration
+    ######
+    metadata['ghc_version'] = cfg['project-version']
+    metadata['git_commit_id'] = cfg['project-git-commit-id']
+    metadata['tables_next_to_code'] = cfg['tables-next-to-code']
+    metadata['unregisterised'] = cfg['ghc-unregisterised']
+    metadata['build_triple'] = cfg['build-platform']
+    metadata['host_triple'] = cfg['host-platform']
+    metadata['target_triple'] = cfg['target-platform']
+    metadata['build_flavour'] = os.environ.get('BUILD_FLAVOUR')
+    metadata['configure_cmdline'] = get_configure_cmdline()
+
+    ######
+    # Information about the bootstrapping environment
+    ######
+    try:
+        lsb_release = run(['lsb_release'])
+    except:
+        lsb_release = 'unknown'
+
+    metadata['bootstrap_environment'] = {
+        'ghc': run([cfg['system-ghc'], '--version']).split('\n')[0],
+        'cc': run([cfg['system-cc'], '--version']).split('\n')[0],
+        'lsb_release': lsb_release,
+    }
+
+    ######
+    # Information about the bootstrapping environment's packages
+    ######
+    if shutil.which('dpkg'):
+        metadata['bootstrap_environment']['package_system'] = 'dpkg'
+        metadata['bootstrap_environment']['installed_packages'] = get_dpkg_packages()
+    elif shutil.which('rpm'):
+        metadata['bootstrap_environment']['package_system'] = 'rpm'
+        metadata['bootstrap_environment']['installed_packages'] = get_rpm_packages()
+    else:
+        metadata['bootstrap_environment']['package_system'] = 'unknown'
+        metadata['bootstrap_environment']['installed_packages'] = None
+
+    ######
+    # The contents of the compiler's global package database
+    ######
+    def call_ghc_pkg(args: List[str]) -> str:
+        return run([ghc_pkg, '--simple-output'] + args).strip()
+
+    metadata['global_packages'] = {
+        pkg: {
+            'version': call_ghc_pkg(['field', pkg, 'version']),
+            'extra-libraries': call_ghc_pkg(['field', pkg, 'extra-libraries']).split(),
+        }
+        for pkg in call_ghc_pkg(['list', '--names-only']).split()
+    }
+
+    ######
+    # Information about the resulting compiler
+    ######
+    metadata['inplace_ghc_info'] = get_ghc_info(ghc)
+
+    json.dump(metadata, sys.stdout, indent=2)
+
+if __name__ == '__main__':
+    main()


=====================================
.gitlab/ci.sh
=====================================
@@ -580,6 +580,7 @@ function install_bindist() {
 function test_hadrian() {
   check_msys2_deps _build/stage1/bin/ghc --version
   check_release_build
+  $TOP/.gitlab/bindist_metadata.py > metadata.json
 
   # Ensure that statically-linked builds are actually static
   if [[ "${BUILD_FLAVOUR}" = *static* ]]; then


=====================================
.gitlab/gen_ci.hs
=====================================
@@ -681,6 +681,7 @@ job arch opsys buildConfig = NamedJob { name = jobName, jobInfo = Job {..} }
       { junitReport = "junit.xml"
       , expireIn = "2 weeks"
       , artifactPaths = [binDistName arch opsys buildConfig ++ ".tar.xz"
+                        ,"metadata.json"
                         ,"junit.xml"]
       , artifactsWhen = ArtifactsAlways
       }



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/083f9d8ea62aac03879d96008ff906009c4759a7

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/083f9d8ea62aac03879d96008ff906009c4759a7
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20230127/2343b23b/attachment-0001.html>