import hashlib
import json
import os
import pathlib
import shutil
import subprocess
import time

root = pathlib.Path(__file__).resolve().parents[2]
local = root/'.local'
evidence = pathlib.Path(__file__).resolve().parent
out = evidence/'binary-compression-records.jsonl'
assert not out.exists(), 'Refuse to overwrite earlier compression records'
skip = {'toolchain', 'linux-qa', 'cross-linux-qa', 'package-font-tools'}
files = []
for base, dirs, names in os.walk(local, followlinks=False):
    base = pathlib.Path(base)
    dirs[:] = [d for d in dirs if d not in skip and d != 'incremental' and not (base/d).is_symlink()]
    for name in names:
        p = base/name
        if p.suffix.lower() == '.exe' and not p.is_symlink() and p.stat().st_size > 1024*1024:
            assert p.resolve().is_relative_to(local.resolve())
            files.append(p)
files.sort(key=lambda p: p.stat().st_size, reverse=True)
(evidence/'binary-compression-candidates.json').write_text(json.dumps(
    [{'path':p.relative_to(root).as_posix(),'bytes':p.stat().st_size} for p in files],indent=2)+'\n',encoding='utf8')
started=time.time()
free_before=shutil.disk_usage(root).free
print(json.dumps({'state':'running','candidates':len(files),'logicalBytes':sum(p.stat().st_size for p in files),'freeBefore':free_before}),flush=True)
def sha(p):
    with p.open('rb') as f:
        return hashlib.file_digest(f,'sha256').hexdigest()
with out.open('x',encoding='utf8') as records, (evidence/'binary-compression.log').open('x',encoding='utf8') as log:
    for i,p in enumerate(files,1):
        before=sha(p)
        process=subprocess.run(['compact.exe','/C','/EXE:XPRESS8K','/Q',str(p)],stdout=subprocess.PIPE,stderr=subprocess.STDOUT)
        log.write(process.stdout.decode('utf-8',errors='replace'));log.flush()
        after=sha(p)
        record={'path':p.relative_to(root).as_posix(),'bytes':p.stat().st_size,'shaBefore':before,'shaAfter':after,'exitCode':process.returncode}
        records.write(json.dumps(record)+'\n');records.flush()
        assert process.returncode == 0 and before == after, record
        if i%25 == 0:
            print(json.dumps({'completed':i,'total':len(files),'freeBytes':shutil.disk_usage(root).free}),flush=True)
result={'state':'complete','files':len(files),'allHashesUnchanged':True,'freeBefore':free_before,'freeAfter':shutil.disk_usage(root).free,'seconds':time.time()-started,'operation':'NTFS executable compression of generated project binaries; no deletions'}
(evidence/'binary-compression-result.json').write_text(json.dumps(result,indent=2)+'\n',encoding='utf8')
print(json.dumps(result),flush=True)