"""Independent reconstruction from primary PDFs and released sequences. No claimant code is loaded. Native MASS defaults run before witness evaluation. """ import csv import hashlib import json import os from pathlib import Path import re import subprocess from decimal import Decimal, localcontext from pypdf import PdfReader ROOT = Path(__file__).resolve().parent os.chdir(ROOT) E = ROOT / 'evidence' text = '\n'.join(p.extract_text() for p in PdfReader(E/'shah_fresh.pdf').pages) entries = list(re.finditer(r'(?m)^([A-Z])\s+([ACGU]{3})\s+', text)) rate = {} for index, match in enumerate(entries): end = entries[index+1].start() if index+1 < len(entries) else len(text) values = re.findall(r'(? 0 rh = ROOT/'runtime/R-expanded/R-fw.pkg/Payload/R.framework/Versions/4.6/Resources' env = dict(os.environ, R_HOME=str(rh), DYLD_LIBRARY_PATH=str(rh/'lib'), R_LIBS_USER=str(ROOT/'runtime/Rlib'), TMPDIR=str(ROOT/'tmp'), OMP_NUM_THREADS='1', OPENBLAS_NUM_THREADS='1', VECLIB_MAXIMUM_THREADS='1') run = subprocess.run([str(rh/'bin/exec/R'),'--vanilla','--slave','-f','standard_referee.R'], env=env, capture_output=True,text=True) (ROOT/'results/native_R.log').write_text(run.stdout+run.stderr) assert run.returncode == 0, (run.returncode, run.stderr) r_order = [int(s)-1 for s in Path('native_order.txt').read_text().split()] assert r_order == order, 'Native and exact dissimilarity orders differ' def evaluate(filename, digits): with localcontext() as ctx: ctx.prec = digits with open(filename,newline='') as f: xy = [[Decimal(r['x']),Decimal(r['y'])] for r in csv.DictReader(f)] assert len(xy) == 23 y = [sum((xy[i][k]-xy[j][k])**2 for k in range(2)).sqrt() for i,j in pairs] y = [y[k] for k in order] blocks = [] for v in y: blocks.append((v,1)) while len(blocks)>1 and blocks[-2][0]*blocks[-1][1] > blocks[-1][0]*blocks[-2][1]: b,n = blocks.pop(); a,m = blocks.pop() blocks.append((a+b,m+n)) fit = [a/n for a,n in blocks for _ in range(n)] stress = 100*(sum((a-b)**2 for a,b in zip(y,fit))/sum(a*a for a in y)).sqrt() return {'stress':str(stress),'blocks':len(blocks)} result = {'sequences':23,'codons':387,'pairs':253,'unique_dissimilarities':len(set(sq)), 'exact_and_native_order_agree':True, 'sha256':{p.name:hashlib.sha256(p.read_bytes()).hexdigest() for p in [E/'paper.pdf',E/'shah_fresh.pdf',E/'sequences.csv',E/'witness.csv']}} for label,filename in [('native_default','native_default.csv'),('witness',E/'witness.csv')]: result[label] = {str(d):evaluate(filename,d) for d in [70,110]} Path('results/recomputed.json').write_text(json.dumps(result,indent=2)+'\n') print(json.dumps(result,indent=2))