File size: 7,696 Bytes
626e1aa d1292d8 626e1aa d1292d8 626e1aa d1292d8 626e1aa d1292d8 626e1aa d1292d8 626e1aa d1292d8 626e1aa d1292d8 626e1aa | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 | """Download the raw source records for every compound in ``compounds_list.py`` into ``references/spectral_raw/<id>/``.
Sources (all public web services, fetched politely with a delay and cached; re-runs skip files already on disk):
* NIST Chemistry WebBook (https://webbook.nist.gov), keyed by CAS:
gc.html Mask=2000 Kovats / Van den Dool retention-index tables (nonpolar + polar columns)
thermo.html Mask=4 phase-change data (boiling point, enthalpy of vaporization)
ir.html Mask=80 list of available IR spectra; each is then saved as ir_<index>.jdx (JCAMP-DX)
uv.html Mask=400 list of UV/Vis spectra; each saved as uv_<index>.jdx
ms_0.jdx Type=Mass NIST/EPA/NIH electron-ionisation mass spectrum (peak list)
* PubChem (https://pubchem.ncbi.nlm.nih.gov), keyed by CAS via the name resolver:
pubchem_props.json CID, formula, MW, SMILES, XLogP, IUPAC name
pubchem_nmr_1h.json PUG-View section "1H NMR Spectra" (HMDB peak lists re-hosted by PubChem)
pubchem_nmr_13c.json PUG-View section "13C NMR Spectra"
pubchem_uv.json PUG-View section "UV Spectra" (lambda-max text records)
Usage: python fetch_sources.py [--ids a,b,c] [--delay 0.6] [--force] [--skip-unlisted]
"""
from __future__ import annotations
import argparse
import json
import re
import sys
import time
from pathlib import Path
import requests
sys.path.insert(0, str(Path(__file__).resolve().parent))
from compounds_list import COMPOUNDS # noqa: E402
ROOT = Path(__file__).resolve().parents[2]
RAW = ROOT / "references" / "spectral_raw"
NIST = "https://webbook.nist.gov/cgi/cbook.cgi"
PUBCHEM = "https://pubchem.ncbi.nlm.nih.gov/rest"
HEADERS = {"User-Agent": "Chemistry-Sim spectral-library fetcher (academic lab simulator; contact: project maintainer)"}
MAX_SPECTRA_PER_TYPE = 4
def nist_id(cas: str) -> str:
return "C" + cas.replace("-", "")
def get(session: requests.Session, url: str, params: dict | None, dest: Path, delay: float, force: bool, log) -> bytes | None:
if dest.exists() and not force:
return dest.read_bytes()
for attempt in range(3):
try:
r = session.get(url, params=params, headers=HEADERS, timeout=60)
time.sleep(delay)
if r.status_code == 200:
dest.write_bytes(r.content)
return r.content
if r.status_code in (404,):
dest.write_bytes(b"")
return b""
log(f" HTTP {r.status_code} for {r.url}; retry {attempt + 1}")
except requests.RequestException as exc:
log(f" {exc!r}; retry {attempt + 1}")
time.sleep(3.0 * (attempt + 1))
return None
def nist_listed_masks(page: bytes) -> set[str]:
"""Masks of the data sections a NIST WebBook compound page links to ('Other data available'), e.g. {'4', '80', '200'}."""
if b"Registry Number Not Found" in page or b'<h1 id="Top">' not in page:
return set()
return {m.decode() for m in re.findall(rb"Mask=([0-9A-F]+)#", page)}
def jcamp_ok(blob: bytes | None) -> bool:
return bool(blob) and b"Spectrum not found" not in blob[:200] and b"##XYDATA" in blob or (bool(blob) and b"##PEAK TABLE" in blob)
def fetch_compound(session, row, delay, force, log, skip_unlisted=False):
sid, name_en, name_zh, cas, cls = row
d = RAW / sid
d.mkdir(parents=True, exist_ok=True)
cid = nist_id(cas)
# --- NIST pages
gc_page = get(session, NIST, {"ID": cid, "Mask": "2000", "Units": "SI"}, d / "gc.html", delay, force, log) or b""
listed = nist_listed_masks(gc_page) if skip_unlisted and gc_page else None
def wanted(mask: str, dest: Path) -> bool:
"""With --skip-unlisted, do not request data NIST's own 'Other data available' list says it does not hold."""
if listed is None or mask in listed or dest.exists():
return True
log(f" skip {dest.name}: Mask={mask} not listed on the NIST page")
return False
if wanted("4", d / "thermo.html"):
get(session, NIST, {"ID": cid, "Mask": "4", "Units": "SI"}, d / "thermo.html", delay, force, log)
ir_page = (get(session, NIST, {"ID": cid, "Mask": "80", "Units": "SI"}, d / "ir.html", delay, force, log) or b"") if wanted("80", d / "ir.html") else b""
uv_page = (get(session, NIST, {"ID": cid, "Mask": "400", "Units": "SI"}, d / "uv.html", delay, force, log) or b"") if wanted("400", d / "uv.html") else b""
for kind, page, typ in (("ir", ir_page, "IR"), ("uv", uv_page, "UVVis")):
indices = sorted({int(m) for m in re.findall(rb"Index=(\d+)", page)})[:MAX_SPECTRA_PER_TYPE]
for idx in indices:
get(session, NIST, {"JCAMP": cid, "Index": str(idx), "Type": typ}, d / f"{kind}_{idx}.jdx", delay, force, log)
if wanted("200", d / "ms_0.jdx"):
get(session, NIST, {"JCAMP": cid, "Index": "0", "Type": "Mass"}, d / "ms_0.jdx", delay, force, log)
# --- PubChem
props = "MolecularFormula,MolecularWeight,SMILES,ConnectivitySMILES,XLogP,IUPACName,Title"
blob = get(session, f"{PUBCHEM}/pug/compound/name/{cas}/property/{props}/JSON", None, d / "pubchem_props.json", delay, force, log)
pc_cid = None
try:
pc_cid = json.loads(blob)["PropertyTable"]["Properties"][0]["CID"]
except Exception:
# fall back to the English name
blob = get(session, f"{PUBCHEM}/pug/compound/name/{name_en}/property/{props}/JSON", None, d / "pubchem_props.json", delay, True, log)
try:
pc_cid = json.loads(blob)["PropertyTable"]["Properties"][0]["CID"]
except Exception:
log(f" PubChem: no CID for {sid} ({cas} / {name_en})")
if pc_cid:
for heading, fname in (("1H NMR Spectra", "pubchem_nmr_1h.json"), ("13C NMR Spectra", "pubchem_nmr_13c.json"), ("UV Spectra", "pubchem_uv.json")):
get(session, f"{PUBCHEM}/pug_view/data/compound/{pc_cid}/JSON", {"heading": heading}, d / fname, delay, force, log)
have = sorted(p.name for p in d.iterdir() if p.stat().st_size > 0)
log(f" {sid}: {len(have)} files ({', '.join(n for n in have if n.endswith('.jdx'))})")
def main():
ap = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
ap.add_argument("--ids", default="", help="comma-separated subset of substance ids")
ap.add_argument("--delay", type=float, default=0.6, help="seconds to wait after every request")
ap.add_argument("--force", action="store_true", help="re-download files already cached")
ap.add_argument("--skip-unlisted", action="store_true",
help="skip thermo/IR/UV/MS requests for data types the compound's NIST page (gc.html) does not list (fewer requests; same result)")
args = ap.parse_args()
rows = COMPOUNDS if not args.ids else [r for r in COMPOUNDS if r[0] in set(args.ids.split(","))]
RAW.mkdir(parents=True, exist_ok=True)
log_path = RAW / "fetch_log.txt"
with open(log_path, "a", encoding="utf-8") as lf:
def log(msg):
print(msg, flush=True)
lf.write(msg + "\n"); lf.flush()
log(f"=== fetch {time.strftime('%Y-%m-%d %H:%M:%S')}: {len(rows)} compounds")
with requests.Session() as session:
for i, row in enumerate(rows, 1):
log(f"[{i}/{len(rows)}] {row[0]} ({row[3]})")
try:
fetch_compound(session, row, args.delay, args.force, log, skip_unlisted=args.skip_unlisted)
except Exception as exc: # keep going; the builder reports what is missing
log(f" FAILED {row[0]}: {exc!r}")
log("=== done")
if __name__ == "__main__":
main()
|