Open-source release: pipeline code, raw logs, derived data, licences, tests

Adds src/ (scripts, raw measurement logs, derived data, report source),
tests/, LICENSE (MIT, code) and LICENSE-DATA.md (CC BY 4.0 for the survey's
data; third-party terms). The report is updated to the current site; the
v1 and v2 data hand-off files are unchanged.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

src/vps_analyze.py

@@ -0,0 +1,45 @@
+"""Turn our own servers' ping logs into live-grid cells.
+
+Each server is pinned to the RIPE anchor in its own city, with its ping to that
+anchor as the allowance (the same rule as the Globalping probes). Server-to-
+server and server-to-anchor measurements then become cells between anchor
+cities, so they share the cable routes already computed for those cities.
+Honolulu has no anchor; it is its own site at the data-centre city (within
+about 1 km of the coordinates used). Every pair gets the page's 25 km slack.
+
+  python3 vps_analyze.py      -> data/vps_cells.json
+"""
+import glob, json, os, re, statistics
+from collections import defaultdict
+from measure import TARGETS
+
+HERE = os.path.dirname(os.path.abspath(__file__))
+STATE = json.load(open(os.path.join(HERE, "data/vps/state.json")))
+ANCHOR = {t[2]: t[0] for t in TARGETS}                 # anchor fqdn -> city
+LOCAL = {"syd": "Sydney", "mel": "Melbourne", "jnb": "Johannesburg", "scl": "Santiago", "sao": "São Paulo",
+         "lax": "Los Angeles", "icn": "Seoul"}         # server -> city of the anchor it is pinned to
+RX = re.compile(r"^(\S+) (\S+) (\S+) .*?(\d+) packets transmitted, (\d+) received.*?= ([\d.]+)/([\d.]+)/([\d.]+)/([\d.]+) ms")
+
+runs = defaultdict(list)                                # (src code, target name) -> per-run minimum RTTs
+for n in STATE["nodes"]:
+    f = os.path.join(HERE, "data/vps", n["code"], "ping.log")
+    if not os.path.exists(f):
+        continue
+    for line in open(f):
+        m = RX.match(line)
+        if m:
+            runs[(n["code"], m.group(2))].append(float(m.group(6)))
+
+site = lambda code_or_target: LOCAL.get(code_or_target) or ANCHOR.get(code_or_target) or ("Honolulu" if code_or_target == "hnl" else None)
+local = {c: min(runs[(c, a)]) for c, city in LOCAL.items() for a, ac in ANCHOR.items() if ac == city and runs.get((c, a))}
+
+cells = []
+for (src, tgt), mins in runs.items():
+    a, b = site(src), site(tgt)
+    if not a or not b or a == b:
+        continue
+    allow = local.get(src, 0.0) + (local.get(tgt, 0.0) if tgt in LOCAL else 0.0)
+    cells.append(dict(src=src, tgt=tgt, a=a, b=b, rtt=round(min(mins), 2), median=round(statistics.median(mins), 2),
+                      runs=len(mins), allow=round(allow, 2), slack_km=25))
+json.dump(dict(cells=cells, local=local, nodes=STATE["nodes"]), open(os.path.join(HERE, "data/vps_cells.json"), "w"), indent=1)
+print(f"{len(cells)} cells from {len(runs)} server-target series; local allowances: { {k: round(v, 2) for k, v in local.items()} }")