LF-single-node script corruption in Virgin.root (embedded LF, zero CR); webBrowser.openURL root cause, #855 unblocker #881

Rationale

LF-single-node script corruption in Virgin.root — a THIRD stored-damage mechanism, distinct from #866's scanner bug and #877's stale DB. Signature: entire script body crammed into ONE outline node with embedded LF (0x0A) bytes and zero CR — a .ut import path that skipped LF→CR normalization. This is the ROOT CAUSE of #865's webBrowser.openURL failure. Repair is cheap and proven: reinstall once via script.newScriptObject (which does LF→CR normalization first-thing) repairs both affected scripts, persists across fileMenu.save + fresh process, verified behaviorally. Works on pre-#866-fix binaries too (no #866 dependency). This is the #855 unblocker.

Current State

LF-node corruption identified (2026-08-10):

• Entire script body in ONE outline node with embedded LF (0x0A) bytes
• Zero CR bytes (native UserTalk line ending is CR 0x0D)
• Signature of .ut import that skipped LF→CR normalization

Confirmed victims:

1. system.verbs.builtins.webBrowser.openURL (1 node, 26 LF) — THIS IS webBrowser.openUrl fails to compile headless; blocks first-run finishInstall flow #865's root cause
2. string.isValidUrl (1 node, 15 LF)

Repair verification (proven, 2026-08-10):

• Reinstall via script.newScriptObject (LF→CR normalization is first-line operation)
• Both scripts repair in place
• Persist across fileMenu.save + fresh process restart
• Behavioral verification: isValidUrl returns true (functional, not just syntax)
• Works on pre-langscan comment-strip scanner ignored single-quoted char constants: 0xC7 inside 'x' literals corrupts scripts cumulatively on install (was misattributed to newScriptObject size/reinstall) #866-fix binaries (no langscan comment-strip scanner ignored single-quoted char constants: 0xC7 inside 'x' literals corrupts scripts cumulatively on install (was misattributed to newScriptObject size/reinstall) #866 dependency on the fix)

Impact:

• webBrowser.openURL broken (blocks Virgin.root ships firstRootRun=false, breaking firstRootRun contract; 9 integration tests depend on skipped init #855 first-run finishInstall)
• string.isValidUrl broken (utility verb unavailable)
• Root cause now identified; unblocker for Virgin.root ships firstRootRun=false, breaking firstRootRun contract; 9 integration tests depend on skipped init #855 is dispatchable now

Desired Outcome

1. Repair the 2 known LF-node scripts (openURL, isValidUrl) via script.newScriptObject
2. Scan Virgin.root corpus for stored-node-text-containing-LF to size the class beyond 2 known
3. Identify provenance (which import path skipped LF→CR normalization)
4. Update webBrowser.openUrl fails to compile headless; blocks first-run finishInstall flow #865 to reflect actual root cause (LF corruption, not missing headless implementation)
5. Unblock Virgin.root ships firstRootRun=false, breaking firstRootRun contract; 9 integration tests depend on skipped init #855 (first-run flow can proceed after repair + flag flip)

Recommendation

Immediate (dispatchable now, no #866 dependency):

For each of webBrowser.openURL and string.isValidUrl:
1. Read current script via protocol: string(@system.verbs.builtins.webBrowser.openURL)
2. Reinstall via script.newScriptObject (normalizes LF→CR automatically)
3. Verify: script compiles in place, behaviorally correct (e.g., isValidUrl("foo@bar.com") returns true)
4. Persist via fileMenu.save

Follow-up:

1. Corpus-wide scan for stored-node-text-containing-LF
2. Identify count and names of all affected scripts
3. Batch repair via script.newScriptObject
4. Investigate provenance (trace which import caused LF→CR skip)

Exit Criteria

• webBrowser.openURL reinstalled and compiles in place
• string.isValidUrl reinstalled and compiles in place
• Both verified behaviorally (openURL callable, isValidUrl returns correct values)
• Corpus scan complete: identify all scripts with embedded LF
• Provenance identified: which import path skipped normalization
• webBrowser.openUrl fails to compile headless; blocks first-run finishInstall flow #865 updated: root cause is LF corruption, not missing headless verb
• Virgin.root ships firstRootRun=false, breaking firstRootRun contract; 9 integration tests depend on skipped init #855 can proceed (first-run unblocked after repair + flag fix)

Related Context

• Found by: langscan comment-strip scanner ignored single-quoted char constants: 0xC7 inside 'x' literals corrupts scripts cumulatively on install (was misattributed to newScriptObject size/reinstall) #866 investigation, 2026-08-10
• Related issues: webBrowser.openUrl fails to compile headless; blocks first-run finishInstall flow #865 (webBrowser.openURL failure — now root-caused), Virgin.root ships firstRootRun=false, breaking firstRootRun contract; 9 integration tests depend on skipped init #855 (first-run flow, now unblocked), 12 Virgin.root scripts genuinely broken in DB (fail to compile in place); webBrowser.openURL + cleanRoot block #855 #879 (broken scripts — some are LF-node corruption, not genuine DB corruption), langscan comment-strip scanner ignored single-quoted char constants: 0xC7 inside 'x' literals corrupts scripts cumulatively on install (was misattributed to newScriptObject size/reinstall) #866 (scanner bug, distinct from this corruption)
• Code: Virgin.root scripts (2 confirmed LF-node victims), import/normalization path
• Repair mechanism: script.newScriptObject (LF→CR normalization), fileMenu.save persistence
• Verification: behavioral tests (isValidUrl returns correct result)
• Dependency: NONE — repair works on pre-langscan comment-strip scanner ignored single-quoted char constants: 0xC7 inside 'x' literals corrupts scripts cumulatively on install (was misattributed to newScriptObject size/reinstall) #866-fix binaries
• Corpus: need to identify full scope of LF-node corruption

Distinct from Other Issues

• NOT langscan comment-strip scanner ignored single-quoted char constants: 0xC7 inside 'x' literals corrupts scripts cumulatively on install (was misattributed to newScriptObject size/reinstall) #866 (scanner/parser bug in compile path) — this is stored data corruption
• NOT databases/Frontier.root is stale (pre-#621, doubled-brace corruption); source of #866 false diagnosis #877 (stale DB) — this is pre-existing damage in current Virgin.root
• NOT 65 Virgin.root scripts fail round-trip idempotence (string->newScriptObject); latent normalization divergence #878 (round-trip non-idempotence) — this is broken persistence, not normalization drift
• NOT 12 Virgin.root scripts genuinely broken in DB (fail to compile in place); webBrowser.openURL + cleanRoot block #855 #879 (genuine compile failures) — this is fixable-by-reinstall, proven cheap repair

Methodology Note

The cheap repair (script.newScriptObject does LF→CR normalization automatically) proves the issue: if reinstall fixes it, the damage is in the stored bytes, not in the code path. Behavioral verification (isValidUrl returns correct result) proves the fix is functional, not just syntactic.