Struct TsProgram 

pub struct TsProgram {
    pub stmts: Vec<TsStmt>,
}

A whole generated TypeScript module, as an ordered sequence of top-level statements. Vec<TsStmt>, plain — no richer container yet (P7.6’s own Artefacts { docs: BTreeMap<PathBuf, Document> } is where a project’s documents get keyed; this is one document’s own tree).

Fields

stmts: Vec<TsStmt>

Implementations

impl TsProgram

pub fn new() -> Self

pub fn push(&mut self, stmt: TsStmt)

pub fn verbatim_content(&self) -> Vec<(VerbatimOrigin, &str)>

Every TsStmtKind::Verbatim/TsExpr::VerbatimExpr leaf’s own text, tagged by its VerbatimOrigin — found by walking every statement/expression container in the tree, not just the top level. #1538’s own real gap: nothing before this walked a whole TsProgram to find every opaque leaf nested inside real structured content (a Verbatim inside an if branch, a VerbatimExpr inside a Call’s arguments, …) — the only prior way to inspect a Verbatim’s content was scanning the whole printed document text (crate::verbatim_violations doesn’t distinguish opaque text from a real node’s own printed output), which also flags real, structured nodes that happen to print the substring any (TsType::named("any") residue, tracked separately by xtask’s own ts_any probe) — a false positive for this check, whose whole point is the opaque escape hatch specifically. TsStmtKind is pub(crate), so only code inside this crate can walk it directly; this method is the public surface a caller outside bynk-ts (bynkc’s own tests/tsc_verify.rs) uses instead.

Every container is walked exhaustively (match, no wildcard arm) so a future TsStmt/TsExpr variant fails to compile here until this walker accounts for it — the same “compile-time construct, not a grep” discipline VerbatimOrigin’s own doc already states for itself. TsStmtKind::Raw is deliberately not walked into any deeper than skipped: it carries no VerbatimOrigin at all (ADR arc-c-lower-rs-permanent-exclusion’s own permanent exclusion, a different bucket from Verbatim’s temporary conversion residue), so it has nothing this method could report. TsType positions are never walked — no TsType variant can hold a Verbatim/ VerbatimExpr leaf, so there is nothing to find there.

A real gap this walker exposed (#1538): no VerbatimOrigin::Emit leaf can currently reach a TsProgram returned from bynk-emit at all. Every real bynk-emit/src/emitter/emit.rs construction site builds a TsExpr::VerbatimExpr, immediately prints it (crate::print_expr/crate::print_stmt), and splices the resulting text into a plain String a caller further up wraps in TsStmtKind::Raw — not tagged with a VerbatimOrigin at all, and so invisible to this walker by design (see TsStmtKind::Raw’s own doc). VerbatimExpr’s own doc promises this content is “visible to crate::verbatim_violations”; today that’s only true at the instant of construction, not once the surrounding document is returned — closing it needs a real Arc C conversion slice, not a change here.

Trait Implementations

impl Debug for TsProgram

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for TsProgram

fn default() -> TsProgram

Returns the “default value” for a type.