Struct PurposeInputReader 

struct PurposeInputReader<'a> {
    control: &'a IndexWorkControl,
    limits: PurposeImportLimits,
    complete_paths: BTreeSet<PathBuf>,
    purpose_filename: String,
    complete_digests: BTreeMap<PathBuf, String>,
}

Operation-owned reader for authored purpose and publication inputs.

Fields

control: &'a IndexWorkControl

Shared cancellation and deadline boundary for the publication.

limits: PurposeImportLimits

Byte and record limits for authored purpose inputs.

complete_paths: BTreeSet<PathBuf>

Inputs that must be consumed completely rather than as header prefixes.

purpose_filename: String

Configured legacy folder-purpose filename.

complete_digests: BTreeMap<PathBuf, String>

Exact digests retained only for complete publication-contract inputs.

Implementations

impl<'a> PurposeInputReader<'a>

fn new( plan: &ScanRuntimePlan, control: &'a IndexWorkControl, limits: PurposeImportLimits, ) -> Self

Create one reader whose cancellation and limits belong to the scan operation.

fn for_complete_paths( control: &'a IndexWorkControl, limits: PurposeImportLimits, complete_paths: BTreeSet<PathBuf>, purpose_filename: String, ) -> Self

Create a bounded reader before a complete runtime plan is available.

fn read_text(&mut self, path: &Path) -> Result<String, CliError>

Read one UTF-8 input, treating non-UTF-8 source headers as purpose-free.

fn read_bytes<R: Read>( &mut self, path: &Path, reader: &mut R, file_limit: u64, require_complete: bool, ) -> Result<Vec<u8>, CliError>

Read bytes with inter-chunk cancellation and aggregate accounting.

fn complete_digest(&self, path: &Path) -> Option<&str>

Return the digest of one complete input already read through this boundary.