spacedriveapp/spacedrive · error

Entry has no content_id

Error message

Entry has no content_id

What it means

The OCR processor requires entry.content_id because extracted text is written into the content_identity row. process() unwraps the Option and errors when None: the entry passed the mime-type check but was never identified (no content identity created), so there is nowhere to store OCR output.

Solutions

1. Ensure the identifier processor runs before OCR in the pipeline ordering
2. Add a content_id requirement to the processor's wants_to_process check so unqualified entries are filtered, not errored
3. Re-run identification on affected entries, then re-queue OCR

Example fix

// before (wants_to_process)
entry.mime_type.as_ref().map_or(false, |m| {
    super::is_ocr_supported(m, self.library.core_context().file_type_registry())
})

// after: also require a content identity, mirroring process()'s hard requirement
entry.mime_type.as_ref().map_or(false, |m| {
    super::is_ocr_supported(m, self.library.core_context().file_type_registry())
}) && entry.content_id.is_some()

// Only enqueue OCR when a content identity exists
if entry.content_id.is_some() {
    ocr_processor.enqueue(entry).await?;
}

fn is_ocr_eligible(entry: &ProcessorEntry, registry: &FileTypeRegistry) -> bool {
    entry.content_id.is_some()
        && entry
            .mime_type
            .as_ref()
            .map_or(false, |m| super::is_ocr_supported(m, registry))
}

match ocr.process(db, &entry).await {
    Ok(res) => { /* aggregate results */ }
    Err(e) if e.to_string().contains("Entry has no content_id") => {
        tracing::debug!(path = %entry.path.display(), "skipping OCR: entry not yet identified");
    }
    Err(e) => return Err(e),
}

When it happens

Trigger: Running the OCR processor on entries the identifier has not processed yet (pipeline ordering wrong); entries whose identification failed or was skipped; manually enqueuing entries into the processor without the identify step.

Common situations: Processor registry ordering changed so OCR runs before identification; a new file picked up mid-write before identify completed; identification errors silently skipped an entry.

Related errors

• ContentIdentity not found
• ContentIdentity not found
• ContentIdentity missing UUID
• Entry not found
• Proxy generation failed

AI-assisted analysis of spacedriveapp/spacedrive@6dfeccf211 (2026-08-16). Data as JSON: /api/errors/6be7474e19f76ab7. Report an issue: GitHub.

Appendix: source

Thrown at core/src/ops/media/ocr/processor.rs:75

		}

		if entry.content_id.is_none() {
			return false;
		}

		entry.mime_type.as_ref().map_or(false, |m| {
			super::is_ocr_supported(m, self.library.core_context().file_type_registry())
		})
	}

	pub async fn process(
		&self,
		db: &sea_orm::DatabaseConnection,
		entry: &ProcessorEntry,
	) -> Result<ProcessorResult> {
		let content_id = entry
			.content_id
			.ok_or_else(|| anyhow::anyhow!("Entry has no content_id"))?;

		debug!("→ Extracting text via OCR for: {}", entry.path.display());

		// Extract text
		let extracted_text = super::extract_text_from_file(&entry.path, &self.languages).await?;

		if extracted_text.is_empty() {
			debug!("No text extracted from: {}", entry.path.display());
			return Ok(ProcessorResult::success(0, 0));
		}

		debug!("✓ Extracted {} characters of text", extracted_text.len());

		// Update content_identity with extracted text
		use crate::infra::db::entities::content_identity;

		let ci = content_identity::Entity::find_by_id(content_id)
			.one(db)

View on GitHub (pinned to 6dfeccf211)