fix: correct duplicate message reporting in Go implementation

- Add DocumentSkippedError custom error type to distinguish between skipped and stored documents
- Fix counter bug where skipped messages were incorrectly reported as "stored"
- Enhance status reporting to show "X skipped as duplicates" for better visibility
- Fix Rust implementation binary attachment handling to support all file types (images, PDFs, etc.)
- Update test scripts to use correct binary names (mail2couch-go, mail2couch-rs)
- Add comprehensive test configurations for implementation comparison

Before: "Summary: Processed 30 messages, stored 30 new messages" (misleading when all were duplicates)
After: "Summary: Processed 30 messages, stored 0 new messages" with detailed "Stored 0/30 messages from INBOX (30 skipped as duplicates)"

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Ole-Morten Duesund 2025-08-04 00:36:01 +02:00
commit 436276f0ef
9 changed files with 127 additions and 18 deletions

View file

@ -705,21 +705,14 @@ impl ImapClient {
// Generate a filename for the attachment
let filename = self.get_attachment_filename(part, index);
// Get the content data (try different methods based on content type)
let body_data = if let Some(text_content) = part.get_text_contents() {
// Text-based attachments
log::debug!("Found text attachment content: {} bytes", text_content.len());
text_content.as_bytes().to_vec()
} else {
// For now, skip attachments without text content
// TODO: Implement binary attachment support with proper mail-parser API
log::debug!("Skipping non-text attachment for part {} (content-type: {})", index, content_type.c_type);
vec![]
};
// Get the binary content data using the proper mail-parser API
// This works for both text and binary attachments (images, PDFs, etc.)
let body_data = part.get_contents().to_vec();
log::debug!("Found attachment content: {} bytes (content-type: {})", body_data.len(), content_type.c_type);
let content_type_str = content_type.c_type.to_string();
// Only create attachment stub if we have actual data
// Create attachment stub - get_contents() always returns the full data
if !body_data.is_empty() {
let attachment_stub = AttachmentStub {
content_type: content_type_str.clone(),