Fills opening_hours, phone, and website gaps when OSM + Overture data
is incomplete. Only fires for business-class POIs (amenity, shop, tourism,
leisure, office, craft). Daily API call cap with SQLite tracking.
cache_put now preserves google columns across cache refreshes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When a query contains no road-type keywords (st, blvd, ave, etc.),
boost amenity/shop/tourism/leisure/office/craft results (+3.0) and
penalize highway/route results (-4.0). This fixes searches like
"starbucks twin falls" where a named service road outranked the
actual business POI due to Photon position tiebreaking.
Also fixes:
- Intent classifier now recognizes full state names ("idaho" not
just "ID") for LOCALITY classification
- Locality-type Photon results now populate _city from name field
so they participate in locality_fuzz scoring
- Trace logging expanded to all candidates with osm_key/value
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Businesses with housenumbers (e.g. M&W Markets at 130 US-30) were
classified as street_address because the housenumber check fired before
the osm_key check. Reorder so osm_key in amenity/shop/tourism/leisure/office
is evaluated first, ensuring businesses get type=poi regardless of
whether they have a street address. Also adds office to the POI key set.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Ingests 20.9M North America places from Overture Maps Foundation
(release 2026-04-15.0) into PostgreSQL. Enriches /api/place responses
with phone, website, and brand data via spatial + fuzzy name matching
when OSM extratags are sparse.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New /api/place/<osm_type>/<osm_id> endpoint returns cleaned OSM tag data
for PlaceDetail panel enrichment. Routes to local Nominatim (Idaho coverage)
first, falls back to Overpass public API for out-of-region queries. Responses
cached in SQLite (data/place_cache.db) with no expiry.
New modules: lib/place_detail.py (proxy + cache), lib/osm_categories.py
(~50 category humanization mappings). Profile YAMLs updated with
place_details config block and has_nominatim_details flag.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add /api/traffic/flow proxy route to hide TomTom API key from frontend
- Add tileset_hillshade and traffic config blocks to all three profiles
- Flip has_hillshade and has_traffic_overlay flags in home and regional profiles
- Minimal profile has config blocks but flags remain false (dormant)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add profile-driven config infrastructure:
- config/profiles/{home,regional_pi,minimal_pi}.yaml templates
- lib/deployment_config.py loader (reads RECON_PROFILE env var)
- GET /api/config returns active profile as JSON (5min cache)
Frontend reads this on startup to determine tile source, defaults,
and feature flags. No existing behavior changed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Accepts lat/lon query params, calls Photon /reverse, returns same
response shape as /api/geocode. Returns 200 with empty results on
no match (graceful degradation for ocean/unmapped areas).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Inverts the /api/geocode chain. Photon is now the primary search
engine; the hand-rolled Netsyms free-text parser is removed.
Address book short-circuits nicknames only ("home", "work") —
full-address queries flow through Photon and address book
entries within 75m annotate matching results with labeled_as.
Coordinate strings detected before search.
Response shape: /api/geocode now returns a ranked candidates
list (always 200 OK, empty list if no match). No more 404 for
unmatched queries. Users can type messy input — wrong case,
missing punctuation, abbreviations, typos — and get results
or close matches.
Netsyms preserved at /api/netsyms/lookup for direct access.
USPS plus4 enrichment of Photon street-address hits is a
planned follow-up.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
lookup() previously did exact-alias-only matching, so "214 north st
filer" missed the home entry with alias "214 north st". Extend to
match when the query begins with an alias followed by a word
boundary, and when an alias appears as a contiguous token sequence
inside the query. Short aliases ("home") keep matching exactly and
also match with trailing text.
Fixes the UX case where typing a known full address falls through
to Netsyms instead of short-circuiting to address_book.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- YAML-backed saved locations (config/address_book.yaml)
- Exact/partial alias matching with case-insensitive lookup
- Flask blueprint: /api/address_book/lookup, /api/address_book/list
- Geocoder short-circuits Photon when address book has exact match
- Test suite for lookup behavior
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add centroid-based query classifier that routes Aurora queries to the
appropriate handler (nav_route, nav_reverse_geocode, direct_answer,
rag_search) before the RAG pipeline runs. Uses TEI embeddings against
pre-computed route centroids from 38 example queries.
- query_router.py: standalone module with lazy centroid init
- query_router_test.py: 7-query test suite (all passing)
- Corresponding recon_rag_tool.py v4.2.0 deployed to Open WebUI DB
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- nav_tools.py: route() geocodes via Photon, routes via Valhalla, returns
summary/maneuvers/polyline. reverse_geocode() for coordinate lookups.
Supports auto/pedestrian/bicycle/truck modes.
- nav_tools_test.py: 5 live tests against local Photon (2322) and Valhalla (8002)
- aurora_nav_tool.py: Open WebUI Tool exposing get_directions to Aurora LLM
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Parse Browsertrix "crawled":N JSON format instead of "N pages"
- Add 3s delay between SIGHUP to kiwix-serve and scan_zims() call
so the OPDS catalog is reloaded before we query it for linking
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Parse "crawled":N from Browsertrix crawlStatus JSON logs instead of
looking for "N pages" pattern. Also check stdout (not just stderr).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
warc2zim (called internally by zimit) requires --name for ZIM metadata.
Without it, argument validation fails with exit code 2.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Must pass `zimit` as command after image name (entrypoint execs args)
- --url → --seeds, --name removed, --lang → --zim-lang, --workers → -w
- Remove --rm so docker logs work after exit, manually rm container
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Extract shared _full_zim_cleanup(source_id) from api_kiwix_remove
- Add SIGHUP to kiwix-serve after kiwix-manage remove
- Delete linked scrape_jobs rows during ZIM removal
- Update api_scraper_delete to do full ZIM cleanup when applicable
- Set chromium_path for single-file browser crawl support
- Add status.db to .gitignore
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New API endpoints: DELETE single job, clear all failed/cancelled.
Dashboard now shows Delete buttons on completed/failed jobs,
Retry+Delete on failed jobs, and a Clear Failed bulk action.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New /kiwix/scraper page with submit form (URL, title, language,
crawl mode), stats cards, and auto-refreshing jobs table with
cancel/retry actions. Kiwix section now has Library/Scraper subnav.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Format: {domain}_{lang}_{YYYY-MM}_{job_id}.zim
Prevents zimwriterfs failures when the same domain is scraped
multiple times in the same month.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
SingleFile CLI has no --crawl-delay option. The invalid flag caused the
process to print help and exit with no output. Added --crawl-no-parent
and --crawl-replace-URLs instead. Removed unused crawl_delay config key.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Complete sources now show "19,344 in Qdrant" instead of misleading
extraction counts. Each status gets contextual progress display:
complete → X in Qdrant, processing → X/Y in Qdrant (%),
extracting → X/Y extracted, detected → dash.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Status was showing COMPLETE after ZIM extraction finished, even when
documents were still queued for enrichment/embedding. Now computes
effective_status by checking actual pipeline state per-source:
- DETECTED: ingest not enabled (gray)
- EXTRACTING: ZIM processor running (blue)
- PROCESSING: extracted but docs still in enricher/embedder queue (amber)
- COMPLETE: all docs fully enriched and embedded in Qdrant (green)
Also fixed _build_kiwix_sources pipeline query to filter by category
per-source instead of returning global kiwix stats for every source.
Progress column now shows "X / Y in Qdrant" when processing, or
"X / Y extracted" otherwise.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Install langdetect package for content-level language detection
- Add _check_language() to enricher.py: reads first 1500 chars of first
page, detects language via langdetect, skips if not in allowed list
- Configurable via config.yaml pipeline.language_filter and
pipeline.allowed_languages (default: en only)
- Catches non-English content from ANY source (PDF, web, ZIM, PeerTube)
before burning Gemini API quota on enrichment
- Add scan_zims retry logic (3 attempts, 2s delay) for upload handler
- Purged 6,483 stale non-English zim_articles rows from DB
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Skip articles with MediaWiki translation suffixes (/es, /fr, /pl, etc.)
before text extraction to avoid wasting Gemini enrichment on translations.
Uses path-based regex matching against ISO 639 language codes.
~5,276 non-English articles already ingested from Appropedia (top: es=837,
zh=765, ru=475, fr=433, ko=407). Purge decision deferred.
- ZIM processor: extract articles from ZIM files, feed into existing enrichment pipeline
- Dashboard: Kiwix tab with library table, ingest toggle, upload, remove
- kiwix-serve on port 8430, wiki.echo6.co behind Authentik
- Citation URLs point to wiki.echo6.co/{zimname}/{article_path}
- Dashboard shows WIKI type badge for ZIM-sourced content
- Appropedia EN (19,445 articles) fully ingested as proof of concept
Adds lib/processors/zim_processor.py which opens a ZIM file via
python-libzim, iterates HTML articles, strips to clean text (lxml),
and feeds each article into the existing RECON enrichment pipeline.
Key features:
- HTML to text via lxml (strips nav/footer/script/style)
- Filters redirects, non-HTML entries, stubs (<200 chars)
- Content hash dedup against existing catalogue
- Creates processing dirs with page files and meta.json
- Registers articles as "extracted" for automatic enrichment
- Checkpointing via zim_sources.last_checkpoint for resume
- Configurable batch size and delay for rate control
- Standalone CLI: python3 -m lib.processors.zim_processor
Tested: 100 Appropedia articles processed in 3s, enricher picks
them up automatically via the existing pipeline.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add lib/zim_monitor.py: polls kiwix-serve OPDS v2 catalog, detects
new ZIMs, reads accurate article count from python-libzim Counter
metadata (not inflated OPDS count), inserts into zim_sources table.
Idempotent on re-run, marks removed ZIMs.
- DB schema: zim_sources, zim_samples, zim_articles tables (created
via sqlite3, not in migrations — matches existing RECON pattern)
- kiwix-tools 3.7.0 installed from binary tarball at /opt/recon/bin/
(Ubuntu 24.04 apt ships 3.5.0 which lacks OPDS v2)
- kiwix.service systemd unit on port 8430
- python-libzim 3.9.0 installed
- Test ZIM: Appropedia EN maxi (496 MB, 19,445 articles)
- Add bin/ to .gitignore (binary tarball, not source)
Upload handler now writes files to the appropriate hopper subfolder
instead of copying directly to /mnt/library/:
- .pdf -> acquired/pdf/
- .txt -> acquired/text/
- .epub, .doc, .docx, .mobi -> acquired/pdf/ (dispatcher format
normalizer converts to PDF before processing)
The dispatcher picks up files and routes through the appropriate
processor (pdf_processor or text_processor) for full metadata
voting, domain classification, and canonical filing.
Changes to api_upload() / _process_upload():
- Relaxed extension check: PDF, TXT, EPUB, DOC, DOCX, MOBI
- Routes to correct hopper subfolder by extension
- Writes meta.json sidecar with original filename and category hint
- Removed: direct library copy, add_to_catalogue, queue_document
- Added: hopper-level dedup check (catches rapid re-uploads)
- Kept: catalogue dedup check for immediate user feedback
Changes to api_upload_status():
- Added fallback: checks acquired/ and processing/ dirs if hash
not yet in documents table (covers gap between upload and
dispatcher pickup)
Template updated: accept attribute and help text now reflect
multi-format support.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Same fix as text_processor — Gemini sometimes returns the literal
string "null" instead of JSON null for empty metadata fields. The
voting logic and Gemini extraction now both treat "null" strings
as None.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds _normalize_formats() to the dispatcher that converts non-standard
document formats to PDF before dispatch. Supports:
- .epub, .mobi -> PDF via ebook-convert (Calibre)
- .doc, .docx -> PDF via LibreOffice headless
Called per-subfolder before _find_pairs() so _find_pairs() only ever
sees standard content files. Conversion failures are logged and
skipped -- the original file stays in acquired/ for manual review.
Also converts 3 staged epub files and cleans up _staging/.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace legacy ingest_channel/ingest_all imports with acquire_batch
from lib.acquisition.peertube. The endpoint now writes flat file pairs
to the hopper and lets the dispatcher handle processing, matching the
Phase 6d architecture. Removes channel/since/process parameters that
were tied to the old direct-ingest path.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New lib/acquisition/peertube.py replaces the removed peertube_scanner_loop.
Polls PeerTube API every 30min, dedupes against catalogue (UUID + title),
writes flat file pairs to data/acquired/stream/ for the dispatcher.
- acquire_batch(): one-shot find-and-acquire with rate limiting
- acquisition_loop(): service thread wrapper (interval from config)
- list_new_videos(): dedup via _build_known_sets() against catalogue
- acquire_one(): fetch VTT, convert, write .tmp then rename atomically
cmd_service(): added peertube-acq daemon thread
cmd_ingest_peertube(): rewired to use acquire_batch(), drops --channel/
--since/--enrich/--process (dispatcher handles full pipeline)
config.yaml: added peertube.poll_interval: 1800
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
recon.py:
- Remove extract stage_loop thread from cmd_service(). Confirmed
vestigial: 0 queued items, silent logs over 24+ hour run. The new
processors do extraction inline in pre_flight().
- Remove cmd_crawl CLI subcommand and its argparse registration.
- Clean up associated imports and variables.
Deleted:
- lib/crawler.py (432 lines) -- old web crawler subsystem, only
referenced by the removed CLI subcommand.
- 24 .bak files (untracked pre-edit safety backups, originals
preserved in git history).
Investigation found the four old loop function definitions
(scanner_loop, peertube_scanner_loop, crawler_scheduler_loop,
organizer_loop) were already deleted in Phase 5c-1.
Modules investigated and KEPT:
- lib/web_scraper.py -- exports chunk_text() used by transcript_processor
- lib/new_pipeline.py -- active Stream B library management CLI tool
- lib/peertube_scraper.py -- only mechanism for transcript ingestion
- lib/extractor.py -- would activate for new PDFs via cmd_run CLI
Service restart verified: 6 threads (dispatcher, enrich, embed,
filing, progress, dashboard), no extract worker, zero errors.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Two bugs in the Recently Completed table:
1. Title showed "Untitled" for all transcripts because the dashboard
read documents.book_title (populated by PDF metadata voting) which
is NULL for transcripts. Fixed by COALESCE(book_title, filename)
in the SQL query -- falls back to catalogue.filename which holds
the real video title.
2. Type showed "WEB" for all transcripts because the type CASE
expression only had web and pdf branches, with web matching any
http% path -- and transcript paths are PeerTube watch URLs.
Fixed by adding a transcript branch keyed on catalogue.source =
stream.echo6.co, evaluated before the web branch.
Also adds badge-transcript CSS (purple) and JS rendering case.
Applied consistently to both the Recently Completed and Sources
table queries.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Transcripts are derived text from PeerTube videos, not primary source
files. They do not belong in library/Domain/Subdomain/ like PDFs.
Change: transcript_processor.pre_flight() now sets organized_at =
CURRENT_TIMESTAMP at the end of successful processing, marking the
transcript as organized in place. The watch URL remains in
catalogue.path and Qdrant download_url so users clicking search
results go to the PeerTube video.
The filing workers path LIKE filter naturally excludes transcripts
since their documents.path is the watch URL, not a filesystem path.
No filing worker changes needed.
Back-fills 2,260 drain items from Phase 5c-2 via one-time SQL.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
