Opensolr Search

Өзгөртүүлөр

1.0.21

• Fix: File / Document / Audio / Video media mappings now emit a real public
  URL instead of the bare attachment ID. Previous versions only resolved
  IMAGE attachments — mapping a PDF / DOCX / MP4 / MP3 attachment ID through
  Facet Mapping silently kept the integer in Solr. Now any image, video,
  audio, or known document MIME (PDF / DOC / DOCX / XLS / XLSX / PPT / PPTX
  / ODT / ODS / TXT) auto-resolves to its public URL.
• Fix: ACF “Image Array” / “File Array” return format no longer pollutes
  Solr with a mix of id + alt + title + url + width + height. The flatten
  pass now detects the ACF attachment shape and emits ONE value per row —
  the url if present, otherwise the ID (which then resolves through the
  normal attachment-URL pipeline). Single-mode AND multi-mode mappings both
  produce clean URL values.
• Fix: Values that are already URLs (http://… / https://…) pass through
  unchanged — covers ACF “Return Format = URL” + hand-rolled custom-field
  patterns. Previously the resolver only fired on numeric IDs, leaving
  everything else as-is, which was already correct for URLs but is now
  documented and explicit.

1.0.20

• New: Date-range quick presets on facet sidebar — every date_range facet
  now renders three pill buttons (Today / Last Week / Last Month) above the
  From/To inputs. One click fills both dates and submits the facet form — no
  more opening the native date picker twice for common queries. Uses LOCAL
  date components (not toISOString) so “Today” stays today in every timezone.
• New: Data Ingestion tab — live “Total: N documents” tally and dynamic
  per-type counts. Check or uncheck a content type and the number next to
  that type flips between its real count and 0, with the grand total at the
  bottom of the fieldset updating instantly. Mirrors the Drupal module’s
  ingestion UX. Include attached files toggle also contributes to the total.

1.0.19

First public release since 1.0.14. Rolls up every intermediate deploy.
* New: Facet Mapping form now matches the Drupal module — Solr Field
Name + Solr Type dropdown + Display Label. Pick a WP field and the
Solr Type auto-fills from the detected suffix; stem auto-fills from
the WP field name. Type hints shown next to every entry in the
WordPress Field dropdown (post_tag → multi, string; price → single,
float; etc.).
* New: WordPress Field dropdown shows every indexable source, not just
a handful of postmeta keys. Core post columns (post_title,
post_content, post_excerpt, post_date, post_status, post_type,
post_author), ALL postmeta (underscore-prefixed included —
_price/_sku/_stock/_thumbnail_id are now mappable), register_meta()
keys, every public taxonomy, all pa* WooCommerce attribute
taxonomies, plus synthetic featured_image_url and gallery_image_urls.
* New: Core post columns and WC synthetic aggregates resolve end-to-end
in both ingestion and tag emission.
* Fix: Facet Mapping saves no longer silently drop every row. The
sanitize callback hooked via register_setting() expected the legacy
flat scalar shape and called is_scalar($dst) on each value — every
row in the new rich {wp_field, meta_name, field_type, label,
solr_field} shape was failing the check and being stripped under
sanitize_option_opensolr_settings. Callback now accepts both shapes.
* Fix: Multi-valued mappings stored as serialized arrays in a single
postmeta row (ACF Checkbox / Select multi / Relationship / Gallery
fields, WooCommerce attributes) now emit one per value and
serialize as Solr arrays. Both emission and ingestion paths flatten
across both WP storage patterns (multi-row AND single-row-with-array).
* Fix: Admin scripts self-invalidate on every deploy via filemtime()
cache-busting. Previously OPENSOLR_VERSION-based cache-busting meant
every intermediate fix shipped to disk without browsers refetching.
* Fix: Facet Mapping table always renders at least one empty row so
the admin has a visible slot to click into.
* Fix: + Add Field Mapping button no longer collides with the
placeholder row’s index.
* Fix: Backend stem fallback — save handler derives the Solr Field
Name from the WP field name when left empty, preventing silent row
drops on incomplete input.

1.0.14

• Fix: image-field Facet Mappings now emit a real image URL. Previously,
  mapping a meta key that stores an attachment ID (ACF image field
  returning “ID”, _thumbnail_id, custom image_attachment_id, etc.) to
  a Solr field emitted the bare integer as the meta content — useless
  as an image source. Both ingestion and meta-tag emission now
  auto-resolve numeric values to wp_get_attachment_url() when the ID
  points at a valid image attachment. SKU, price, and other numeric
  fields are untouched (guarded by wp_attachment_is_image()).

1.0.13

• Fix: Persistent Filters “Exclude” mode was silently saved as “Include”.
  The register_setting() sanitize callback used a op: include/exclude
  shape that didn’t match the mode: +/- shape the admin form writes
  and the query builder reads, so any - (Exclude) selection was
  overwritten with the default on save. Existing stuck filters need to
  be re-saved after upgrading to pick up the correct mode.

1.0.12

• New: combobox autocomplete on “Solr field for result image” in Search
  Display. Focusing the input drops down the full alphabetical list of
  Solr fields indexed on your site (from the Luke handler); typing
  filters by case-insensitive substring match. Keyboard navigation
  (arrows + Enter) and mouse picking supported. New admin-only REST
  endpoint /opensolr/v1/luke-fields with a manage_options
  permission_callback feeds the list. JS and CSS enqueued via
  wp_enqueue_script / wp_enqueue_style (no inline tags); URL and
  nonce passed via wp_localize_script; all output escaped; DOM-based
  rendering avoids innerHTML. Plain typing still works for field names
  that aren’t indexed yet.

1.0.11

• New: “Solr field for result image” input in Search Display settings.
  Default og_image. Lets admins point result thumbnails at any
  single-valued Solr field they’ve mapped through Facet Mapping
  (WordPress post meta, WooCommerce gallery, custom URL field, etc.).
  Applies to the search results page, the AI Reader modal, and the
  AI Reader metadata endpoint. Multi-valued sources normalized to the
  first element. Empty or non-alphanumeric input falls back to og_image.

1.0.10

• Fixed: Facet Mapping entries targeting multi-valued Solr fields
  (_sm/_fm/_im/_dtm) now work correctly with both the Web Crawler
  and Data Ingestion paths. Previously the ingest path was broken entirely
  for field mappings (reading config keys the admin UI never wrote), and
  the meta-tag emission concatenated array values into a single
  comma-joined string. Now:
  • Ingest parses the suffix out of solr_field to determine multi vs
    single, and writes arrays directly to Solr.
  • Meta-tag emission writes one <meta property="opensolr:X_sm"> tag
    per value so the Opensolr crawler can aggregate them back into an
    array (DOM order preserved).
  • Hardcoded emissions for product_images_sm and variable product
    attribute *_sm fields updated to the same repeated-tag pattern.
    Pairs with a crawler update that aggregates repeated opensolr:* tags.

1.0.9

• Security: register_setting() sanitize_callback now applies per-field
  sanitization (sanitize_text_field, sanitize_email, absint, float clamping,
  strict enum allowlists, recursive array sanitization for facets / filters /
  content type lists). Previous callback only merged input, relying on
  caller-side sanitization.
• Security: REST /queue-stats endpoint now requires manage_options. The
  endpoint feeds the Data Crawler admin UI (live preview counts for arbitrary
  content-type overrides), so it’s an admin-configuration surface and should
  not be public.
• Compliance: JSON-LD structured-data blocks (WebSite SearchAction, Article,
  BreadcrumbList) now output via wp_print_inline_script_tag() instead of
  echoing raw tags. Same XSS protection (JSON_HEX_TAG | JSON_HEX_AMP
  prevents script breakout) — just routed through WordPress’s official helper.
• Dependency: Chart.js bundled library updated from 4.5.0 to 4.5.1.

1.0.7

• New: Lexical Only search mode in Search Tuning → Search Mode. Explicit
  opt-out from vector — no embedding API call is made, search runs straight
  through the lexical edismax path. Useful for deterministic BM25 ranking,
  avoiding embed API latency, or users on plans without vector who want
  explicit control instead of the implicit server-side fallback.
• New: When Lexical Only is selected, the Minimum Relevance Score slider
  auto-zeros itself. That threshold is calibrated for the combined hybrid
  score and would filter out every result in pure lexical mode. Backend
  forces the value to 0 on save and at query time; JS zeroes the slider
  live the moment the radio is picked. A yellow note explains the coupling.
• Change: Lexical-only fallback now uses defType=edismax + raw q (idiomatic
  Solr style) instead of q={!bool should=$lexicalQuery}. Same behavior,
  cleaner in the Query Inspector, matches the opensolr search_lexical()
  method for cross-platform consistency.
• Change: Plugin version bumped so browsers pick up the new admin-ui JS
  handler for the bool_mode radios.

1.0.5

• New: Minimum Relevance Score slider in Search Tuning. Drops results whose
  combined {!bool}(lexical + vector) score falls below the threshold, cleaning
  noise from facet counts and result totals. Applied as a Solr frange
  post-filter on the hybrid score. Needed because the {!func}-wrapped lexical
  side of the {!bool} matches every doc (score 0 for non-lexical hits) and
  KNN topK returns N candidates regardless of cosine similarity — without this
  filter numFound balloons to the entire base filter set and facet counts get
  polluted. Default 0.3; legitimate matches easily clear it. Slider moves in
  0.1 increments, 0 disables the filter.
• Change: Plugin version bumped so browsers pick up the new admin-ui JS.
  Without the bump, cached old JS had no listener for the new slider and its
  live label wouldn’t update while dragging.

1.0.4

• Improvement: The Semantic ↔ Lexical Balance slider is now a true balance dial.
  Previously it just scaled the qf/pf field-weight boosts linearly (same effect
  as lowering the weights themselves). The lexical score is now reshaped via
  pow(BM25 + 1, lexical_weight) – 1, so the slider changes the SHAPE of the
  lexical contribution curve, not just its magnitude:
  • 1.0 = linear (full BM25, named entities dominate)
  • 0.5 = sqrt-like compression (balanced, new default)
  • 0.2 = heavy compression (vector takes over)
  • 0.1 = lexical almost off
• Change: Default Title field weight raised from 0.03 to 0.1 and default
  Semantic ↔ Lexical Balance raised from 0.20 to 0.50. These defaults work
  better with the new pow reshape across a wide range of corpora.
• Removed: Misleading “field weights are automatically overridden to 0”
  banner in Search Tuning — no longer applies after the hybrid ranking fix
  in 1.0.3.

1.0.3

• Fix: Hybrid search ranking ignored query keywords. In vector hybrid mode the
  lexical qf was hardcoded to zero, so proper nouns, product names and other
  specific terms contributed nothing to scoring — results were ranked purely
  on semantic similarity, which pushed near-duplicate neighbours ahead of exact
  matches. Field weights from Search Tuning now apply alongside the KNN vector
  score as designed.

1.0.2

• Fix: Filters button on mobile search page now toggles the facet sidebar
  (class-name mismatch between template, CSS and JS — JS was listening on
  the old selectors left over from an earlier refactor).

1.0.1

• Plugin-review compliance pass (no functional changes):
• Moved all inline <script> and <style> blocks into enqueued .js and .css files (wp_enqueue_script / wp_enqueue_style / wp_localize_script)
• Refactored all $wpdb->prepare() calls to inline placeholders (fixes Plugin Check InterpolatedNotPrepared + UnfinishedPrepare warnings)
• Tightened JSON-LD output escaping: JSON_HEX_TAG | JSON_HEX_AMP to prevent any possible breakout
• esc_url_raw → esc_url wherever a URL is written into an output context
• Added opt-out for local analytics logging (Search Display → Analytics toggle)
• Documented every external service, endpoint and data flow in the “External services” section above
• Updated bundled libraries: Chart.js 4.5.0, marked 15.0.12, SortableJS 1.15.6

1.0.0

• Initial release
• Hybrid vector + keyword search
• Faceted navigation with list, slider, date range, and hierarchical widgets
• AI Hints and AI Reader streaming
• Autocomplete with query suggestions
• WooCommerce product support
• Web Crawler and Data Ingestion indexing
• Search analytics with CTR tracking
• Query Elevation (pin/exclude results)
• Persistent admin filters
• Multilingual support (WPML, Polylang)
• Dark theme
• Embeddable search widget mode
• Meta tag and JSON-LD injection for SEO
• XML sitemap generation