feat(archives): Dublin Core + EAD3 export + OAI-PMH 2.0#127
Merged
fabiodalez-dev merged 11 commits intomainfrom May 4, 2026
Merged
feat(archives): Dublin Core + EAD3 export + OAI-PMH 2.0#127fabiodalez-dev merged 11 commits intomainfrom
fabiodalez-dev merged 11 commits intomainfrom
Conversation
- Dublin Core XML: GET /admin/archives/{id}/dc.xml with full ISAD(G)→DC
crosswalk; discovery <link rel="alternate"> in detail view (#124)
- EAD3 export: GET /admin/archives/export.ead3 downloads full collection
as EAD3 XML with ISAD(G)→EAD3 crosswalk (#125)
- OAI-PMH 2.0: public GET|POST /archives/oai endpoint — Identify,
ListMetadataFormats (oai_dc + ead3), ListIdentifiers, ListRecords,
GetRecord, ListSets; cursor-based pagination via resumptionToken (#126)
- Export buttons added to archives list view
- Translations added to it_IT, en_US, de_DE
Closes #124, #125, #126
|
Warning Rate limit exceeded
To keep reviews running without waiting, you can enable usage-based add-on for your organization. This allows additional reviews beyond the hourly cap. Account admins can enable it under billing. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (7)
📝 WalkthroughWalkthroughLa PR aggiunge tre endpoint di interoperabilità (Dublin Core per unità, esportazione bulk EAD3, endpoint OAI‑PMH pubblico), estende i layout per emettere link in ChangesInteroperabilità Archivistica e OAI‑PMH
Sequence DiagramsequenceDiagram
actor Client
participant OAI as OAI-PMH Handler
participant DB as Database
participant Serializer as DC/EAD3 Serializer
Client->>OAI: GET /archives/oai?verb=Identify
OAI->>OAI: Validate verb & params
OAI->>Serializer: Build Identify response
Serializer-->>OAI: XML Identify element
OAI-->>Client: 200 + OAI Identify XML
Client->>OAI: GET /archives/oai?verb=ListRecords&metadataPrefix=oai_dc
OAI->>DB: Query non-deleted units (from/until/set filters, cursor)
DB-->>OAI: Unit rows + cursor
OAI->>Serializer: Build DC record per unit
Serializer-->>OAI: oai_dc:dc elements
OAI->>OAI: Encode resumptionToken(cursor, prefix, filters)
OAI-->>Client: 200 + ListRecords (records + resumptionToken)
Client->>OAI: GET /archives/oai?verb=ListRecords&resumptionToken=...
OAI->>OAI: Decode resumptionToken → cursor, prefix, filters
OAI->>DB: Query units from cursor with filters
DB-->>OAI: Next batch + new cursor
OAI->>Serializer: Build next batch of records
Serializer-->>OAI: Record elements
OAI-->>Client: 200 + ListRecords (next batch)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minuti Possibly related PRs
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Review rate limit: 0/1 reviews remaining, refill in 16 minutes and 21 seconds.Comment |
- oaiIdentify: read repositoryName and adminEmail from ConfigStore instead of hardcoded placeholders (breaks every non-default install) - oaiListRecords: filter from/until on updated_at, not created_at; OAI-PMH datestamps track last-modified, so harvesters using incremental selects would miss updated records - ListMetadataFormats: add ead3 as a supported prefix (the EAD3 writer existed but was not advertised or reachable via OAI); handle metadataPrefix=ead3 in both oaiListRecords and oaiGetRecord - marcxml schema URL: remove duplicate namespace prefix that was incorrectly embedded in the <schema> element value - exportEad3CollectionAction: document <eadlist> as a Pinakes-specific container (not part of EAD3 schema) so consuming tools know to strip the outer wrapper
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@storage/plugins/archives/ArchivesPlugin.php`:
- Around line 4575-4593: The resumption token handling is done after validating
$metadataPrefix, so a continuation request that supplies only resumptionToken
triggers cannotDisseminateFormat; move token parsing ahead of the
$metadataPrefix validation so a token can override/restore
metadataPrefix/from/until/set; specifically, in the block handling $token and
$cursor (variables $token, $cursor, $parts) parse and decode the token first to
populate $metadataPrefix, $from, $until, $set and only then validate
$metadataPrefix via the existing in_array check and oaiError call; also replace
the fragile colon-joined token generation/consumption used around the code that
builds tokens (currently at the token creation around lines 4709-4714) with a
robust encoding (e.g., urlencoded components, JSON or base64-encoded payload) so
date/time and values containing ':' are safe to split and restore.
- Around line 4670-4702: The code incorrectly wraps a <header> inside a <record>
even when $identifiersOnly is true (i.e. $verbElement = 'ListIdentifiers'),
which breaks OAI-PMH schema; change the loop so that when $identifiersOnly is
true you do NOT call $xw->startElement('record') / $xw->endElement() and instead
write the <header> (identifier, datestamp, optional setSpec) as a direct child
of the top-level verb element; keep the existing metadata branch
(fetchAuthoritiesForArchivalUnit, writeDublinCoreRecord, writeEad3Document,
writeArchivalUnitMarcRecord) only inside the !$identifiersOnly path and ensure
corresponding $xw->startElement('metadata') / $xw->endElement() calls remain
balanced.
In `@storage/plugins/archives/views/show.php`:
- Around line 502-504: The <link rel="alternate" ... href="<?=
$e(url('/admin/archives/' . $id . '/dc.xml')) ?>"> tag is currently rendered in
the view body and should be moved into the page <head>; remove the inline
emission from show.php and instead emit the same link element into the template
engine's head slot/layout (use the head/layout slot API your renderer provides)
so the exact href expression $e(url('/admin/archives/' . $id . '/dc.xml')) is
output in the document head rather than the body.
- Line 504: Il link in storage/plugins/archives/views/show.php currently punta
alla route protetta "/admin/archives/{id}/dc.xml" usando href="<?=
$e(url('/admin/archives/' . $id . '/dc.xml')) ?>", impedendo a harvester non
autenticati di recuperare il DC; update il view per usare un endpoint pubblico
(es. "/archives/{id}/dc.xml") o creare un equivalente pubblico controller/route
che esponga il record Dublin Core e poi sostituire la chiamata a
url('/admin/archives/' . $id . '/dc.xml') con url('/archives/' . $id .
'/dc.xml') (oppure implementare e collegare una strategia machine-to-machine
esplicita se si vuole mantenere protezione), mantenendo l'uso della variabile
$id e della funzione url()/escaper $e.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 47f3c9fd-b7b8-4bb3-a70f-e10d57533130
📒 Files selected for processing (6)
locale/de_DE.jsonlocale/en_US.jsonlocale/it_IT.jsonstorage/plugins/archives/ArchivesPlugin.phpstorage/plugins/archives/views/index.phpstorage/plugins/archives/views/show.php
Comment on lines
+502
to
+504
| <?php /* Dublin Core discovery link — discoverable by aggregators and web crawlers. */ ?> | ||
| <link rel="alternate" type="application/rdf+xml" title="Dublin Core" | ||
| href="<?= $e(url('/admin/archives/' . $id . '/dc.xml')) ?>"> |
There was a problem hiding this comment.
Sposta il discovery link nel <head> della pagina
Alle Line 502-504 il tag <link rel="alternate"> viene emesso nel body della view: così può non essere considerato dai consumer di metadati. Per la discovery va pubblicato nell’<head> (via layout/slot head del template engine).
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@storage/plugins/archives/views/show.php` around lines 502 - 504, The <link
rel="alternate" ... href="<?= $e(url('/admin/archives/' . $id . '/dc.xml')) ?>">
tag is currently rendered in the view body and should be moved into the page
<head>; remove the inline emission from show.php and instead emit the same link
element into the template engine's head slot/layout (use the head/layout slot
API your renderer provides) so the exact href expression
$e(url('/admin/archives/' . $id . '/dc.xml')) is output in the document head
rather than the body.
|
|
||
| <?php /* Dublin Core discovery link — discoverable by aggregators and web crawlers. */ ?> | ||
| <link rel="alternate" type="application/rdf+xml" title="Dublin Core" | ||
| href="<?= $e(url('/admin/archives/' . $id . '/dc.xml')) ?>"> |
There was a problem hiding this comment.
URL di discovery su /admin non compatibile con harvester pubblici
Alla Line 504 l’href usa /admin/archives/{id}/dc.xml: se la route è protetta, aggregatori e crawler non autenticati non possono leggere il record Dublin Core. Serve un endpoint pubblico equivalente (o una strategia di accesso machine-to-machine esplicita).
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@storage/plugins/archives/views/show.php` at line 504, Il link in
storage/plugins/archives/views/show.php currently punta alla route protetta
"/admin/archives/{id}/dc.xml" usando href="<?= $e(url('/admin/archives/' . $id .
'/dc.xml')) ?>", impedendo a harvester non autenticati di recuperare il DC;
update il view per usare un endpoint pubblico (es. "/archives/{id}/dc.xml") o
creare un equivalente pubblico controller/route che esponga il record Dublin
Core e poi sostituire la chiamata a url('/admin/archives/' . $id . '/dc.xml')
con url('/archives/' . $id . '/dc.xml') (oppure implementare e collegare una
strategia machine-to-machine esplicita se si vuole mantenere protezione),
mantenendo l'uso della variabile $id e della funzione url()/escaper $e.
Owner
Author
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
Owner
Author
|
All three CodeRabbit findings (reviewed against
|
Owner
Author
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
1 similar comment
✅ Actions performedFull review triggered. |
There was a problem hiding this comment.
Actionable comments posted: 6
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@app/Views/frontend/layout.php`:
- Around line 210-212: Non fare echo diretto di $headExtra: smantella il sink
XSS e invece accetta/converti dati strutturati (es. un array di link/meta) e
renderizza esplicitamente ogni elemento usando escaping sugli attributi;
sostituisci l'echo di $headExtra con un renderer che itera su qualcosa come
$headExtra['links'] / $headExtra['metas'] e costruisce tag <link> e <meta> con
tutti gli attributi passati attraverso htmlspecialchars(..., ENT_QUOTES,
'UTF-8') (es. per href/action/src usare htmlspecialchars(url(...), ENT_QUOTES,
'UTF-8')), oppure rimuovi l'injection libero se non servono dati dinamici;
assicurati di riferirti a $headExtra nel codice esistente per la migrazione.
In `@app/Views/layout.php`:
- Around line 64-66: Non stampare più la stringa raw $headExtra: sostituisci
l'echo diretto in layout.php con un rendering sicuro che accetti dati
strutturati (es. un array $headExtraItems o una funzione renderHeadExtra) e non
HTML arbitrario; per ogni item, valida che il tipo sia consentito (es. 'link') e
che contenga solo gli attributi ammessi ('rel','href','type','title'), sanitizza
ogni attributo usando htmlspecialchars($value, ENT_QUOTES, 'UTF-8') e valida gli
URL con filter_var(..., FILTER_VALIDATE_URL) prima di inserirli, e se devi
mantenere compatibilità con stringhe legacy, non stamparle raw ma eseguire
htmlspecialchars sul contenuto intero o deprecarle.
In `@storage/plugins/archives/ArchivesPlugin.php`:
- Around line 4061-4063: The response is using the wrong MIME for OAI-DC: change
the Content-Type header set via response->withHeader('Content-Type',
'application/rdf+xml; charset=utf-8') to use application/xml (or another
OAI-PMH-consistent type) while keeping charset; also update the discovery links
(the link elements referenced around the discovery code paths that currently
declare type="application/rdf+xml") to use the same corrected type attribute so
the payload (<oai_dc:dc>) is served and advertised consistently; ensure the
Content-Disposition filename logic (the slug) remains unchanged.
- Around line 4110-4113: The in_array check inside the loop over $authorities
that maps roles to dc:creator must be tightened: remove 'photographer' and
'associated' so only 'creator' maps to dc:creator (update the condition in the
block that writes dc:creator where $auth['role'] is checked), and drop any
reference to the dead 'photographer' role (it’s not in AUTHORITY_ROLES); ensure
the writer call $xw->writeElementNs('dc','creator',...) only runs for role
'creator' to conform with Dublin Core crosswalk rules.
- Around line 4210-4215: The exported file is labeled as `.ead3.xml` but the
code emits a Pinakes-specific `<eadlist>` wrapper
(`$xw->startElement('eadlist')` / `$xw->writeAttribute('xmlns:xsi', ...)`) which
makes the XML non-conformant with EAD3; update the export behaviour to avoid
misleading consumers by either (A) removing the custom wrapper and emitting a
single valid `<ead>` per export (or producing multiple separate EAD3 files/ZIP)
or (B) keep the wrapper but change the filename/Content-Type to a generic `.xml`
(or append a `.nonstandard`/`-eadlist.xml` marker) and document the non-standard
wrapper in headers; locate the EAD bulk-export logic in ArchivesPlugin.php where
`$xw->startElement('eadlist')` is called and implement one of these two fixes
consistently for filename generation and HTTP Content-Type/header output.
- Around line 4429-4435: La logica che mappa le authority ISAAR in EAD3 va
aggiornata: invece di trattare tutto ciò che non è "corporate" come "persname",
rileva il valore "family" in $authType e usa il tag "famname" per quei casi;
aggiorna anche l'attributo di encodinganalog di conseguenza (es. usare '600' per
family), mantenendo invariati gli altri attributi ($role -> relator, testo da
authorised_form). Modifica la selezione del tag/encoding vicino alle variabili
$authType, $tag e alla chiamata a $xw->writeAttribute('encodinganalog') in modo
che 'corporate' => 'corpname'/'710', 'family' => 'famname'/'600', altrimenti
'persname'/'700'.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: c2f0a636-540a-4c84-af32-1b6ba45cea4d
📒 Files selected for processing (8)
app/Views/frontend/layout.phpapp/Views/layout.phplocale/de_DE.jsonlocale/en_US.jsonlocale/it_IT.jsonstorage/plugins/archives/ArchivesPlugin.phpstorage/plugins/archives/views/index.phptests/archives-plugin.unit.php
- headExtra raw-echo XSS sink: replace string-based $headExtra with structured $headLinks array in both layouts; each link's attributes are individually escaped with htmlspecialchars(..., ENT_QUOTES, 'UTF-8') — no arbitrary HTML injection possible - MIME type: change application/rdf+xml → application/xml for Dublin Core XML downloads and discovery <link> elements; oai_dc:dc is not RDF/XML and rdf+xml causes parse failures in strict clients - dc:creator crosswalk: remove 'photographer' (dead code, not in AUTHORITY_ROLES) and 'associated' (semantically wrong per DC spec) from the role filter; only role='creator' maps to dc:creator - Tests: update headExtra checks → headLinks; add negative assertion that raw headExtra echo is absent
- famname: family-type authorities now emit <famname> in EAD3 controlaccess instead of <persname>; encodinganalog stays 700 (added entry) for both person and family, 710 for corporate — unchanged - EAD3 bulk export filename: rename archives_export.ead3.xml → archives_export.xml to avoid claiming pure EAD3 compliance when the root element is the Pinakes-specific <eadlist> container
Owner
Author
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
fabiodalez-dev
added a commit
that referenced
this pull request
May 2, 2026
…AD3, fix broken links - Fix broken links: create guida/arricchimento.md and en/guide/bulk-enrichment.md - Add guida/dischi.md + en/guide/discs.md: media types, dynamic labels, Discogs/MusicBrainz/Deezer plugins - Add guida/collane.md + en/guide/series.md: series hierarchy (cycles, seasons, spin-offs) from v0.5.9.6 - Update guida/archivi.md + en/guide/archives.md: unified search (PR #120) + interoperability section (OAI-PMH 2.0, Dublin Core XML, EAD3 bulk export — PR #127) - Update _sidebar.md + en/_sidebar.md: add Dischi/Discs and Collane/Series entries - Update changelog: add v0.5.9.6 entry (series hierarchy) and "In arrivo" section (PR #118/119/120/127) - Update tecnico/requisiti.md + en/technical/requirements.md: note on zip extension installer check (PR #118)
Owner
Author
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
Missing closing `>` caused unclosed <link rel="alternate"> tags on every public archival unit detail page. Admin layout already correct.
There was a problem hiding this comment.
Actionable comments posted: 6
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@app/Views/frontend/layout.php`:
- Around line 210-221: The output loop over $headLinks renders $hl['href']
without URL-normalization/validation; update the loop that builds $out (the
foreach over $headLinks and inner foreach using $hl) to pass the href through
url() (or at minimum filter_var(..., FILTER_VALIDATE_URL)) before escaping, i.e.
compute a validated/normalizedHref from $hl['href'] and then use
htmlspecialchars(normalizedHref, ENT_QUOTES, 'UTF-8') when appending the href
attribute; ensure you only emit the href attribute when validation succeeds to
match the fix applied in app/Views/layout.php (the same $headLinks/$hl
handling).
In `@app/Views/layout.php`:
- Around line 64-75: The href attribute in the head links loop is only escaped
with htmlspecialchars but must be validated and normalized as a URL before
output; update the foreach block that processes $headLinks/$hl so that for
'href' you first verify filter_var($hl['href'], FILTER_VALIDATE_URL) (and/or
ensure an allowed scheme like http/https) or normalize via your url(...) helper,
then pass the result through htmlspecialchars(url(...), ENT_QUOTES, 'UTF-8')
before appending to $out—leave other attributes escaped as-is; ensure invalid or
non-absolute href values are skipped or sanitized to an empty value to prevent
injection.
In `@storage/plugins/archives/ArchivesPlugin.php`:
- Around line 4562-4585: oaiListMetadataFormats currently ignores an optional
identifier — update the oaiListMetadataFormats(XMLWriter $xw) implementation to
accept and check the incoming identifier parameter, lookup/validate the
corresponding item (use the existing repository/item lookup methods used
elsewhere in this class), and when identifier is present: if item not found
write an <error code="idDoesNotExist">…</error>, if item exists but has no
supported metadata formats write an <error code="noMetadataFormats">…</error>,
otherwise output only the metadataFormat elements supported for that item; keep
the current repository-wide behavior when identifier is not supplied. Ensure you
reference oaiListMetadataFormats and reuse the class’s existing item
lookup/format-detection utilities so responses conform to OAI-PMH v2.0.
- Around line 4475-4504: The <request> element is being written (via $xw on
XMLWriter) with verb/argument attributes before OAI-PMH argument validation,
causing non-conformant responses for badVerb/badArgument; fix by deferring
serialization of the <request> element until after you validate $verb and other
OAI args (or alternatively write the <request> inside each verb handler after
its validation), and ensure oaiError() responses produce a <request> containing
only the base URL (no attributes) when emitting badVerb/badArgument errors;
update the code that currently starts/writes/ends the 'request' element (the
$xw->startElement('request') / writeAttribute / $xw->endElement() sequence) so
it executes after validation or is moved into handlers, and adjust
oaiError()/error paths to render the attribute-less <request>.
- Around line 4292-4295: Il valore passato a eventdatetime/@standarddatetime può
essere in formato DB TIMESTAMP ("YYYY-MM-DD HH:MM:SS"); modifica il codice
intorno a ArchivesPlugin.php dove si costruisce $createdAt (usando
$row['created_at'], startElement('eventdatetime') e writeAttribute) per
normalizzare sempre in ISO 8601 UTC: se $row['created_at'] è in "Y-m-d H:i:s"
crea un oggetto DateTime/DateTimeImmutable con timezone UTC e formatta con
"Y-m-d\\TH:i:s\\Z", altrimenti se è già ISO lasciarlo, e infine usare quella
stringa sia per writeAttribute('standarddatetime', $createdAt) che per
text($createdAt).
In `@tests/archives-plugin.unit.php`:
- Around line 74-75: Le asserzioni sono fragili: per la prima
(decodeOaiResumptionToken) controlla non solo la presenza ma l'ordine rispetto
alla validazione di metadataPrefix—usa strpos sul contenuto sorgente per
ottenere le posizioni di "decodeOaiResumptionToken($token)" e della chiamata di
validazione del metadataPrefix e assert che la posizione di
decodeOaiResumptionToken sia minore; per la seconda (if (!$identifiersOnly) {
... $xw->startElement('record')) rimuovi la dipendenza dall'indentazione e dal
newline letterale e usa una regex permissiva o due strpos col controllo di
ordine (es. assicurati che "if (!$identifiersOnly)" compaia e che, da quella
posizione in poi, sia presente "$xw->startElement('record')" permettendo
spazi/blocchi bianchi variabili) in modo che la verifica controlli la struttura
logica non l'indentazione.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: e76ef7d0-7551-4941-829a-5a1446c5753d
📒 Files selected for processing (4)
app/Views/frontend/layout.phpapp/Views/layout.phpstorage/plugins/archives/ArchivesPlugin.phptests/archives-plugin.unit.php
- layout.php + frontend/layout.php: validate href with filter_var + scheme check before htmlspecialchars (blocks javascript:/data: URIs) - ArchivesPlugin: normalize EAD3 eventdatetime/@standarddatetime to ISO 8601 UTC via DateTimeImmutable (DB TIMESTAMP has space not T) - ArchivesPlugin: defer <request> attributes until after verb validation so badVerb responses emit attribute-less <request> per OAI-PMH §2.2 - ArchivesPlugin: oaiListMetadataFormats now accepts identifier param and returns idDoesNotExist when the record is not found - tests: replace fragile whitespace-sensitive str_contains with regex for the identifiersOnly record-wrapper assertion
Owner
Author
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
Replace existence-only str_contains with strpos order check: verify that decodeOaiResumptionToken($token) appears before 'cannotDisseminateFormat' in the plugin source, ensuring the structural invariant (decode before metadataPrefix validation) is enforced and not just presence-checked.
Owner
Author
|
@coderabbitai full review |
✅ Actions performedFull review triggered. |
…bels 'Esporta EAD3' → EN: 'Export EAD3', DE: 'EAD3 exportieren' 'OAI-PMH' is protocol-neutral, kept identical across all locales.
This was referenced May 4, 2026
Closed
8 tasks
6 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
GET /admin/archives/{id}/dc.xml) — crosswalk ISAD(G)→DC con tutti i campi disponibili;<link rel="alternate">inserito nella pagina di dettaglio per la discovery automatica (archives: Dublin Core XML export per unità archivistiche #124)GET /admin/archives/export.ead3) — esportazione dell'intera raccolta archivistica in formato EAD3 (Library of Congress); pulsante dedicato nella lista unità (archives: EAD3 export — raccolta completa delle unità archivistiche #125)GET|POST /archives/oai, endpoint pubblico) — protocol verbs: Identify, ListMetadataFormats (oai_dc+ead3), ListIdentifiers, ListRecords, GetRecord, ListSets; paginazione conresumptionTokena 100 record per pagina; filtrifrom/untilsuupdated_at(archives: OAI-PMH 2.0 — endpoint di harvesting per aggregatori nazionali #126)it_IT,en_US,de_DERoadmap correlata (future)
Test plan
GET /admin/archives/{id}/dc.xml— verifica XML Dublin Core ben formato con namespace correttiGET /admin/archives/export.ead3— download file EAD3, valida struttura XMLGET /archives/oai?verb=Identify— risposta Identify con repository infoGET /archives/oai?verb=ListMetadataFormats— restituisceoai_dceead3GET /archives/oai?verb=ListRecords&metadataPrefix=oai_dc— paginazione con resumptionTokenGET /archives/oai?verb=GetRecord&identifier=oai:pinakes:archival_unit:1&metadataPrefix=oai_dc— record singoloGET /archives/oai?verb=ListSets— rispostanoSetHierarchy<link rel="alternate" type="application/rdf+xml">presente nel<body>della pagina di dettaglio@coderabbitai full review
🤖 Generated with Claude Code
Summary by CodeRabbit
New Features
UI
Traduzioni
Tests