diff --git a/NEWS b/NEWS index d371eaea7c..65782857b8 100644 --- a/NEWS +++ b/NEWS @@ -6,6 +6,47 @@ releases. For more information about the current release, please consult RELEASE-NOTES. For more information about changes, please consult ChangeLog. +Invenio v1.1.6 -- released 2015-05-21 +------------------------------------- + +Security fixes +~~~~~~~~~~~~~~ + ++ WebStyle: + + - Adds back the `HttpOnly` cookie attribute in order to better + protect against potential XSS vulnerabilities. (#3064) + +Improved features +~~~~~~~~~~~~~~~~~ + ++ installation: + + - Apache virtual environments are now created with appropriate + `WSGIDaemonProcess` user value, taken from the configuration + variable `CFG_BIBSCHED_PROCESS_USER`, provided it is set. This + change makes it easier to run Invenio under non-Apache user + identity. + + - Apache virtual environments are now created with appropriate + `WSGIPythonHome` directive so that it would be easier to run + Invenio from within Python virtual environments. + +Bug fixes +~~~~~~~~~ + ++ global: + + - Replaces `invenio-demo.cern.ch` by `demo.invenio-software.org` + which is the new canonical URL of the demo site. (#2867) + ++ testutils: + + - Switches off SSL verification when running the test suite. Useful + for Python-2.7.9 where self-signed SSL certificates (that are + usually used on development installations) would cause apparent + test failures. (#2868) + Invenio v1.0.9 -- released 2015-05-21 ------------------------------------- diff --git a/RELEASE-NOTES b/RELEASE-NOTES index da10c7620d..93cc7042f4 100644 --- a/RELEASE-NOTES +++ b/RELEASE-NOTES @@ -1,759 +1,102 @@ --------------------------------------------------------------------- -Invenio v1.2.0 is released -March 3, 2015 -http://invenio-software.org/ --------------------------------------------------------------------- +============================ + Invenio v1.1.6 is released +============================ -Invenio v1.2.0 was released on March 3, 2015. +Invenio v1.1.6 was released on May 21, 2015. -This new stable release introduces new Invenio release series v1.2. -Invenio v1.2 brings over 1200 commits on top of Invenio v1.1, authored -by over 50 contributors. Invenio v1.2 contains many new features and -improvements, such as the new document model, the new authority record -control, and much more. Invenio v1.2 is recommended to all Invenio -installations using v1.1 or previous stable release series. +About +----- -What's new: ------------ +Invenio is a digital library framework enabling you to build your own +digital library or document repository on the web. - *) BatchUploader: apache error codes; insert or replace mode; - authorize via CIDR; add holdingpen directory; several - improvements; bibtask logs via email (#1255); multiple - improvements (#603); fix for permission checking (#1747 #1748) +This old stable release update is recommended to all Invenio sites +using v1.1.5 or previous releases. - *) BibAuthorID: user prefs and session fix; inactivation of - test_save_matrix() (#1678); merge and manage fixes; caches badly - stored in user settings; fix 'create new person' ticketing issue; - leftover print statement; disables debug output; Claiming page is - now reloaded after commit.; hepnames match; - add_cname_to_hepname_record(); hotfix in name comparison; remove - changes tempfile.rootdir; graceful external system query; adds - webuser user merge utility; fix arXiv redirect link; improvements - and bug fixes; improvements and bug fixes; DOIs from ORCID check; - WaP daemon and BAI interface fixes; fix in templates handlers; - hotfixes for authorpages and webauthorprofile daemon; Help pages - and messages; a new hope; use defaultdict from containerutils - - *) BibAuthority: new names for authority collections; source file - mode fix; separate Authorities collection (#1605); initial release - (#1602); fix for unit test suite - - *) BibCatalog: ticket_id type is now string (#2096); better error - reporting; requestor on ticket submit; ticket_submit() docstring - update (#2094); improve RT search error handling; return empty - list if no search params; RT discovery; email content cleanup; bug - fix; pylint fixes; refactoring; adds bibcatalog bin to ignored - files; add daemon task (#1528); default email backend (#872); new - email ticketing backend (#872) - - *) BibCheck: $$9 bibcheck to DOIs (#1955); improvements in DOI checks - (#1955); allow filtering by subfield contents (#2474); last_run - correct update; properly cumulates records; compatiblity with - dateutil 2.2; improve url plugin and tasklet; improve url plugin; - adds --config option; improve exception handling crossref queries; - add retry download to crossrefutils; improve doi plugin; avoid - checking dummy records; add option to consider deleted records; - new BibCheck module - - *) BibCirculation: library creation and other fixes (#2550 #2551 - #2552 #2562 #2373); fix for CERN returnees; fix for typo; missing - web tests; minor spelling error fix; fix for mandatory library - type (#1519); email ID changes and test fixes (#1479); admin guide - cleanup; patron-driven acquisition and more (#1280); personid CERN - attribute; ILL improvement; CERN LDAP improvements (#1186); set - colour of some buttons; fix for ILL title and request type; fix - for library ID variable name; various updates; fixed notes link; - code cleaning; better ill/purchase search; auto-fill for purchase - request; remove reference to apache_user; optimize CERN LDAP - query; temporary barcode for new copies; extended ILL to manage - acquisition; make statuses customizable; lots of small fixes; - 'cancelled' status for ILL request; sorting last issued loans; - edit ill request details; loan and renew process enhancement; - added budget_code to crcILLREQUEST; edit library type; arrival - date and library merge; extended item statuses; improved - book_title_from_MARC; fixed multi-barcode loan; fixed user - interface loan renewal; pylint and kwalitee cleaning; small fixes - on printing & intrface; daemon for overdue letters; email alerts - for new requests; delete a copy of a book; added CERN id in - borrower profile (#207); use new URL handler for admin pages; - avoid multiple loan creation (#305) - - *) BibClassify: ontology cache check improvement (#2672); always use - invenio code; raises an exception if rdflib is missing; unit tests - temp dir fix; remove ability to run as standalone (#1459) - - *) BibConvert: lxml support for local document() (#2497) - - *) BibDocFile: pickle support fix (#2549); decompose_file_url() and - subformat (#2556 #2557); bibdocfile.BibDoc memory fix (#2082 - #2136); change name failure raises exception (#2071); more robust - decompose_bibdocfile_url() (#1957); escape file URLs in /files tab - (#2067); fix type of bibrec-bibdoc connection (#1759); get_icon() - for smallest size icon (#1350 #1743); undefined variable fix; - register downloads with recid (#1831 #1832); new web tests; - bibdocmoreinfo query typo fix (#1706); textification with OCR fix - (#1676); get_file() exact_docformat support; display counts in - tabs; fix "delete" CLI option; no access to filesystem; preferred - extension (#1619); load plugins at global level; migration script - fix; fixes wrong variable name; error reporting changes; CERN AFS - awareness (#1388); retry mkstemp in case of failure; CERN AFS - awareness (#1388); fix for bibdoc unattached to record (#1551); - improve BibDoc display in Files tab; raise exception in - _build_file_list(); additional mimetypes support; fix version in - register_download() (#1532); fix for BibDocFile instantiation - (#1317); implements format renaming (#1318); allow doctype - renaming (#980); revert md5 property patch (#1249); new document - data model; fix for display of hidden icons; change_name missing - parameter fix (#1818) - - *) BibEdit: only notifications on error; kwalitee improvements; add - email notification on submit; user name in BibSched column; - wrongly displayed HP changeset bug; autocompletion of fields from - KBs (#1258 #73); author names into history revisions; duplicate - code removal; new RT ticket through UI dialog; int object is not - iterable fix; InvalidCache exception on clone; modal submission - preview window; check for record in BibSched queue fix; debugging - all user actions; holding pen fix; prevent deletion of managed - DOIs fields (#1445); fix revert when no 005 in history; Holding - Pen fix; add AJAX profiling option; adds affiliation guessing; - bibHOLDINGPEN from TEXT to longblob; bibupload xml file path - conflicts; support for simple ticketing; small merging fix; - BibEdit web test improvements.; fix textmarc2xmlmarc unit test; - record from history instead of bibfmt; BibCatalog and other - improvements; fixes errors in case of deleted records; several - bugfixes; moves files cache to the database; open DOI source in - new window; fixes pdf detection; several improvements; HoldingPen - multiple improvements; update - CFG_BIBEDIT_EXTEND_RECORD_WITH_COLLECTION_TEMPLATE; fixes date - parsing problem; hide authors when they exceed max number; - multiple fixes and improvements (#1190); send latest timestamp - when reverting; update admin help page and shortcuts; remove extra - holding pen call; show/hide specific parts of the record; merge - record with template; custom errors for AJAX requests; better - holdingpen integration (#87); tab switch between fields; minor - fixes; fix perform_doi_search function; refactoring and fixes; use - perform_request_search on search; add version to bibedit css name; - sort HoldingPen changes alphabetically; add direct link to - dx.doi.org; save changes periodically; avoid sync request to see - if record has pdf; change cache folder; amend textmarc to xmlmarc - unit test (#1269); import CrossRef data; improvements and fixes - (#761 #1032); css changes; allow opening deleted records (#573); - delete cache if record not modified; add extraction of references - from URL; several fixes; introduce textmarc editor; cnum - generation on conference records; multiple improvements (#696); - revert to master version (#792 #63 #118 #125); fix return binding - on Jeditable cells; fix jEditable callback when pressing return; - fix input default value; disable preview button when reverting - record; fix jEditable callback when pressing return; hide delete - record button by default; add field in specific position (#583); - fix apply all HP changes (#125); clean JavaScript code (#63); - extract css into a separate file (#118); upgrade to jQuery 1.5 - - *) BibEditMulti: only notifications on error; add email notification - on submit; adds support for hidden fields (#707); allow non- - delayed processing and priority change; several improvements; - display all MARC fields (#1489); fix for multilanguage interface - (#1331); multiple improvements and fixes (#1146 #1147 #1148 #1130 - #1149 #1156 #1158) - - *) BibEncode: support for FFmpeg >= v0.9; updated for latest - BibDocFile APIs; fix uuid Python 2.4 compatibility (#1478) - - *) BibExport: update Google Scholar exporter; hidden files and - recrawling - - *) BibField: new CFG_BIBFORMAT_HIDDEN_RECJSON_FIELDS (#2197 #2396); - better create_record error catching (#2510); fix copyright field - names (#1933); backported improvements from pu (#1687); no caching - of calculated fields; change recid field type to integer (#1633); - improvements backported from next; fix number_of_copies field - (#1625); new upgrade recipe to remove json cache; new field - filtering for `get_record`; elimination of None values in recjson; - `schema` in `split_blob`; `is_empty` update; fix for '__eq__' to - better compare recs; fix for `is_type_isbn`; continuable vs fatal - errors; bibdoc integration; new decorator @only_if_master_format; - better handling of calculated fields; fix for misbehavior when - parsing rules; new producer section added to config; clean up of - atlantis.cfg demo file (#1557); volume subfield addition; creation - date addition and keyword fix; new fft field in `atlantis.cfg`; - Python-2.4 compatibility fix (#1533); legacy_export_as_marc - escaping fix (#1509); new calculated fields; virtual field aliases - amended (#1530); new @persistent_identifier "decorator" (#1500); - bug fixes when using decorators (#1502); fix for lxml - compatibility; Python-2.4 compliance; initial release (#1300) - - *) BibFormat: new BFO for authority records (#1699 #1749); links to - public resources of authors (#1700 #1749); better display of - authority records (#1749 #1699); ORCID display for authors; - removal of obsolete BFX engine (#2563 #2124); recjson update using - bibreformat (#1708 #2220); PEP8 fixes in bibreformat (#2220); add - sponsor information to copyright (#1975); larger column - `format.code` (#2072); advertise /doi URL in DC output; improve - Dublin Core output (#320 #1213); configurable /record; new - bfe_oai_identifier element; new bfe_date parsing/formatting - options; plot file identification improvements (#1514); fix for - eval_format_element return type; file rights fix; new - bfe_arxiv_link; affiliation improvements; left over print in - format_record; reworks exceptions handling; only save default site - lang on the fly; look for missing caches by default; fix snippet - generation; needs_2nd_pass in bibreformat; duplicate code in - bibreformat.py; improves and updates bfe_plots; fixes tests; fix - external function for libxslt; improve Google Scholar support - (#1513); Displays the DOI in the EndNote; dublin core export now - includes DOI; to fixup to removes old php format; small fixes; - several fixes; fix in date comparison; second formatting pass - (#1464); lazy missing formats updates; empty record check; - progress display improvement; initial example of Twitter Card - support; fix for snippet generation; author links for mobile app; - initial release of mobile app formats; new Solr fulltext snippet - facility (#1301); QR-code format element (#1441); add DataCite XSL - stylesheet; remove 0248_a field from title; fixes last run date - for HDREF (#1236) - - *) BibIndex: ambiguous SQL query fix for MariaDB-5.5 (#2759); - tag.recjsonvalue NOT NULL (#1947 #2259); fix new-old record - incremental indexing (#2693); clean up after authority regression - test (#2448); author ID performance improvements (#1952); upgrade - recipe for `tag.recjson_value`; recjson fields in admin interface; - indexing non-MARC standards; abstraction layer for terms - retrieval; WordTable API changes; move helper functions to utils - file; changes in WordTable argument list; PEP8 compliance fixes; - fix for virtual index filtering; new DOI index (#1655); virtual - index queue dupe optimisation; new 'all-virtual' CLI option; minor - engine refactoring; documentation for virtual indexes; new pattern - for tokenizer inheritance (#1704); new abstract class for indexes; - separate class for virtual index (#1661); common words in virtual - index (#1653); fixes admin regression tests; - BibIndexDefaultTokenizer upgrade; bad word check optimized; - consistency check optimizations (#1436); ingestion health and - "unneeded" indexes (#1632); index type in admin interface; virtual - global index (#1574); indexing only affected indexes (#1573); - clean warning messages in test suite (#1615); filetype and - itemcount tokenizer fix (#1609); new index 'filetype' (#473); new - index 'itemcount'; tokenisation of authority records; fix syntax - error in bibstat; support for CJK languages (#285); pluginutils - for tokenizers (#852); centralisation of tokenizers (#852); new - regression tests; centralisation of LaTeX/HTML treatment (#852); - centralisation of stopword treatment (#852); centralisation of - synonym treatment (#852); fix for external fulltext indexing; - rework of error handling (#1075); move of text extraction to - BibDocFile; new exacttitle index (#1397); new filename index - (#1717) - - *) BibKnowledge: searchtype parameter in KB export (#2570 #2581); fix - get_kbt_items_for_bibedit (#1879 #1895); lxml port - get_kbt_items_for_bibedit - - *) BibMatch: allow tests to login over plain http; Fix validator - problem; use other author comparison function; more print - statements; improves get_longest_words; improve fuzzy queries; - validation fixes - - *) BibMerge: adds CFG_SITE_RECORD as script data (#2580 #2237); - `onclickSubmitButton` missing comma fix (#2230); prevent loss of - DOI when merging records (#1446); delete cache of master record - before submission; change order of updates; add subfield sorting - and interface fixes; several fixes; add 981__a field to master - record; delete duplicate record first (#1645) - - *) BibRank: fix path for download history graph (#2554 #2374); fix of - similar-to-recid result order (#1745 #2236); missing selfcites for - collaborations; record ID citations catchup; citation blobs in - Redis (#1689); adds a new option to disable bibsort (#1617); minor - refactoring in word ranking; handle missing files when removing - graphs; more leanient date handling in citation graphs; more - leanient date handling in citation graphs; remove outdated import - in citations tab; exception when gnuplot is not available; unicode - recid in citation indexer; Added ISBN, recid and HBL identifiers; - fix bad variable name; fixes for sorting; missing drop table - rnkSELFCITEDICT; new way to generate graphs (#1244); consider only - one year in citation graph; fixes for cited by sort; function to - get citations of a single record; fixes sorting; optimized cited - by sorting; filter citations on collections (#1504); logging of - citation changes (#1426); store selfcites in a table (#1417); - citesummary optimizations (#1481 #1217); handle records with - mulitple journals (#1394); optimized cited by sorting; no citerank - error when no citations (#1624); better Solr regression tests; - faster Solr indexer; new multi-tag Solr indexer; index latest - records first in Solr; increase rnkDOWNLOADS.file_format size; - self cites upgrade recipe improvement (#1482); detect external - word similarity ranker; storing citation indexer warnings in DB - (#1210); optimisations in citation algorithm (#1073); selfcites - fixes; fix for citation indexer checks; citation indexer sanity - checks and alerts (#1091); reference linking improvements (#950); - citation indexer date check change (#946); fix for missing Python - files in Makefile; two algorithms for self-citations (#945); - change import to defining module; better exception handling in - Solr indexer (#1199); better default mode in Solr indexer (#1192); - more invalid Solr character replacements; new Solr and Xapian - ranking bridge (#1084 #1168) - - *) BibRecord: namespaces ignored for lxml (#2604); search & compare - subfields; new API records_identical(); new API - identical_records(); record_get_field_values with filtering - (#1550); filter field instances (#1550) - - *) BibReformat: chunking of updated records query - - *) BibSched: email-logs-on-error parameter (#2205); check schSTATUS - when detecting status; pep8 for bibtask.py; pep8 for bibched.py; - subdirs for bibsched logs; fixes a bug with --profile cli option; - fix priority for the same sequence id; increase max log file size - to 5Mb; display mode for non-periodical tasks; adds more task - changing commands; get_modified_records_since() (#1538); monitor - auto mode selection bug; invalid sql in monitor history tab; - setting to continue on errors; many improvements (#1177 #991); - error when switching to manual; refactoring and improvements - (#1274 #1275 #1449); enhanced write_message(); motd update check; - problem parsing task CLI options (#1330); interface responsiveness - improvements (#1303); priority in automatic mode; CLI-started - tasks host field; kwalitee fixes; shell output leakage upon task - kill (#1343); single error reporting (#1342); scheduling algorithm - improvements (#1281); fixes task chain-sleeping (#1304); fixes - monotasks for multi-node (#1304); fixes for multi-node setup - (#925); new --email-logs-to bibtask CLI (#1252); subprocess - instead of deprecated popen2; new web UI for BibSched live view - - *) BibSort: improved washers (#2283 #1754); add check before - deleting; fix typos and CLI arguments - - *) BibUpload: creation_date based on incoming 005 (#2693 #1604 - #2684); faster recjson deletion after updates (#1708); no reload() - in regression tests (#1702); --append only new fields (#1440); - removed print statement; do not always process MoreInfo; - CFG_BIBUPLOAD_EXTERNAL_OAIID_TAG fix; new - CFG_BIBUPLOAD_MATCH_DELETED_RECORDS (#1438); affected fields and - insert mode; 8564_s support; less useless computation; less - verbose; messages cleaning; ingore 856 tag order in conflict - (#1606); smarter conflict report; smarter conflict detection; no - tickets in pretend mode; ticket creation fix; improves utf-8 - checks; task error messages (#1449); utf-8 encoding; encoding - checks; regression tests cleanup; matching existing records - (#1438); pretend holding pen fix (#1618); fill affected_fields in - hstRECORD (#1572); fix for inserts with 005 (#1595); conflicting - revision ticket queue; smart record uploader fixes; BibCatalog - connection; sensible history and other goodies (#498 #1250 #871); - bibrec timestamp bug (#1431); smart record uploader (#816 #864 - #897); check DOI uniqueness (#1160) - - *) DocExtract: new CMS PAS report numbers; additional report numbers; - extract page-end from references; removes stdout ouput from tests; - rework of regression tests; fixes regression tests; improves - bibrecord; increases compiled regexp cache size; preload - docextract author regexp; using -i instead of -r; preload kbs on - wsgi load; re-enable caching of kbs; outdated import in webtool; - reduce verbose in tests; fixes DESY-THIS rn recognition; 5 digits - arxiv numbers detection; do not create old tickets; lower - bibupload priority; webinterface text box fix; optional unidecode - dependency; help messages & compatibility warnings (#1220); - several improvements; move mislabelled regression tests (#1309); - journal rawref search fix (#1306); nose-friendly refextract tests; - fix reference extractor unit tests; refextract unit tests file - name fix; preparing for merging into master; multiple fixes (#966 - #958); new docextract and refextract modules (#944 #1014) - - *) HepData: updates to formats; fixes unit tests; clean hepdata.js - inclusion; new HepData module; adds hepdataharvest bin to ignored - files - - *) HepNames: update form migrated to INSPIRE - - *) I18N: PO file update for the release of v1.2.0; more complete - POTFILES.in; fix wrong msgids in Persian translation; updates to - the Persian translation; POTFILES.in update; initial Persian - (Farsi) translation; infrastructure for Persian (Farsi); several - fixes in Spanish translation; Catalan and Spanish updates to - Search Guide; Catalan and Spanish updates to Search Tips - - *) InvenioConnector: allow logins over plain http; fix for CDS - authentication - - *) OAIHarvest: fix identifier parsing (#2408); conversion argument - name upgrade (#1753); error reporting fix (#1804 #1812); respect - hidden fields; do not launch BibIndex when done; bibindex priority - to 4; only update lastrun on successful harvest; small daemon - enhancement; fixes missing import; priority of single harvest - tasks; improves arXiv identifier harvesting; several improvements - (#547); sample OAI-ArXiv conversion update (#678); CERN-specific - "arXiv" doctype; consider source_id for selective post-processing; - configurable selective post-processing (#1477) - - *) OAIRepository: lower priority to updating uploads (#2525); fix for - hidden OAI tags (#2642); more lenient time limit for tests; do not - report cache not found errors; allows running slow machine; - oai_get_recid() for merged/deleted records (#1429); marcxml - created in shared directory; forcing clients to re-harvest (#1218) - - *) PdfChecker: clean up after regression tests (#2448); log full list - of updated recids; skip records without unique ids in 037 tags; - new module for arxiv pdf checker - - *) RefExtract: avoid double encoding (#2602); refactored book - handling; improved book search; addresses warning in tests; - removes leftover print; only accpet digits as numeration; allows - more lines between title and numeration; changes condition for - ticket creation; improves docextract, refextract - - *) SolrUtils: fix of similar-to-recid result order (#1745 #2236); - better exception handling in indexers; faster snippet factility; - support entire full-text indexing; cleaner schema.xml; support - high count of logical clauses - - *) WebAccess: automatically fetch SSO nicknames (#2583); CERN- - specific authorization message; fixes user details page links; - CFG_ACCESS_CONTROL_LEVEL_SITE=1 support (#1501); remove Facebook - testing credentials; update check for "external" account at CERN; - ORCID support (#1124); OpenID and OAuth authentication (#1124); - new CERN auth method support - - *) WebAlert: update tests for newly introduced records - - *) WebApiKey: unit Vs. regression tests - - *) WebAuthorList: fix import from recid; fix import from record id; - ignore empty affiliations when exporting; add new author list - manager tool - - *) WebAuthorProfile: reenable profile pages; disable if not - available; compatiblity with atlantis; fixes unit tests; recompute - link as a post action; new regression test suite - - *) WebBasket: 'move item' improvements; new 'move item' functionality - (#1547); Create Basket link in the main display (#1333); correct - referer when adding to basket (#1194); fix copying external items - - *) WebComment: fix for get_first_comments_or_remarks (#2522 #2523); - more prominent subscription link (#2434); deleted record message; - "Your Comments" page; link to "Your Comments" after posting; "Your - Comments" page (#974) - - *) WebJournal: Indico seminars widget improvement (#1980 #1981); fix - image dimension retrieval; sample Twitter Card markup; new image - template; better exception handling when caching; structured cache - (#1544); CERN-specific fix; fix for what's new widget test case - - *) WebLinkback: safer notification email; clean regression test suite - (#1285); fix for importing CFG_DATABASE values; pending linkback - notification emails (#1247); minor improvement; better - documentation; module optional (#1245); better global /linkbacks - page; better /linkbacks tab display; better URL title display; fix - for DB name in regression tests; fix for user_info passing; unit - test module rename; auto-increment regression test fix (#1136); - initial release (#627 #857 #1136); fix truncated FSF address in - docstring - - *) WebMessage: English corrections in output messages (#1849) - - *) WebSearch: optional refersto/citedby record limit (#2711); removal - of hard-coded Holdings tab (#2592 #2664); new test case for - pattern-limit queries (#1750 #1751); search results pattern limit - fix (#1750 #1751); proper re-raise in RSS handling (#2084 #2598); - fix for the number of printed records (#2512); inverted collection - scores (#2058); stemming and '*' (#2468); smarter journal hint - (#2352); new Journal Hint Service (#2352); kwalitee fix (#2352); - richer `/record` and `/search` API docs (#2303); fix for record - numbering in pagination (#1762 #1763); new Add-to-Search Interface - (#622 #271 #1738); CERN-specific video latest additions (#2068); - CERN-specific lecture latest additions (#2068); improve detection - of record owners (#2068); better retrieval of record tabs (#2068); - fix IndexError in is_hosted_collection (#1764); CERN-specific hack - for latest additions (#1976); CERN-specific collection sorting - (#2017); fix for 'rhs is of unknown type' (#1819); resolve - (internal) DOIs (#1322); anyfield in CFG_WEBSEARCH_SYNONYM_KBRS - (#1493); faster collection children cache (#1739); initial support - for recjson output; fix detailed record page tab tracebacks; - update collection page markup; fix search URL in timeout message; - CERN-specific collection sorting; better sort order in citation - tabs (#1307); timestamp detection fix for empty sites; sorting - fixes (#1674); reverse order and scores; rg parameter with - of='id'; reworks async downloader; wgsi.errors in fake request; - handle case in /collections/ urls; fixes - regression tests; stdout.flush conflicting with mod_wsgi; changes - citation tab count; search API changes for record sorting (#1657); - fix sorting options (#801); spires date parsing errors fixes; no - 'back to search' on empty session; display deleted records in - citation log; make query parser use Invenio datetime; fix mixed - parameter for re.sub(); bibfmt on innodb; outdated import in - citations tab; takes into account new record in tests; fix HepData - templates; change CFG_SITE_URL to CFG_BASE_URL; fixes search bug - with --empty hitset; add cataloguer: search unit; SPIRES date - format 11/93; change CFG_SITE_URL to CFG_BASE_URL; testing fixing - subject lookup; add record edit link in brief; pep8 fixes; INSPIRE - vol to use volume field; INSPIRE texkey in 035__%; handles - selfcites searches; spires syntax and quotes; INSPIRE fulltext - warning update; find doi in search engine (#1051); CV output - formats (#314); correct record sums from hosted colls (#1651); - search services (#1278); custom i18n collection boxes (#1286); - item count regression test activation; fix for hidden-field admin - access test; support for intbitset output format (#1460); fix - missing cc info in req object; CERN-specific hack update; display - number of hits in mobile output; Greek translation of Search - Guide; fix Python 2.4 syntax error; most popular field values - optimisation (#1096); fix Search Guide reformatting; fix Search - Guide mismatch tags; Search Guide reformat and pretty-print; fix - browsing deleted/restricted records (#1292); webcoll performance - improvements; permitted restricted colls for guests; empty unit - test suite for summarizer; removal of excess summariser tests; - revamping of citesummary pages (#134); summarizer unit test - update; fix searching with limits; enforcement of record view - restrictions; translation-friendly overview box; restricted record - search improvements; fix for regression test link targets; fix for - double display of the footer; better restricted collection search - (#1161); "p=el*;rm=citation" test inactivation (#1174); sorting - test amendments; mixed ranking/sorting test amendments; wildcard - limit parameter is 0 in p_r_s; bugfix for empty set sorting; - refactored perform_request_search() (#542); add regression tests - for "em" parameter; add "em" parameter; include 'cc' in RSS - 's (#2013 #2014); fix for "--language" option - (#1399 #2219) - - *) WebSession: no differentiation between guests (#2786 #2813); CSRF - token in profiling settings (#1855); disable ORCID login (#1667); - user preference to enable profiling; new Redis session storage - backend (#1688); fixes session_cleanup; session_param_get() - default value (#1294) - - *) WebStat: fix for custom query summary graph (#2553 #2375); default - query in the Custom query summary (#2388); list link fix for - system health UI page (#1713); ingestion health monitor fix - (#1631); use Invenio instead of CDS in pages; new ingestion - monitor (#936); no wildcard limit for custom summary data; add - bibcirculation config variables - - *) WebStyle: richer documentation on record page tabs (#216); ping - handler returning 200 status code (#2700); POST handling fix - (#1951); req object with no headers; fix gotoadmin CLI parmeters - parsing (#1427); move charset higher in the document; move of lang - and dir attributes to html; fixes for /goto CERN-HR plugin - example; silence client disconnected errors; blocking read in - handle_file_post; add missing icons to Makefile; - WebInterfaceDisabledPages(); quote canonical and alternate URLs - (#1515); /info pages using webdoc infrastructure; more accurate - "Restricted" flag display (#867); inactivate two regression tests - (#1293); goto engine typo fix; canonical and alternate URLs (#1251 - #368); new /goto URL handler (#1178); memory leak fix in session - handling (#571) - - *) WebSubmit: Set_Embargo optional and functional (#2699); link to - successfully created record (#1641); more robust JavaScript check - (#1741); print white space instead of None (#1741); support for - elements' custom_level (#1741); `test_revise_picture_admin` test - fix (#2142); `deferRelatedFormatsCreation` param fix (#2142); - Link_Records error message fix (#1734); fix access restrictions in - /uploadfile (#1703 #2066); allow record owners to upload files; - allow image conversion of .tiff (#1909); grant access to the - superadmin as owner (#2065); doilookup function in webinterface - (#2025); guests support improved in /direct URL (#1240); rotate - created icons according to EXIF (#1516); web tests for DEMOART and - DEMOPIC; DEMOART uses bibdocfile_managedocfile; skip pdf - optimization if pdfopt is missing; fixes for openoffice handling; - INSPIRE specific amendments; add traceback info on error; new - Run_PlotExtractor function (#1506); source file mode fix; login - offer to guests on action page; fix for icon creation for - bibdocfiles; jquery-ui update for photo submission - - *) containerutils: new Python-2.4 defaultdict - - *) crossrefutils: new Fundref-based APIs - - *) dataciteutils: refactor DataCite API wrapper (#1457); DataCite DOI - support and test cases - - *) dateutils: adds __add__ to our custom datetime; strptime for - Python-2.4 compatibility; fix for strftime() function; consolidate - localtime_to_utc; day ranges; fix for unit test suite; new - get_time_estimator function - - *) dbdump: partial dumps; ignore with regexp (#579); dump on detached - slave (#1282); fix compress mode; add option to ignore tables; add - slave support; improve error handling - - *) dbquery: fix for importing CFG_DATABASE values; more reinstall- - friendly dbquery - - *) demo site: fix double 245 MARC field - - *) docker: more complete configuration - - *) errorlib: Sentry logging improvements (#2535 #2546); tags context - fix for sentry (#2623); fix Sentry context syntax issue (#1960 - #2147); context support in sentry (#1960 #2147); support for - Sentry logging (#1726); makes SMS messages shorter; time - independent tests; hostname in error notifications (#1546); wrap - warnings to invenio.err (#1616) - - *) filedownloadutils: add verbose to download_url(); utility for file - retrieval (#1076) - - *) general: new CFG_SCOAP3_SITE flag; optional remote debugger; test - fixes; Propagating exceptions in debug mode; unit-tests fixes - - *) git: ignore KDevelop4 project files - - *) global: PEP-8 style in block comments (#1904); test suite original - modification date fix (#2737); removal of INSERT DELAYED SQL - statements (#2268 #2269); removal of leftover files; - InvenioTestCase in test suite (part 2); InvenioTestCase in test - suite; cdsweb.cern.ch becomes cds.cern.ch - - *) htmlutils: render MathML by MathJax; improve js string escaping - - *) importutils: fix None values error; Makefile clean up; Python 2.4 - support and test case; initial release - - *) installation: new release_1_2_0 upgrade recipe; - 2015_03_03_tag_value upgrade recipe; 2013_09_16_aidPERSONIDDATA - fix; 2014_08_12_format upgrade recipe fix; all upgrade recipes in - tabcreate (#1753); richer uninstall-jquery-plugins (#2418); - python-twitter requirement update (#2015); lxml recommended; - location of demo_table_jui.css; location of jquery.omniwindow.js; - location of jquery.blockUI.js; location of sly.min.js; location of - parsley.js; Redis and Nydus pre-requisites; jinja2 prerequisite - (#1677); move h5py to extra requirements; h5py requirement clean- - up; update to MathJax-2.3; add h5py dependency; use custom faster - jeditable; ColVis.js on invenio-software.org; table creation fix; - Python-2.6 and pip requirements; fix for BibAuthority upgrade - recipe; fix for rnkDOWNLOADS upgrade recipe; help for BibAuthorID - email settings; support for Apache-2.4 (#1552); help for `--load- - bibfield-conf` step; oaiREPOSITORY_last_updated upgrade; fix for - table drops and upgrades; selfcites upgrade recipe add-on; more - gentle idxINDEX.indexer recipe; maint-1.1-to-master upgrade recipe - (#1198); fix for 2012_10_29 upgrade recipe; fix for duplicate /css - alias - - *) intbitset: initialization from iterator (#1698); no crash when - intbitset is on rhs (#1287); atomic installation; union() and - isdisjoint() support; type checking for operators - - *) inveniocfg: adds option to failt tests on first error; restore - wrapping showarning after running unit tests; do not capture - warning in unit tests; workaround bibfmt corruption; fixes - BibSched check in upgrader; new derived config CFG_BASE_URL - - *) inveniogc: guest users gc optimization (#428 #1950); clean up gc - tasks (#1950); delete refextract logs after 7 days (from 28); - BibEdit related improvements.; add new session deletion mode; - delete BibEdit temporary files - - *) kwalitee: even stricter PEP-8 compliance - - *) mailutils: better email header type detection (#2713); support - invalid senders (#2256 #2385); fix for send_email() error on DEV - site (#1744); extend send_email with BCC option; send_email() with - attachments (#1253) - - *) mathpreview: js-based math preview panel (#1221) +Security fixes +-------------- - *) oaiharvest: fixes harvest() web interface (#2524) ++ WebStyle: - *) plotextractor: recid parsing fix (#2566); sanity in plotextractor - tests; do not add FFT if there is no location; remove dummy - caption generation; fixes arg parsing and more; more shell - argument escaping; process files of a record; fix CLI parameters - parsing + - Adds back the `HttpOnly` cookie attribute in order to better + protect against potential XSS vulnerabilities. (#3064) - *) redisutils: initial release +Improved features +----------------- - *) sequtils: increases size of seqSTORE.seq_value; no texkey if no - year; increases size of seqSTORE.seq_value; fix texkey generation; - add start_date parameter to CnumSeq; wait for BibUpload to finish; - new seq generator for texkeys ++ installation: - *) shellutils: Mac OS compatibility (#1184) + - Apache virtual environments are now created with appropriate + `WSGIDaemonProcess` user value, taken from the configuration + variable `CFG_BIBSCHED_PROCESS_USER`, provided it is set. This + change makes it easier to run Invenio under non-Apache user + identity. - *) solrutils: clean unit and regression test suite (#1284); add - search and ranking tests; fix for ranking result display; better - collection filter generator; removal of unused code; better - invalid character handling (#1197); add documentation + - Apache virtual environments are now created with appropriate + `WSGIPythonHome` directive so that it would be easier to run + Invenio from within Python virtual environments. - *) testutils: wait for element to be displayed/hidden; default to - assertEqual in py26; add new relative url function; - regression_tests fix; new JavaScript unit test framework +Bug fixes +--------- - *) textmarc2xmlmarc: remove content regexp check (#1267) ++ global: - *) textutils: wash_for_utf8() simplification (#1755); - translate_to_ascii() unknown chars fix (#1754); show_diff() API - clean-up (#1465); fix old import statement; sharp-s to ss; - unidecode verision; add ALA-LC transliteration (#1092); create - function to show diff view + - Replaces `invenio-demo.cern.ch` by `demo.invenio-software.org` + which is the new canonical URL of the demo site. (#2867) - *) urlutils: new function get_relative_url(); use hashlib instead of - md5 if possible ++ testutils: - *) xmlmarc2textmarc: order only by tags + - Switches off SSL verification when running the test suite. Useful + for Python-2.7.9 where self-signed SSL certificates (that are + usually used on development installations) would cause apparent + test failures. (#2868) -Download: ---------- +Download +-------- - - - +- http://invenio-software.org/download/invenio-1.1.6.tar.gz +- http://invenio-software.org/download/invenio-1.1.6.tar.gz.md5 +- http://invenio-software.org/download/invenio-1.1.6.tar.gz.sig -Installation notes: -------------------- +Installation +------------ Please follow the INSTALL file bundled in the distribution tarball. -Upgrade notes: --------------- +Upgrade +------- -We recommend to test the upgrade on a development server first. Please proceed as follows: - a) Stop your bibsched queue and your Apache server. +a) Stop your bibsched queue and your Apache server. - b) Install the update: +b) Install the update:: - $ tar xvfz invenio-1.2.0.tar.gz - $ cd invenio-1.2.0 + $ tar xvfz invenio-1.1.6.tar.gz + $ cd invenio-1.1.6 $ sudo rsync -a /opt/invenio/etc/ /opt/invenio/etc.OLD/ - $ sh /opt/invenio/etc/build/config.nice # (1) + $ sh /opt/invenio/etc/build/config.nice $ make - $ sudo -u www-data make check-upgrade - $ sudo -u www-data make install # (2) - $ sudo diff -r -u /opt/invenio/etc.OLD/ \ - /opt/invenio/etc/ # (3) + $ make check-upgrade + $ sudo -u www-data make install + $ sudo rsync -a /opt/invenio/etc.OLD/ \ + --exclude invenio-autotools.conf \ + /opt/invenio/etc/ # (1) $ sudo -u www-data /opt/invenio/bin/inveniocfg --update-all - $ sudo -u www-data /opt/invenio/bin/inveniocfg --upgrade # (4) - - Notes: - - (1) If you are upgrading from previous Invenio stable release - series, you may need to upgrade your system packages. For - example, Python-2.6 is now the minimal required Python - version. Hence you may need to run: - - $ sudo pip install -r requirements.txt - $ sudo pip install -r requirements-extras.txt - $ ./configure ... - - as indicated in the INSTALL file. - - (2) After installing new files from the release tarball, you will - most probably also want to upgrade your jQuery and other - runtime dependencies, by running: - - $ sudo -u www-data make install-jquery-plugins - - and similar statements as indicated in the INSTALL file. - - (3) Due to many new features brought by this new release series, - please study the configuration changes carefully. You may - need to adapt several CFG variables, `etc` files, and any - local templates you may have customised previously. We - recommend to test the upgrade carefully on a development - server. + $ sudo -u www-data /opt/invenio/bin/inveniocfg --upgrade - (4) This command will perform the database upgrade, which may take - considerable time due to upgrading the document model. Note - that the command will also inform you about any further - actions you should take, for example you may be asked to run: + (1) If you are upgrading from previous stable release series + (v0.99 or v1.0), please don't run this rsync command but + diff, in order to inspect changes and adapt your old + configuration to the new Invenio v1.1 release series. - $ inveniocfg --load-bibsched-conf - $ inveniocfg --reset-recjson-cache - $ webcoll -u admin -f +c) Restart your Apache server and your bibsched queue. - c) Restart your Apache server and your bibsched queue. +Happy hacking and thanks for flying Invenio. -- end of file - +| Invenio Development Team +| Email: info@invenio-software.org +| IRC: #invenio on irc.freenode.net +| Twitter: http://twitter.com/inveniosoftware +| GitHub: http://github.com/inveniosoftware +| URL: http://invenio-software.org