Repository: tika Updated Branches: refs/heads/TIKA-1508 03d38248f -> 01320372f
TIKA-2006 -- add mime definitions for iCal and vCalendar Project: http://git-wip-us.apache.org/repos/asf/tika/repo Commit: http://git-wip-us.apache.org/repos/asf/tika/commit/4d308fd7 Tree: http://git-wip-us.apache.org/repos/asf/tika/tree/4d308fd7 Diff: http://git-wip-us.apache.org/repos/asf/tika/diff/4d308fd7 Branch: refs/heads/TIKA-1508 Commit: 4d308fd7015391c9cdfd13ba6990dcd6e8496138 Parents: 06633cc Author: tballison <[email protected]> Authored: Wed Jun 15 07:13:15 2016 -0400 Committer: tballison <[email protected]> Committed: Wed Jun 15 07:13:15 2016 -0400 ---------------------------------------------------------------------- CHANGES.txt | 2 ++ .../org/apache/tika/mime/tika-mimetypes.xml | 11 +++++++++++ .../java/org/apache/tika/mime/TestMimeTypes.java | 6 ++++++ .../test/resources/test-documents/testICalendar.ics | 15 +++++++++++++++ .../test/resources/test-documents/testVCalendar.vcs | 10 ++++++++++ 5 files changed, 44 insertions(+) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/tika/blob/4d308fd7/CHANGES.txt ---------------------------------------------------------------------- diff --git a/CHANGES.txt b/CHANGES.txt index 6008b51..59d2451 100644 --- a/CHANGES.txt +++ b/CHANGES.txt @@ -1,5 +1,7 @@ Release 1.14 - ??? + * Add mime definitions of iCal and vCalendar (TIKA-2006). + * Upgrade to PDFBox 2.0.2 (TIKA-1996). * Add configurable maximum threshold for number of events extracted http://git-wip-us.apache.org/repos/asf/tika/blob/4d308fd7/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml ---------------------------------------------------------------------- diff --git a/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml b/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml index 8a79844..ca84d94 100644 --- a/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml +++ b/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml @@ -5494,8 +5494,14 @@ </mime-type> <mime-type type="text/calendar"> + <magic priority="50"> + <match value="BEGIN:VCALENDAR" type="string" offset="0"> + <match value="VERSION:2.0" type="string" offset="15:30"/> + </match> + </magic> <glob pattern="*.ics"/> <glob pattern="*.ifb"/> + <sub-class-of type="text/plain"/> </mime-type> <mime-type type="text/css"> @@ -6250,6 +6256,11 @@ </mime-type> <mime-type type="text/x-vcalendar"> + <magic priority="50"> + <match value="BEGIN:VCALENDAR" type="string" offset="0"> + <match value="VERSION:1.0" type="string" offset="15:30"/> + </match> + </magic> <glob pattern="*.vcs"/> <sub-class-of type="text/plain"/> </mime-type> http://git-wip-us.apache.org/repos/asf/tika/blob/4d308fd7/tika-parsers/src/test/java/org/apache/tika/mime/TestMimeTypes.java ---------------------------------------------------------------------- diff --git a/tika-parsers/src/test/java/org/apache/tika/mime/TestMimeTypes.java b/tika-parsers/src/test/java/org/apache/tika/mime/TestMimeTypes.java index 46d3f5d..3f22842 100644 --- a/tika-parsers/src/test/java/org/apache/tika/mime/TestMimeTypes.java +++ b/tika-parsers/src/test/java/org/apache/tika/mime/TestMimeTypes.java @@ -1031,6 +1031,12 @@ public class TestMimeTypes { assertTypeByData("application/pkcs7-signature", "testPKCS17Sig-v3.xml.p7m"); assertTypeByData("application/pkcs7-signature", "testPKCS17Sig-v4.xml.p7m"); } + + @Test + public void testVandICalendars() throws Exception { + assertType("text/calendar", "testICalendar.ics"); + assertType("text/x-vcalendar", "testVCalendar.vcs"); + } private void assertText(byte[] prefix) throws IOException { assertMagic("text/plain", prefix); http://git-wip-us.apache.org/repos/asf/tika/blob/4d308fd7/tika-parsers/src/test/resources/test-documents/testICalendar.ics ---------------------------------------------------------------------- diff --git a/tika-parsers/src/test/resources/test-documents/testICalendar.ics b/tika-parsers/src/test/resources/test-documents/testICalendar.ics new file mode 100644 index 0000000..cbd9bee --- /dev/null +++ b/tika-parsers/src/test/resources/test-documents/testICalendar.ics @@ -0,0 +1,15 @@ +BEGIN:VCALENDAR +VERSION:2.0 +METHOD:PUBLISH +BEGIN:VEVENT +DTSTART:20010701T213000Z +DTEND:20010701T223000Z +LOCATION:Banque Nationale Stage () +UID:[email protected] +DTSTAMP:20141020T130403Z +SUMMARY:Susi Hyldgaard - Festival International de Jazz de Montréal +DESCRIPTION:SUSI HYLDGAARD\n\nAs the new millennium opens, the prevailing style is the mélange... but you still have to know how to measure out the ingredients. Susi Hyldgaardâs facility with jazz and world music is stunning the critics. Want proof? Sheâs been compared to Björk, Neneh Cherry and Cassandra Wilson but the singer/pianist/accordionist sets herself apart by virtue of a thoroughly personal style, spontaneous and blazingly passionate. Without a doubt, one of the most striking talents to emerge from the Danish music scene in the last five years. \n\nhttp://www.montrealjazzfest.com/program/concert.aspx?id=3579\n +PRIORITY:5 +CLASS:PUBLIC +END:VEVENT +END:VCALENDAR \ No newline at end of file http://git-wip-us.apache.org/repos/asf/tika/blob/4d308fd7/tika-parsers/src/test/resources/test-documents/testVCalendar.vcs ---------------------------------------------------------------------- diff --git a/tika-parsers/src/test/resources/test-documents/testVCalendar.vcs b/tika-parsers/src/test/resources/test-documents/testVCalendar.vcs new file mode 100644 index 0000000..b6ea937 --- /dev/null +++ b/tika-parsers/src/test/resources/test-documents/testVCalendar.vcs @@ -0,0 +1,10 @@ +BEGIN:VCALENDAR +VERSION:1.0 +BEGIN:VEVENT +DTSTART:20121201T210000 +DTEND:20121201T220000 +LOCATION;ENCODING=QUOTED-PRINTABLE:Mississippi Studios (3939 N Mississippi Ave., Portland, OR) +DESCRIPTION;ENCODING=QUOTED-PRINTABLE: +SUMMARY;ENCODING=QUOTED-PRINTABLE:Battleme, My Goodness, the Ax +PRIORITY:3END:VEVENT +END:VCALENDAR \ No newline at end of file
