more fixed.

diff --git a/magic/magic.mime b/magic/magic.mime
index 781528621609854d44ce46605cc84f0e3a422e3b..c4591b8b1da4286ca267f8b1303370c7ff1b8b1a 100644 (file)
--- a/magic/magic.mime
+++ b/magic/magic.mime
@@ -99,9 +99,7 @@
 
 # svg
 
-0      string          \<?xml
-#                      text/xml
->38    string          \<\!DOCTYPE\040svg      image/svg+xml
+38     string          \<\!DOCTYPE\040svg      image/svg+xml
 
 
 # xml
@@ -111,8 +109,8 @@
 #------------------------------------------------------------------------------
 # Java
 
-0      short           0xcafe
->2     short           0xbabe          application/java
+0      beshort         0xcafe
+>2     beshort         0xbabe          application/java
 
 #------------------------------------------------------------------------------
 # audio:  file(1) magic for sound formats
@@ -135,7 +133,7 @@
 # DEC systems (e.g. DECstation 5000) use a variant of the Sun/NeXT format
 # that uses little-endian encoding and has a different magic number
 # (0x0064732E in little-endian encoding).
-0      lelong          0x0064732E      
+0      lelong          0x0064732E
 >12    lelong          1               audio/x-dec-basic
 >12    lelong          2               audio/x-dec-basic
 >12    lelong          3               audio/x-dec-basic
@@ -148,25 +146,25 @@
 
 # Bytes 0-3 of AIFF, AIFF-C, & 8SVX audio files are "FORM"
 #                                      AIFF audio data
-8      string          AIFF            audio/x-aiff    
+8      string          AIFF            audio/x-aiff
 #                                      AIFF-C audio data
-8      string          AIFC            audio/x-aiff    
+8      string          AIFC            audio/x-aiff
 #                                      IFF/8SVX audio data
-8      string          8SVX            audio/x-aiff    
+8      string          8SVX            audio/x-aiff
 
 
 
 # Creative Labs AUDIO stuff
 #                                      Standard MIDI data
-0      string  MThd                    audio/unknown   
+0      string  MThd                    audio/unknown
 #>9    byte    >0                      (format %d)
 #>11   byte    >1                      using %d channels
 #                                      Creative Music (CMF) data
-0      string  CTMF                    audio/unknown   
+0      string  CTMF                    audio/unknown
 #                                      SoundBlaster instrument data
-0      string  SBI                     audio/unknown   
+0      string  SBI                     audio/unknown
 #                                      Creative Labs voice data
-0      string  Creative\ Voice\ File   audio/unknown   
+0      string  Creative\ Voice\ File   audio/unknown
 ## is this next line right?  it came this way...
 #>19   byte    0x1A
 #>23   byte    >0                      - version %d
@@ -408,10 +406,7 @@
 0      string          \<!--   text/html
 0      string          \<h1    text/html
 0      string          \<H1    text/html
-0      string          \<!doctype\ HTML        text/html
-0      string          \<!DOCTYPE\ HTML        text/html
-0      string          \<!doctype\ html        text/html
-0      string          \<!doctype\ HTML        text/html
+0      string/c        \<!doctype\ html        text/html
 
 #------------------------------------------------------------------------------
 # images:  file(1) magic for image formats (see also "c-lang" for XPM bitmaps)
@@ -530,7 +525,11 @@
 #
 
 0      string          \376\067\0\043                  application/msword
-0      string          \320\317\021\340\241\261        application/msword
+# disable this one because it applies also to other
+# Office/OLE documents for which msword is not correct. See PR#2608.
+# from magic file of the apache
+#0     string          \320\317\021\340\241\261        application/msword
+512    string          \354\245\301                    application/msword
 0      string          \333\245-\0\0\0                 application/msword
 
 
@@ -566,7 +565,7 @@
 #0     string          \367\203        TeX generic font data
 #0     string          \367\131        TeX packed font data
 #0     string          \367\312        TeX virtual font data
-#0     string          This\ is\ TeX,  TeX transcript text     
+#0     string          This\ is\ TeX,  TeX transcript text
 #0     string          This\ is\ METAFONT,     METAFONT transcript text
 
 # There is no way to detect TeX Font Metric (*.tfm) files without
@@ -603,7 +602,7 @@
 0       belong             0x000001B3     video/mpv
 0       belong&0xFF5FFF1F  0x47400010     video/mp2t
 0       belong             0x00000001
->4      byte&1F            0x07           video/h264
+>4      byte&0x1F         0x07           video/h264
 
 # FLI animation format
 0      leshort         0xAF11                          video/fli
@@ -658,7 +657,7 @@
 #
 # Executables
 #
-0      string          \177ELF 
+0      string          \177ELF
 >16    leshort         0               application/octet-stream
 >16    leshort         1               application/x-object
 >16    leshort         2               application/x-executable
@@ -712,19 +711,19 @@
 4      string  Standard\ Jet\ DB       application/msaccess
 
 ## magic for XBase files
-#0      byte       0x02        
+#0      byte       0x02
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
-#0      byte       0x03        
+#0      byte       0x03
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
-#0      byte       0x04        
+#0      byte       0x04
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
-#0      byte       0x05        
+#0      byte       0x05
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
@@ -740,7 +739,7 @@
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
-#0      byte       0x83        
+#0      byte       0x83
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
@@ -748,7 +747,7 @@
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
-#0      byte       0x8e        
+#0      byte       0x8e
 #>8     leshort          >0
 #>>12   leshort    0   application/x-dbf
 #
@@ -769,7 +768,7 @@
 0       beshort         0x9501                  text/PGP key security ring
 0       beshort         0x9500                  text/PGP key security ring
 0       beshort         0xa600                  text/PGP encrypted data
-0       string          -----BEGIN\040PGP       text/PGP armored data 
+0       string          -----BEGIN\040PGP       text/PGP armored data
 >15     string          PUBLIC\040KEY\040BLOCK- public key block
 >15     string          MESSAGE-                message
 >15     string          SIGNED\040MESSAGE-      signed message
@@ -777,7 +776,7 @@
 0       beshort         0x8501                  data
 #
 # GnuPG Magic:
-# 
+#
 0       beshort         0x9901                  text/GnuPG key public ring
 0       beshort         0x8501                  text/OpenPGP data
 
@@ -787,7 +786,7 @@
 #
 #       http://www.macromedia.com/software/flash/open/
 #
-0      string          FWS             
+0      string          FWS
 >3     byte            x                       application/x-shockwave-flash
 
 # The following paramaters are created for Namazu.
@@ -914,7 +913,7 @@
 #0     string  \000\000\001\000        image/x-ico
 
 # Quark Xpress 3 Files:
-# (made the mimetype up) 
+# (made the mimetype up)
 0      string  \0\0MMXPR3\0    application/x-quark-xpress-3
 
 # EET archive
@@ -924,3 +923,13 @@
 # From: Denis Knauf, via gentoo.
 0      string  fLaC            audio/x-flac
 0      string  CWS             application/x-shockwave-flash
+
+# Hangul Document Files:
+# Reversed-engineered HWP magic numbers
+# From: Won-Kyu Park <wkpark@kldp.org>
+512    string          R\0o\0o\0t\0    application/x-hwp
+
+0      string/c        BEGIN:VCARD     text/x-vcard
+0      string          WordPro\0       application/vnd.lotus-wordpro
+0      string          WordPro\r\373   application/vnd.lotus-wordpro
+0      string          CPC\262         image/cpi