* <many> add support for searching attachment mime-types

- updated manpages
  - some cleanups

  Note, requires a --rebuild

NEWS

@@ -3,6 +3,8 @@
 ** Release 0.9.8 <>
 
    - '--descending' has  been renamed into '--reverse'
+   - search for attachment MIME-type using 'attmime:' or 'y:'
+   - experimental emacs-based mail client
 
  * Release 0.9.7 <>
 

man/mu-easy.1

@@ -1,6 +1,6 @@
-.TH MU-EASY 1 "August 2011" "User Manuals"
+.TH MU-EASY 1 "November 2011" "User Manuals"
 
-.SH NAME 
+.SH NAME
 
 mu easy \- a quick introduction to mu
 
@@ -24,7 +24,7 @@ you won't be able to index/search your mail.
 Before you can search e-mails, you'll first need to index them:
 
 .nf
-  \fB$ mu index\fR 
+  \fB$ mu index\fR
 .fi
 
 The process can take a few minutes, depending on the amount of mail you have,
@@ -54,7 +54,7 @@ some examples for common cases.
 First, let's search for all messages sent to Julius Caesar regarding fruit:
 
 .nf
-\fB$ mu find t:julius fruit\fR 
+\fB$ mu find t:julius fruit\fR
 .fi
 
 This should return something like:
@@ -73,7 +73,7 @@ date/sender/subject. However, we can change this using the \fI--fields\fR
 parameter (see the \fBmu-find\fR man page for the details):
 
 .nf
-  \fB$ mu find --fields="t s" t:julius fruit\fR 
+  \fB$ mu find --fields="t s" t:julius fruit\fR
 .fi
 
 In other words, display the 'To:'-field (t) and the subject (s). This should
@@ -90,7 +90,7 @@ is, it displays messages that match all the parameters. However, we can use
 logical OR as well:
 
 .nf
-  \fB$ mu find t:julius OR f:socrates\fR 
+  \fB$ mu find t:julius OR f:socrates\fR
 .fi
 
 In other words, display messages that are either sent to Julius Caesar
@@ -106,7 +106,7 @@ a 'summary' of the first lines of the message using the \fI--summary\fR
 option, which will 'summarize' the first \fIn\fR lines of the message:
 
 .nf
-  \fB$ mu find --summary napoleon m:/archive\fR 
+  \fB$ mu find --summary napoleon m:/archive\fR
 .fi
 
 .nf
@@ -130,21 +130,21 @@ version 0.9 or later.
 
 Get all important messages which are signed:
 .nf
-  \fB$ mu find flag:signed prio:high \fR 
+  \fB$ mu find flag:signed prio:high \fR
 .fi
 
 Get all messages from Jim without an attachment:
 .nf
-  \fB$ mu find from:jim AND NOT flag:attach\fR 
+  \fB$ mu find from:jim AND NOT flag:attach\fR
 .fi
 
 Get all unread messages where the subject mentions Ångström:
 .nf
-  \fB$ mu find subject:Ångström flag:unread\fR 
+  \fB$ mu find subject:Ångström flag:unread\fR
 .fi
 which is equivalent to:
 .nf
-  \fB$ mu find subject:angstrom flag:unread\fR 
+  \fB$ mu find subject:angstrom flag:unread\fR
 .fi
 because does mu is case-insensitive and accent-insensitive.
 
@@ -159,7 +159,7 @@ Get all messages received today:
   \fB$ mu find date:today..now\fR
 .fi
 
-Get all message we got in the last two weeks about emacs:
+Get all messages we got in the last two weeks about emacs:
 .nf
   \fB$ mu find date:2w..now emacs\fR
 .fi
@@ -182,6 +182,25 @@ filename, for example:
 .fi
 will get you all message with an attachment starting with 'pic'.
 
+If you want to find attachments with a certain MIME-type, you can use the
+following:
+
+Get all messages with PDF attachments:
+.nf
+  \fB$ mu find attmime:application/pdf\fR
+.fi
+
+or even:
+
+Get all messages with image attachments:
+.nf
+  \fB$ mu find 'attmime:image/*'\fR
+.fi
+
+Note that (1) the '*' wildcard can only be used as the rightmost thing in a
+search query, and (2) that you need to quote the search term, because
+otherwise your shell will interpret the '*' (expanding it to all files in the
+current directory -- probably not what you want).
 
 .SH DISPLAYING MESSAGES
 
@@ -195,14 +214,14 @@ need its path. To get the path (think \fBl\fRocation) for our first example we
 can use:
 
 .nf
-  \fB$ mu find --fields="l" t:julius fruit\fR 
+  \fB$ mu find --fields="l" t:julius fruit\fR
 .fi
 
 And we'll get someting like:
 .nf
   /home/someuser/Maildir/archive/cur/1266188485_0.6850.cthulhu:2,
 .fi
-We can now display this message: 
+We can now display this message:
 
 .nf
   \fB$ mu view /home/someuser/Maildir/archive/cur/1266188485_0.6850.cthulhu:2,\fR
@@ -225,7 +244,7 @@ find \fIcontacts\fR, that is, names + addresses. Without any search
 expression, \fBmu cfind\fR lists all of your contacts.
 
 .nf
-  \fB$ mu cfind julius\fR 
+  \fB$ mu cfind julius\fR
 .fi
 
 will find all contacts with 'julius' in either name or e-mail address. Note
@@ -237,7 +256,7 @@ program. For example, to export your contact information to a \fBmutt\fR
 address book file, you can use something like:
 
 .nf
-  \fB$ mu cfind --format=mutt-alias > ~/mutt-aliases \fR 
+  \fB$ mu cfind --format=mutt-alias > ~/mutt-aliases \fR
 .fi
 
 Then, you can use them in \fBmutt\fR if you add something like \fBsource

man/mu-find.1

@@ -1,4 +1,4 @@
-.TH MU FIND 1 "July 2011" "User Manuals"
+.TH MU FIND 1 "November 2011" "User Manuals"
 
 .SH NAME
 
@@ -123,6 +123,7 @@ search fields and their abbreviations:
 	date,d          Date-Range
 	size,z          Message size
 	attach,a        Attachment filename
+	attmime,y       Attachment MIME-type
 	tag,x           Tag for the message (contents of the \fIX-Label\fR field)
 .fi
 
@@ -474,6 +475,23 @@ Find all unread messages with attachments:
 .fi
 
 
+Find all messages with PDF-attachments:
+
+.nf
+  $ mu find attmime:application/pdf
+.fi
+
+Find all messages with attached images:
+
+.nf
+  $ mu find 'attmime:image/*'
+.fi
+
+Note[1]: the argument needs to be quoted, or the shell will interpret the '*'
+Note[2]: the '*' wild card can only be used as the last (rightmost) part of a
+search term.
+
+
 .SS Integrating mu find with mail clients
 
 .TP

src/mu-msg-fields.c

@@ -34,24 +34,31 @@ enum _FieldFlags {
 	FLAG_GMIME	         = 1 << 0, /* field retrieved through
 					    * gmime */
 	FLAG_XAPIAN_INDEX        = 1 << 1, /* field is indexed in
-			          	    * xapian */
+			          	    * xapian (i.e., the text
+			          	    * is processed */
 	FLAG_XAPIAN_TERM         = 1 << 2, /* field stored as term in
-			          	    * xapian */
+			          	    * xapian (so it can be searched) */
 	FLAG_XAPIAN_VALUE        = 1 << 3, /* field stored as value in
-			          	    * xapian */
+			          	    * xapian (so the literal
+			          	    * value can be
+			          	    * retrieved) */
 	FLAG_XAPIAN_CONTACT      = 1 << 4, /* field contains one or more
 					    * e-mail-addresses */
 	FLAG_XAPIAN_ESCAPE       = 1 << 5, /* field needs escaping for
-					    * xapian */
+					    * xapian (so the xapian
+					    * query does not get
+					    * confused) */
 	FLAG_XAPIAN_BOOLEAN      = 1 << 6, /* use 'add_boolean_prefix'
 					    * for Xapian queries */
 	FLAG_XAPIAN_PREFIX_ONLY  = 1 << 7, /* whether this fields
 					    * matches only when the
 					    * prefix is explicitly
-					    * included */
+					    * included in the search
+					    * query -- e.g., the text
+					    * body */
 	FLAG_NORMALIZE	         = 1 << 8, /* field needs flattening for
 					    * case/accents */
-	FLAG_DONT_CACHE          = 1 << 9  /* don't cache this field in
+	FLAG_DONT_CACHE          = 1 << 9, /* don't cache this field in
 					    * the MuMsg cache */
 };
 typedef enum _FieldFlags	FieldFlags;
@@ -81,8 +88,14 @@ static const MuMsgField FIELD_DATA[] = {
 		MU_MSG_FIELD_ID_ATTACH,
 		MU_MSG_FIELD_TYPE_STRING,
 		"attach" , 'a', 'A',
-		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_NORMALIZE |
+		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_NORMALIZE | FLAG_DONT_CACHE
-		FLAG_DONT_CACHE
+	},
+
+	{
+		MU_MSG_FIELD_ID_ATTACH_MIME_TYPE,
+		MU_MSG_FIELD_TYPE_STRING,
+		"attmime" , 'y', 'Y',
+		FLAG_XAPIAN_TERM | FLAG_XAPIAN_ESCAPE
 	},
 
 	{
@@ -213,6 +226,13 @@ static const MuMsgField FIELD_DATA[] = {
 		"tag", 'x', 'X',
 		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_XAPIAN_PREFIX_ONLY |
 		FLAG_NORMALIZE | FLAG_XAPIAN_ESCAPE
+	},
+
+	{	/* special, internal field, to get a unique key */
+		MU_MSG_FIELD_ID_UID,
+		MU_MSG_FIELD_TYPE_STRING,
+		"uid", 0, 'U',
+		FLAG_XAPIAN_TERM
 	}
 
 	/* note, mu-store also use the 'Q' internal prefix for its uids */

src/mu-msg-fields.h

@@ -29,7 +29,9 @@ G_BEGIN_DECLS
 enum _MuMsgFieldId {
 
 	/* first all the string-based ones */
-	MU_MSG_FIELD_ID_ATTACH	    = 0,
+	MU_MSG_FIELD_ID_ATTACH	            = 0,
+	MU_MSG_FIELD_ID_ATTACH_MIME_TYPE, /* mime-type */
+
 	MU_MSG_FIELD_ID_BCC,
 	MU_MSG_FIELD_ID_BODY_HTML,
 	MU_MSG_FIELD_ID_BODY_TEXT,
@@ -40,6 +42,9 @@ enum _MuMsgFieldId {
 	MU_MSG_FIELD_ID_PATH,
 	MU_MSG_FIELD_ID_SUBJECT,
 	MU_MSG_FIELD_ID_TO,
+
+	MU_MSG_FIELD_ID_UID, /* special, generated from path */
+
 	/* MU_MSG_STRING_FIELD_ID_NUM, see below */
 
 	/* string list items... */
@@ -59,7 +64,7 @@ typedef guint8 MuMsgFieldId;
 
 /* some specials... */
 static const MuMsgFieldId MU_MSG_FIELD_ID_NONE = (MuMsgFieldId)-1;
-#define MU_MSG_STRING_FIELD_ID_NUM (MU_MSG_FIELD_ID_TO + 1)
+#define MU_MSG_STRING_FIELD_ID_NUM (MU_MSG_FIELD_ID_UID + 1)
 
 #define mu_msg_field_id_is_valid(MFID) \
 	((MFID) < MU_MSG_FIELD_ID_NUM)

src/mu-msg-file.c

@@ -594,7 +594,7 @@ get_body (MuMsgFile *self, gboolean want_html)
 	/* note, str may be NULL (no body), but that's not necessarily
 	 * an error; we only warn when an actual error occured */
 	if (err)
-		g_warning ("error occured while retrieving %s body"
+		g_warning ("error occured while retrieving %s body "
 			   "for message %s",
 			   want_html ? "html" : "text", self->_path);
 	return str;

src/mu-store-priv.hh

@@ -225,9 +225,4 @@ private:
 };
 
 
-/* Xapian DB prefix for the UID value */
-#define MU_STORE_UID_PREFIX "Q"
-
-
-
 #endif /*__MU_STORE_PRIV_HH__*/

src/mu-store-read.cc

@@ -50,6 +50,8 @@ _MuStore::get_uid_term (const char* path)
 	unsigned djbhash, bkdrhash, bkdrseed;
 	unsigned u;
 	static char hex[18];
+	static const char uid_prefix =
+		mu_msg_field_xapian_prefix(MU_MSG_FIELD_ID_UID);
 
 	djbhash  = 5381;
 	bkdrhash = 0;
@@ -60,9 +62,8 @@ _MuStore::get_uid_term (const char* path)
 		bkdrhash = bkdrhash * bkdrseed + path[u];
 	}
 
-	snprintf (hex, sizeof(hex),
+	snprintf (hex, sizeof(hex), "%c%08x%08x",
-		  MU_STORE_UID_PREFIX "%08x%08x",
+		  uid_prefix, djbhash, bkdrhash);
-		  djbhash, bkdrhash);
 
 	return hex;
 }

src/mu-store-write.cc

@@ -405,19 +405,32 @@ struct PartData {
 static void
 each_part (MuMsg *msg, MuMsgPart *part, PartData *pdata)
 {
+	static const std::string
+		att (prefix(MU_MSG_FIELD_ID_ATTACH)),
+		mime (prefix(MU_MSG_FIELD_ID_ATTACH_MIME_TYPE));
+
 	if (mu_msg_part_looks_like_attachment (part, TRUE) &&
 	    (part->file_name)) {
 
 		char val[MuStore::MAX_TERM_LENGTH + 1];
 		strncpy (val, part->file_name, sizeof(val));
 
-		/* now, let's create a terms... */
+		/* now, let's create a term... */
 		mu_str_normalize_in_place (val, TRUE);
 		mu_str_ascii_xapian_escape_in_place (val);
 
 		pdata->_doc.add_term
-			(prefix(pdata->_mfid) +
+			(att + std::string(val, 0, MuStore::MAX_TERM_LENGTH));
-			 std::string(val, 0, MuStore::MAX_TERM_LENGTH));
+
+		/* save the mime type */
+		if (part->type) {
+			gchar *str;
+			str = g_strdup_printf ("%s/%s", part->type, part->subtype);
+			pdata->_doc.add_term
+				(mime + std::string(str, 0, MuStore::MAX_TERM_LENGTH));
+			g_free (str);
+		} else
+			pdata->_doc.add_term (mime + "application/octet-stream");
 	}
 }
 
@@ -482,9 +495,12 @@ add_terms_values (MuMsgFieldId mfid, MsgDoc* msgdoc)
 	case MU_MSG_FIELD_ID_BODY_TEXT:
 		add_terms_values_body (*msgdoc->_doc, msgdoc->_msg, mfid);
 		break;
-	case MU_MSG_FIELD_ID_ATTACH:
+	case MU_MSG_FIELD_ID_ATTACH: /* also takes care of MU_MSG_FIELD_ID_ATTACH_MIME */
 		add_terms_values_attach (*msgdoc->_doc, msgdoc->_msg, mfid);
 		break;
+	case MU_MSG_FIELD_ID_ATTACH_MIME_TYPE:
+	case MU_MSG_FIELD_ID_UID:
+		break; /* already taken care of elsewhere */
 	default:
 		if (mu_msg_field_is_numeric (mfid))
 			add_terms_values_number (*msgdoc->_doc, msgdoc->_msg,