From c28acc38624deae7053621be113c102ca86472a7 Mon Sep 17 00:00:00 2001 From: Oleg Broytman Date: Mon, 17 Mar 2014 20:19:29 +0400 Subject: [PATCH] Update documentation --- mimedecode.docbook | 82 +++++++++++++++++++++++++--------------------- 1 file changed, 44 insertions(+), 38 deletions(-) diff --git a/mimedecode.docbook b/mimedecode.docbook index 02ee7a5..7d767f3 100644 --- a/mimedecode.docbook +++ b/mimedecode.docbook @@ -122,7 +122,7 @@ - Here is the solution - mimedecode.py! + Here is a solution - mimedecode.py! @@ -131,8 +131,8 @@ message, and decodes to stdout or an output file. If the file is not an RFC822 message it is just copied to the output one-to-one. If the file is a simple RFC822 message it is decoded as one part. If it is a MIME message - with multiple parts ("attachments") all parts are decoded. Decoding can be - controlled by command-line options. + with multiple parts ("attachments") all non-multipart subparts are decoded. + Decoding can be controlled by the command-line options. @@ -140,21 +140,22 @@ -r and -R options. Then, Subject and Content-Disposition headers (and all headers listed with -d and -p options) are examined. If any of those exists, they are decoded according to RFC2047. Content-Disposition header is not - decoded - only its "filename" parameter. Encoded header parameters violate - the RFC, but widely deployed anyway by ignorant coders who never even heard - about RFCs. Correct parameter encoding specified by RFC2231. This program - decodes RFC2231-encoded parameters, too. + decoded (if it was not listed in option -d) - only its "filename" parameter. + Encoded header parameters violate the RFC, but widely deployed anyway by + ignorant coders who never even heard about RFCs. Correct parameter encoding + specified by RFC2231. This program decodes RFC2231-encoded parameters, too. Then the body of the message (or the current part) is decoded. Decoding starts with looking at header Content-Transfer-Encoding. If the header specifies non-8bit encoding (usually base64 or quoted-printable), the body - converted to 8bit. Then, if its content type is multipart (multipart/related - or multipart/mixed, e.g) every part is recursively decoded. If it is not - multipart, mailcap database is consulted to find a way to convert the body - to plain text. (I have no idea how mailcap can be configured on OSes other - than POSIX, please don't ask me; real OS users can consult my example at + is converted to 8bit (can be prevented with -B). Then if its content type + is multipart (multipart/related or multipart/mixed, e.g) every part is + recursively decoded. If it is not multipart, mailcap database is consulted + to find a way to convert the body to plain text (can be prevented with + options -Bbei). (The author has no idea how mailcap can be configured on + OSes other than POSIX, please don't ask; users can consult an example at http://phdru.name/Software/dotfiles/mailcap.html). The decoding process uses the first copiousoutput filter it can find. If there are no filters the body just passed as is. @@ -162,14 +163,14 @@ Then Content-Type header is consulted for charset. If it is not equal to the - current locale charset and recoding is allowed the body text is recoded. - Finally message headers and the body are flushed to stdout. + current locale charset and recoding is allowed (see options -Cc) the body + text is recoded. Finally message headers and the body are flushed to stdout. - Please be warned that in the following options asterisk is a shell + Please be reminded that in the following options asterisk is a shell metacharacter and should be escaped or quoted. Either write -d \*,-h1,-h2 or -d '*,-h1,-h2' or such. @@ -221,7 +222,7 @@ -f charset - Force this charset to be the current default charset instead of + Force this charset to be used for recoding instead of charset from the current locale. @@ -254,10 +255,10 @@ This variant completely changes headers decoding. First, the list of - headers to decode is cleared. Then all the headers are decoded - except the given list of exceptions (headers listed with '-'). In - this mode it would be meaningless to give more than one -d options - but the program doesn't enforce it. + headers to decode is cleared (as with -D). Then all the headers are + decoded except the given list of exceptions (headers listed with + '-'). In this mode it would be meaningless to give more than one -d + options but the program doesn't enforce the limitation. @@ -345,7 +346,7 @@ Add the parameters(s) to a list of headers parameters to remove; - the parameters will be decoded only for the given header(s). + the parameters will be removed only for the given header(s). Initially the list is empty. @@ -457,10 +458,11 @@ --save-message mask - Append mask to a list of content types to save to a file; + Append mask to lists of content types to save to files; --save-headers saves only decoded headers of the message (or - subpart); --save-body saves only decoded body; --save-message saves - the entire message or subpart (headers + body). + the current subpart); --save-body saves only decoded body; + --save-message saves the entire message or subpart (headers + + body). @@ -492,14 +494,15 @@ cat input_file | mimedecode.py -o output_file The 5 list options (-Bbeit) require more explanation. They allow a user to control body decoding with great flexibility. Think about said mail archive; for example, its maintainer wants to put there only texts, convert - PDF/Postscript to text, pass HTML and images as is (decoding base64 to html - but left images in base64), and ignore everything else. Easy: + PDF/Postscript to text, pass HTML and images decoding base64 to html but + leaving images encoded, and ignore everything else. This is how it could be + done: - mimedecode.py -t application/pdf -t application/postscript -b text/html - -B 'image/*' -i '*/*' + mimedecode.py -t application/pdf -t application/postscript -t text/plain + -b text/html -B 'image/*' -i '*/*' @@ -507,17 +510,19 @@ cat input_file | mimedecode.py -o output_file When the program decodes a message (non-MIME or a non-multipart subpart of a MIME message), it consults Content-Type header. The content type is searched in all 5 lists, in order "text-binary-ignore-error". If found, appropriate - action performed. If not found, the program search the same lists for + action is performed. If not found, the program searches the same lists for "type/*" mask (the type of "text/html" is just "text"). If found, - appropriate action performed. If not found, the program search the same - lists for "*/*" mask. If found, appropriate action performed. If not found, - the program uses default action, which is to decode everything to text (if - mailcap specifies a filter). + appropriate action is performed. If not found, the program searches the same + lists for "*/*" mask. If found, appropriate action is performed. If not + found, the program uses the default action, which is to decode everything to + text (if mailcap specifies a filter). This algorithm allows more specific + content types to override less specific: -b image/* will be processed + earlier than -B */*. Initially all 5 lists are empty, so without any additional parameters -the program always uses the default decoding. + the program always uses the default decoding (as -t */*). @@ -525,14 +530,15 @@ the program always uses the default decoding. program to save every non-multipart subpart (only headers, or body, or the entire subpart: headers + body) that corresponds to the given mask to a file. Before saving the message (or the subpart) is decoded according to all other - options and placed to the output stream as usual. Filename for the file is + options and is placed to the output stream as usual. Filename for the file is created using "filename" parameter from the Content-Disposition header, or "name" parameter from the Content-Type header if one of those exist; a serial counter is prepended to the filename to avoid collisions; if there are no name/filename parameters, or the name/filename parameters contain forbidden characters (null, slash, backslash) the filename is just the serial counter. The file is saved in the directory set with -O (default is the current - directory). + directory). The save options are processed before option -e so the user can + save the message that causes the error. @@ -560,8 +566,8 @@ the program always uses the default decoding. From/To/Cc/Reply-To/Mail-Followup-To/Subject headers and filenames. Other than that output is correct MIME message. The program does not try to guess whether the headers are correct. For example, if a message header states - that charset is iso8859-5, but the body is actually in utf-8 the program - will recode the message with the wrong charset. + that charset is iso8859-1, but the body (HTML, for example) is actually in + utf-8 the program will recode the message with the wrong charset. -- 2.39.5