The lib email parse problem...

[叮叮当当]

Re: The lib email parse problem...

yes, the special is i must choose exactly one section to destruct,
instead of processing all subparts.


John Machin 写道：

Tim Roberts wrote:

"????" <guotie.9@gmail .comwrote:

>i know how to use email module lib.
>
>the question is about how to handle the rfc 1521 mime
>mulitpart/alternitave part .
>
>i know emai can handle mulitpart , but the subpart alternative is
>special .

No, it's not. A multipart/alternative section is constructed exactly the
same as any other multipart section. It just so happens that it will have
exactly two subsections, one text/plain and one text/html.

>
I was under the impression that it was a little more general than that
... see e.g. http://www.freesoft.org/CIE/RFC/1521/18.htm
>
My guess is that the OP meant special in the sense that the reader
needs to choose one subpart, instead of processing all subparts.
>
Cheers,
John
>
>
>
>

--
- Tim Roberts, timr@probo.com
Providenza & Boekelheide, Inc.

[John Machin]

Re: The lib email parse problem...

On 30/08/2006 4:44 PM, 叮叮当当 wrote:

yes, the special is i must choose exactly one section to destruct,
instead of processing all subparts.

So have you tried to use the example I posted yesterday? Do you still
have any problems? Note: it is generally a good idea to post a message
when you have overcome a problem -- that lets would-be helpers know that
they are "off the case" :-)

Cheers,
John

[neoedmund]

Re: The lib email parse problem...

myself wrote a multipart parser in java(i customise it because i need
get information of "upload progress"). and i think it's also easy to
implement in python, i've not have time done it, or i'll post it.
but if you're no other special needs, just use email lib, it's quick to
program and if you really not need some part, just drop it.
there's anything wrong with email lib?

叮叮当当 wrote:

yes, the special is i must choose exactly one section to destruct,
instead of processing all subparts.
>
>
John Machin 写道：
>

Tim Roberts wrote:

"????" <guotie.9@gmail .comwrote:
>
i know how to use email module lib.

the question is about how to handle the rfc 1521 mime
mulitpart/alternitave part .

i know emai can handle mulitpart , but the subpart alternative is
special .
>
No, it's not. A multipart/alternative section is constructed exactlythe
same as any other multipart section. It just so happens that it willhave
exactly two subsections, one text/plain and one text/html.

I was under the impression that it was a little more general than that
... see e.g. http://www.freesoft.org/CIE/RFC/1521/18.htm

My guess is that the OP meant special in the sense that the reader
needs to choose one subpart, instead of processing all subparts.

Cheers,
John

--
- Tim Roberts, timr@probo.com
Providenza & Boekelheide, Inc.

[叮叮当当]

Re: The lib email parse problem...

thanks.

i have use a temp method to overcome it .

i still think the email lib should give the boundary border to parse
mail.

code is as following:

def parse_mail_cont ent(self, mail):
content = ''
alter = False
subty = ''
html = ''
plain = ''
for part in mail.walk():
if part.is_multipa rt():
if part.get_conten t_subtype() == 'alternative':
alter = True
else:
alter = False
continue
if part.get_conten t_maintype() == 'text':
if part.get_filena me():
continue
ty = part.get_conten t_subtype()
ch = part.get_conten t_charset()
if alter and ty == 'plain':
subty = 'plain'
if ch:
plain = unicode(part.ge t_payload(decod e =
True),ch).encod e('utf-8')
else:
plain = part.get_payloa d(decode =
True).decode('g b2312').encode( 'utf-8')
elif alter and ty == 'html':
subty = 'html'
if ch:
html = unicode(part.ge t_payload(decod e =
True),ch).encod e('utf-8')
else:
html = part.get_payloa d(decode =
True).decode('g b2312').encode( 'utf-8')
elif not alter:
if subty == 'html':
content += html
elif subty == 'plain':
content += plain
alter = False
subty = ''
if ch:
content += unicode(part.ge t_payload(decod e =
True),ch).encod e('utf-8')
else:
content += part.get_payloa d(decode =
True).decode('g b2312').encode( 'utf-8')
elif alter:
if subty == 'html':
content += html
elif subty == 'plain':
content += plain
alter = False
subty = ''
if alter:
if subty == 'html':
content += html
elif subty == 'plain':
content += plain
return content

thanks very much.

John Machin wrote:

On 30/08/2006 4:44 PM, 叮叮当当 wrote:

yes, the special is i must choose exactly one section to destruct,
instead of processing all subparts.

>
So have you tried to use the example I posted yesterday? Do you still
have any problems? Note: it is generally a good idea to post a message
when you have overcome a problem -- that lets would-be helpers know that
they are "off the case" :-)

Cheers,
John

[Fredrik Lundh]

Re: The lib email parse problem...

"????" wrote:

i have use a temp method to overcome it .
>
i still think the email lib should give the boundary border to parse
mail.

the email lib you're using is a PARSER, and it's already PARSING the
mail for you.

(if you have trouble structuring your program when someone else is doing
the parsing for you, what makes you think it would be easier if you had to
do the parsing yourself as well ?)

wrt. your temp method, I think you'll find that a recursive solution would
be a lot easier to get right without having to resort to code duplication like
in your example; I think John Machin posted an example earlier in this
thread.

</F>