replacing \n characters in a hash

[Johhny]

Hello,

I am currently trying to write some scripts to get information from the
xmlrpc for redhat network. One of the issues I am having is trying to
strip off the special characters in the hash that is returned. Here is
an example of the information returned within the hash :

===SNIP===
{'errata_update _date': '2005-12-06', 'errata_topic': 'Updated
libc-client packages that fix a buffer overflow issue are
now\navailable. \n\nThis update has been rated as having moderate
security impact by the Red\nHat Security Response Team.',
'errata_type': 'Security Advisory', 'errata_notes': '',
'errata_synopsi s': 'Moderate: libc-client security update',
'errata_referen ces': '', 'errata_last_mo dified_date': '2006-01-25
10:37:24', 'errata_issue_d ate': '2005-12-06', 'errata_descrip tion':
'C-client is a common API for accessing mailboxes.\n\nA buffer overflow
flaw was discovered in the way C-client parses user\nsupplied
mailboxes. If an authenticated user requests a specially
crafted\nmailbo x name, it may be possible to execute arbitrary code on
a server that\nuses C-client to access mailboxes. The Common
Vulnerabilities and Exposures\nproj ect has assigned the name
CVE-2005-2933 to this issue.\n\nAll users of libc-client should upgrade
to these updated packages, which\ncontain a backported patch that
resolves this issue.'}
===SNIP===


What I would like to do is remove the \n characters from
'errata_topic'. Which is this section of the hash.

Updated libc-client packages that fix a buffer overflow issue are
now\navailable. \n\nThis update has been rated as having moderate
security impact by the Red\nHat Security Response Team.

What I had attempted to do is use the replace() function but it
consistantly comes up with the following errors:

Traceback (most recent call last):
File "rhn_errata.py" , line 63, in ?
errata_package = errata_package. strip('\n','')
AttributeError: 'dict' object has no attribute 'strip'

where errata_package is JUST the errata_topic hash value.

Any advice would be great on how to do that.

Regards,

Johhny

[Carsten Haese]

Re: replacing \n characters in a hash

On Thu, 2006-01-26 at 09:24, Johhny wrote:[color=blue]
> Hello,
>
> I am currently trying to write some scripts to get information from the
> xmlrpc for redhat network. One of the issues I am having is trying to
> strip off the special characters in the hash that is returned. Here is
> an example of the information returned within the hash :
>
> ===SNIP===
> {'errata_update _date': '2005-12-06', 'errata_topic': 'Updated
> libc-client packages that fix a buffer overflow issue are
> now\navailable. \n\nThis update has been rated as having moderate
> security impact by the Red\nHat Security Response Team.',
>
> [...]
>
> What I had attempted to do is use the replace() function but it
> consistantly comes up with the following errors:
>
> Traceback (most recent call last):
> File "rhn_errata.py" , line 63, in ?
> errata_package = errata_package. strip('\n','')
> AttributeError: 'dict' object has no attribute 'strip'
>
> where errata_package is JUST the errata_topic hash value.[/color]

errata_package is obviously not just the errata_topic hash value,
because that would be a string, and python for some reason seems to
believe that errata_package is a dictionary.

Also note that once you correct that problem, python will complain that
strip() doesn't take 2 parameters, since you actually mean replace().

-Carsten

[Johhny]

Re: replacing \n characters in a hash

Hello,

Thankyou for your response,
If I check that the errara_package value is with a print I get the
following.

===SNIP===
Updated libc-client packages that fix a buffer overflow issue are now
available.

This update has been rated as having moderate security impact by the
Red
Hat Security Response Team.
===SNIP===

Notice that it has formatted the output with the \n's.

So i dont understand why its reporting as a dictionary rather than just
the string.

[Carsten Haese]

Re: replacing \n characters in a hash

On Thu, 2006-01-26 at 09:49, Johhny wrote:[color=blue]
> Hello,
>
> Thankyou for your response,
> If I check that the errara_package value is with a print I get the
> following.
>
> ===SNIP===
> Updated libc-client packages that fix a buffer overflow issue are now
> available.
>
> This update has been rated as having moderate security impact by the
> Red
> Hat Security Response Team.
> ===SNIP===
>
> Notice that it has formatted the output with the \n's.
>
> So i dont understand why its reporting as a dictionary rather than just
> the string.[/color]

Posting relevant bits of your code might help.

-Carsten

[Johhny]

Re: replacing \n characters in a hash

Hello,

Here is the code (minus my details section).

server = xmlrpclib.Serve rProxy(url)

session = server.auth.log in(username,pas sword)

#functions.

def getErrata():

channel_label = 'rhel-i386-as-4'

errata =
server.channel. software.list_e rrata(session,c hannel_label,st art_date,end_da te)

return errata



def getPackage(advi sory):

Package = server.errata.g et_details(sess ion,advisory)

return Package



errata = getErrata()

for vals in errata:

print "%s\t\t%s\t\t%s \t%s\t%s" %
(vals['errata_advisor y'],vals['errata_issue_d ate'],vals['errata_update_ date'],vals['errata_last_mo dified_date'],vals['errata_type'],)

errata_info = getPackage(vals['errata_advisor y'],)

print errata_info['errata_topic']

errata_package = errata_info['errata_topic']

print getPackage(vals['errata_advisor y'])

I have not got any of the section in to replace the \n's as I was
trying to work out why its not seeing what I thought was a string as a
dict.

[Larry Bates]

Re: replacing \n characters in a hash

Johhny wrote:[color=blue]
> Hello,
>
> I am currently trying to write some scripts to get information from the
> xmlrpc for redhat network. One of the issues I am having is trying to
> strip off the special characters in the hash that is returned. Here is
> an example of the information returned within the hash :
>
> ===SNIP===
> {'errata_update _date': '2005-12-06', 'errata_topic': 'Updated
> libc-client packages that fix a buffer overflow issue are
> now\navailable. \n\nThis update has been rated as having moderate
> security impact by the Red\nHat Security Response Team.',
> 'errata_type': 'Security Advisory', 'errata_notes': '',
> 'errata_synopsi s': 'Moderate: libc-client security update',
> 'errata_referen ces': '', 'errata_last_mo dified_date': '2006-01-25
> 10:37:24', 'errata_issue_d ate': '2005-12-06', 'errata_descrip tion':
> 'C-client is a common API for accessing mailboxes.\n\nA buffer overflow
> flaw was discovered in the way C-client parses user\nsupplied
> mailboxes. If an authenticated user requests a specially
> crafted\nmailbo x name, it may be possible to execute arbitrary code on
> a server that\nuses C-client to access mailboxes. The Common
> Vulnerabilities and Exposures\nproj ect has assigned the name
> CVE-2005-2933 to this issue.\n\nAll users of libc-client should upgrade
> to these updated packages, which\ncontain a backported patch that
> resolves this issue.'}
> ===SNIP===
>
>
> What I would like to do is remove the \n characters from
> 'errata_topic'. Which is this section of the hash.
>
> Updated libc-client packages that fix a buffer overflow issue are
> now\navailable. \n\nThis update has been rated as having moderate
> security impact by the Red\nHat Security Response Team.
>
> What I had attempted to do is use the replace() function but it
> consistantly comes up with the following errors:
>
> Traceback (most recent call last):
> File "rhn_errata.py" , line 63, in ?
> errata_package = errata_package. strip('\n','')
> AttributeError: 'dict' object has no attribute 'strip'
>
> where errata_package is JUST the errata_topic hash value.
>
> Any advice would be great on how to do that.
>
> Regards,
>
> Johhny
>[/color]
You must use 'errata_topic' as index into your dictionary
to get the string and then replace the \n chars.

Example assumes that the dict is pointed to by d):

errata_topic_te xt=d['errata_topic'].replace('\n',' ')

Hope this helps.

Larry Bates

[Fredrik Lundh]

Re: replacing \n characters in a hash

Johhny wrote:

[color=blue]
> for vals in errata:
>
> print "%s\t\t%s\t\t%s \t%s\t%s" %
>[/color]
(vals['errata_advisor y'],vals['errata_issue_d ate'],vals['errata_update_ date'],vals['errata_last_mo dified_date'],vals['errata_type'],
)[color=blue]
>
> errata_info = getPackage(vals['errata_advisor y'],)
>
> print errata_info['errata_topic']
>
> errata_package = errata_info['errata_topic'][/color]

add

print type(errata_pac kage), repr(errata_pac kage)

here, and let us know what it prints.
[color=blue]
> print getPackage(vals['errata_advisor y'])[/color]

</F>

[Steve Holden]

Re: replacing \n characters in a hash

Johhny wrote:[color=blue]
> Hello,
>
> Here is the code (minus my details section).
>
> server = xmlrpclib.Serve rProxy(url)
>
> session = server.auth.log in(username,pas sword)
>
> #functions.
>
> def getErrata():
>
> channel_label = 'rhel-i386-as-4'
>
> errata =
> server.channel. software.list_e rrata(session,c hannel_label,st art_date,end_da te)
>
> return errata
>
>
>
> def getPackage(advi sory):
>
> Package = server.errata.g et_details(sess ion,advisory)
>
> return Package
>
>
>
> errata = getErrata()
>
> for vals in errata:
>
> print "%s\t\t%s\t\t%s \t%s\t%s" %
> (vals['errata_advisor y'],vals['errata_issue_d ate'],vals['errata_update_ date'],vals['errata_last_mo dified_date'],vals['errata_type'],)
>
> errata_info = getPackage(vals['errata_advisor y'],)
>
> print errata_info['errata_topic']
>
> errata_package = errata_info['errata_topic'][/color]

errata_package = errata_info['errata_topic'].replace('\n', " ")
[color=blue]
>
> print getPackage(vals['errata_advisor y'])
>
> I have not got any of the section in to replace the \n's as I was
> trying to work out why its not seeing what I thought was a string as a
> dict.
>[/color]

regards
Steve
--
Steve Holden +44 150 684 7255 +1 800 494 3119
Holden Web LLC www.holdenweb.com
PyCon TX 2006 www.python.org/pycon/

[Johhny]

Re: replacing \n characters in a hash

Hello,

In response to that the output is this :

<type 'str'> 'Updated libc-client packages that fix a buffer overflow
issue are now\navailable. \n\nThis update has been rated as having
moderate security impact by the Red\nHat Security Response Team.'

[Johhny]

Re: replacing \n characters in a hash

Hello All,

thanks for your help. I got it working and learnt some more things
which is always great. Your assitance has been very useful.

Regards,

Johhny

[Fredrik Lundh]

Re: replacing \n characters in a hash

(since this is a newsgroup, can you please quote the message
you're replying to).

Johhny wrote:
[color=blue]
> In response to that the output is this :
>
> <type 'str'> 'Updated libc-client packages that fix a buffer overflow
> issue are now\navailable. \n\nThis update has been rated as having
> moderate security impact by the Red\nHat Security Response Team.'[/color]

<type 'str> means that it *is* a string, after all. looks like you didn't
post the code you were using :-(

replacing the print statement with

errata_package = errata_package. replace("\n", " ")

should do what you want.

</F>