Missed version

Change to example.com/40 for iniplist with leftover bits to ignore.

.gitignore

@@ -0,0 +1,8 @@
+*.pyc
+build/
+test/*.out
+test/*.tstout
+test/*.log
+test.db
+dist
+MANIFEST

CREDITS

@@ -1,5 +1,5 @@
 Jim Niemira (urmane@urmane.org) wrote the original C module and some quick
-and dirty python to use it.  Stuart D. Gathman (stuart@bmsi.com) took that
+and dirty python to use it.  Stuart D. Gathman (stuart@gathman.org) took that
 kludge and added threading and context objects to it, wrote a proper OO
 wrapper (Milter.py) that handles attachments, did lots of testing, packaged
 it with distutils, and generally transformed it from a quick hack to a
@@ -7,7 +7,6 @@ real, usable Python extension.
 
 Other contributors (in random order):
 
-
 Daniel Troeder
   for pointing out a typo in @noreply
 arkanes@irc.freenode.net
@@ -44,4 +43,4 @@ Business Management Systems - http://www.bmsi.com
   for hosting the website, and providing paying clients who need milter service
   so I can work on it as part of my day job.
 
-If I have left anybody out, send me a reminder: stuart@bmsi.com
+If I have left anybody out, send me a reminder: stuart@gathman.org

Doxyfile

@@ -1,4 +1,4 @@
-# Doxyfile 1.5.7.1
+# Doxyfile 1.6.1
 
 # This file describes the settings to be used by the documentation system
 # doxygen (www.doxygen.org) for a project
@@ -31,7 +31,7 @@ PROJECT_NAME           = pymilter
 # This could be handy for archiving the generated documentation or
 # if some version control system is used.
 
-PROJECT_NUMBER         = 0.9.6
+PROJECT_NUMBER         = 1.0
 
 # The OUTPUT_DIRECTORY tag is used to specify the (relative or absolute)
 # base path where the generated documentation will be put.
@@ -54,11 +54,11 @@ CREATE_SUBDIRS         = NO
 # information to generate all constant output in the proper language.
 # The default language is English, other supported languages are:
 # Afrikaans, Arabic, Brazilian, Catalan, Chinese, Chinese-Traditional,
-# Croatian, Czech, Danish, Dutch, Farsi, Finnish, French, German, Greek, 
+# Croatian, Czech, Danish, Dutch, Esperanto, Farsi, Finnish, French, German,
-# Hungarian, Italian, Japanese, Japanese-en (Japanese with English messages), 
+# Greek, Hungarian, Italian, Japanese, Japanese-en (Japanese with English
-# Korean, Korean-en, Lithuanian, Norwegian, Macedonian, Persian, Polish, 
+# messages), Korean, Korean-en, Lithuanian, Norwegian, Macedonian, Persian,
-# Portuguese, Romanian, Russian, Serbian, Serbian-Cyrilic, Slovak, Slovene, 
+# Polish, Portuguese, Romanian, Russian, Serbian, Serbian-Cyrilic, Slovak,
-# Spanish, Swedish, and Ukrainian.
+# Slovene, Spanish, Swedish, Ukrainian, and Vietnamese.
 
 OUTPUT_LANGUAGE        = English
 
@@ -207,6 +207,17 @@ OPTIMIZE_FOR_FORTRAN   = NO
 
 OPTIMIZE_OUTPUT_VHDL   = NO
 
+# Doxygen selects the parser to use depending on the extension of the files it parses.
+# With this tag you can assign which parser to use for a given extension.
+# Doxygen has a built-in mapping, but you can override or extend it using this tag.
+# The format is ext=language, where ext is a file extension, and language is one of
+# the parsers supported by doxygen: IDL, Java, Javascript, C#, C, C++, D, PHP,
+# Objective-C, Python, Fortran, VHDL, C, C++. For instance to make doxygen treat
+# .inc files as Fortran files (default is PHP), and .f files as C (default is Fortran),
+# use: inc=Fortran f=C. Note that for custom extensions you also need to set FILE_PATTERNS otherwise the files are not read by doxygen.
+
+EXTENSION_MAPPING      =
+
 # If you use STL classes (i.e. std::string, std::vector, etc.) but do not want
 # to include (a tag file for) the STL sources as input, then you should
 # set this tag to YES in order to let doxygen match functions declarations and
@@ -394,6 +405,10 @@ SORT_MEMBER_DOCS       = YES
 
 SORT_BRIEF_DOCS        = NO
 
+# If the SORT_MEMBERS_CTORS_1ST tag is set to YES then doxygen will sort the (brief and detailed) documentation of class members so that constructors and destructors are listed first. If set to NO (the default) the constructors will appear in the respective orders defined by SORT_MEMBER_DOCS and SORT_BRIEF_DOCS. This tag will be ignored for brief docs if SORT_BRIEF_DOCS is set to NO and ignored for detailed docs if SORT_MEMBER_DOCS is set to NO.
+
+SORT_MEMBERS_CTORS_1ST = NO
+
 # If the SORT_GROUP_NAMES tag is set to YES then doxygen will sort the
 # hierarchy of group names into alphabetical order. If set to NO (the default)
 # the group names will appear in their defined order.
@@ -468,7 +483,8 @@ SHOW_DIRECTORIES       = YES
 SHOW_FILES             = YES
 
 # Set the SHOW_NAMESPACES tag to NO to disable the generation of the
-# Namespaces page.  This will remove the Namespaces entry from the Quick Index
+# Namespaces page.
+# This will remove the Namespaces entry from the Quick Index
 # and from the Folder Tree View (if specified). The default is YES.
 
 SHOW_NAMESPACES        = YES
@@ -511,7 +527,7 @@ WARNINGS               = YES
 # for undocumented members. If EXTRACT_ALL is set to YES then this flag will
 # automatically be disabled.
 
-WARN_IF_UNDOCUMENTED   = YES
+WARN_IF_UNDOCUMENTED   = NO
 
 # If WARN_IF_DOC_ERROR is set to YES, doxygen will generate warnings for
 # potential errors in the documentation, such as not documenting some
@@ -552,7 +568,10 @@ WARN_LOGFILE           =
 # directories like "/usr/src/myproject". Separate the files or directories
 # with spaces.
 
-INPUT                  = mime.py doc/mainpage.py doc/milter.py Milter
+INPUT                  = mime.py \
+                         doc/mainpage.py \
+                         doc/milter.py \
+                         Milter
 
 # This tag can be used to specify the character encoding of the source files
 # that doxygen parses. Internally doxygen uses the UTF-8 encoding, which is
@@ -636,14 +655,17 @@ IMAGE_PATH             =
 # by executing (via popen()) the command <filter> <input-file>, where <filter>
 # is the value of the INPUT_FILTER tag, and <input-file> is the name of an
 # input file. Doxygen will then use the output that the filter program writes
-# to standard output.  If FILTER_PATTERNS is specified, this tag will be 
+# to standard output.
+# If FILTER_PATTERNS is specified, this tag will be
 # ignored.
 
 INPUT_FILTER           =
 
 # The FILTER_PATTERNS tag can be used to specify filters on a per file pattern
-# basis.  Doxygen will compare the file name with each pattern and apply the 
+# basis.
-# filter if there is a match.  The filters are a list of the form: 
+# Doxygen will compare the file name with each pattern and apply the
+# filter if there is a match.
+# The filters are a list of the form:
 # pattern=filter (like *.cpp=my_cpp_filter). See INPUT_FILTER for further
 # info on how filters are used. If FILTER_PATTERNS is empty, INPUT_FILTER
 # is applied to all files.
@@ -693,7 +715,8 @@ REFERENCES_RELATION    = YES
 # If the REFERENCES_LINK_SOURCE tag is set to YES (the default)
 # and SOURCE_BROWSER tag is set to YES, then the hyperlinks from
 # functions in REFERENCES_RELATION and REFERENCED_BY_RELATION lists will
-# link to the source code.  Otherwise they will link to the documentstion.
+# link to the source code.
+# Otherwise they will link to the documentation.
 
 REFERENCES_LINK_SOURCE = YES
 
@@ -767,6 +790,11 @@ HTML_HEADER            =
 
 HTML_FOOTER            =
 
+# If the HTML_TIMESTAMP tag is set to YES then the generated HTML
+# documentation will contain the timesstamp.
+
+HTML_TIMESTAMP         = NO
+
 # The HTML_STYLESHEET tag can be used to specify a user-defined cascading
 # style sheet that is used by each HTML page. It can be used to
 # fine-tune the look of the HTML output. If the tag is left blank doxygen
@@ -875,20 +903,37 @@ QCH_FILE               =
 
 # The QHP_NAMESPACE tag specifies the namespace to use when generating
 # Qt Help Project output. For more information please see
-# <a href="http://doc.trolltech.com/qthelpproject.html#namespace">Qt Help Project / Namespace</a>.
+# http://doc.trolltech.com/qthelpproject.html#namespace
 
 QHP_NAMESPACE          = org.doxygen.Project
 
 # The QHP_VIRTUAL_FOLDER tag specifies the namespace to use when generating
 # Qt Help Project output. For more information please see
-# <a href="http://doc.trolltech.com/qthelpproject.html#virtual-folders">Qt Help Project / Virtual Folders</a>.
+# http://doc.trolltech.com/qthelpproject.html#virtual-folders
 
 QHP_VIRTUAL_FOLDER     = doc
 
+# If QHP_CUST_FILTER_NAME is set, it specifies the name of a custom filter to add.
+# For more information please see
+# http://doc.trolltech.com/qthelpproject.html#custom-filters
+
+QHP_CUST_FILTER_NAME   =
+
+# The QHP_CUST_FILT_ATTRS tag specifies the list of the attributes of the custom filter to add.For more information please see
+# <a href="http://doc.trolltech.com/qthelpproject.html#custom-filters">Qt Help Project / Custom Filters</a>.
+
+QHP_CUST_FILTER_ATTRS  =
+
+# The QHP_SECT_FILTER_ATTRS tag specifies the list of the attributes this project's
+# filter section matches.
+# <a href="http://doc.trolltech.com/qthelpproject.html#filter-attributes">Qt Help Project / Filter Attributes</a>.
+
+QHP_SECT_FILTER_ATTRS  =
+
 # If the GENERATE_QHP tag is set to YES, the QHG_LOCATION tag can
 # be used to specify the location of Qt's qhelpgenerator.
 # If non-empty doxygen will try to run qhelpgenerator on the generated
-# .qhp file .
+# .qhp file.
 
 QHG_LOCATION           =
 
@@ -905,21 +950,19 @@ ENUM_VALUES_PER_LINE   = 4
 
 # The GENERATE_TREEVIEW tag is used to specify whether a tree-like index
 # structure should be generated to display hierarchical information.
-# If the tag value is set to FRAME, a side panel will be generated
+# If the tag value is set to YES, a side panel will be generated
 # containing a tree-like index structure (just like the one that
 # is generated for HTML Help). For this to work a browser that supports
-# JavaScript, DHTML, CSS and frames is required (for instance Mozilla 1.0+, 
+# JavaScript, DHTML, CSS and frames is required (i.e. any modern browser).
-# Netscape 6.0+, Internet explorer 5.0+, or Konqueror). Windows users are 
+# Windows users are probably better off using the HTML help feature.
-# probably better off using the HTML help feature. Other possible values 
-# for this tag are: HIERARCHIES, which will generate the Groups, Directories,
-# and Class Hierarchy pages using a tree view instead of an ordered list;
-# ALL, which combines the behavior of FRAME and HIERARCHIES; and NONE, which
-# disables this behavior completely. For backwards compatibility with previous
-# releases of Doxygen, the values YES and NO are equivalent to FRAME and NONE
-# respectively.
 
 GENERATE_TREEVIEW      = NO
 
+# By enabling USE_INLINE_TREES, doxygen will generate the Groups, Directories,
+# and Class Hierarchy pages using a tree view instead of an ordered list.
+
+USE_INLINE_TREES       = NO
+
 # If the treeview is enabled (see GENERATE_TREEVIEW) then this tag can be
 # used to set the initial width (in pixels) of the frame in which the tree
 # is shown.
@@ -934,6 +977,13 @@ TREEVIEW_WIDTH         = 250
 
 FORMULA_FONTSIZE       = 10
 
+# When the SEARCHENGINE tag is enable doxygen will generate a search box for the HTML output. The underlying search engine uses javascript
+# and DHTML and should work on any modern browser. Note that when using HTML help (GENERATE_HTMLHELP) or Qt help (GENERATE_QHP)
+# there is already a search function so this one should typically
+# be disabled.
+
+SEARCHENGINE           = NO
+
 #---------------------------------------------------------------------------
 # configuration options related to the LaTeX output
 #---------------------------------------------------------------------------
@@ -1010,6 +1060,10 @@ LATEX_BATCHMODE        = NO
 
 LATEX_HIDE_INDICES     = NO
 
+# If LATEX_SOURCE_CODE is set to YES then doxygen will include source code with syntax highlighting in the LaTeX output. Note that which sources are shown also depends on other settings such as SOURCE_BROWSER.
+
+LATEX_SOURCE_CODE      = NO
+
 #---------------------------------------------------------------------------
 # configuration options related to the RTF output
 #---------------------------------------------------------------------------
@@ -1146,8 +1200,10 @@ GENERATE_PERLMOD       = NO
 PERLMOD_LATEX          = NO
 
 # If the PERLMOD_PRETTY tag is set to YES the Perl module output will be
-# nicely formatted so it can be parsed by a human reader.  This is useful 
+# nicely formatted so it can be parsed by a human reader.
-# if you want to understand what is going on.  On the other hand, if this 
+# This is useful
+# if you want to understand what is going on.
+# On the other hand, if this
 # tag is set to NO the size of the Perl module output will be much smaller
 # and Perl will parse it just the same.
 
@@ -1234,8 +1290,10 @@ SKIP_FUNCTION_MACROS   = YES
 # Optionally an initial location of the external documentation
 # can be added for each tagfile. The format of a tag file without
 # this location is as follows:
+#
 # TAGFILES = file1 file2 ...
 # Adding location for the tag files is done as follows:
+#
 # TAGFILES = file1=loc1 "file2 = loc2" ...
 # where "loc1" and "loc2" can be relative or absolute paths or
 # URLs. If a location is present for each tag, the installdox tool
@@ -1462,12 +1520,3 @@ GENERATE_LEGEND        = YES
 # the various graphs.
 
 DOT_CLEANUP            = YES
-
-#---------------------------------------------------------------------------
-# Configuration::additions related to the search engine   
-#---------------------------------------------------------------------------
-
-# The SEARCHENGINE tag specifies whether or not a search engine should be 
-# used. If set to NO the values of all tags below this one will be ignored.
-
-SEARCHENGINE           = NO

MANIFEST.in

@@ -10,6 +10,7 @@ include testmime.py
 include testutils.py
 include test.py
 include sample.py
+include sgmllib.py
 include milter-template.py
 include test/*
 include Milter/*.py

Milter/__init__.py

@@ -8,12 +8,17 @@
 # Copyright 2001,2009 Business Management Systems, Inc.
 # This code is under the GNU General Public License.  See COPYING for details.
 
-__version__ = '0.9.8'
+from __future__ import print_function
+__version__ = '1.0.2'
 
 import os
 import re
 import milter
-import thread
+try:
+  import thread
+except:
+  # libmilter uses posix threads
+  import _thread as thread
 
 from milter import *
 from functools import wraps
@@ -21,12 +26,6 @@ from functools import wraps
 _seq_lock = thread.allocate_lock()
 _seq = 0
 
-## @fn set_flags(flags)
-# @brief Enable optional %milter actions.
-# Certain %milter actions need to be enabled before calling milter.runmilter()
-# or they throw an exception. 
-# @param flags Bit ored mask of optional actions to enable
-
 def uniqueID():
   """Return a unique sequence number (incremented on each call).
   """
@@ -49,6 +48,12 @@ OPTIONAL_CALLBACKS = {
   'header':(P_NR_HDR,P_NOHDRS)
 }
 
+MACRO_CALLBACKS = {
+  'connect': M_CONNECT,
+  'hello': M_HELO, 'envfrom': M_ENVFROM, 'envrcpt': M_ENVRCPT,
+  'data': M_DATA, 'eom': M_EOM, 'eoh': M_EOH
+}
+
 ## @private
 R = re.compile(r'%+')
 
@@ -107,7 +112,7 @@ def rejected_recipients(klass):
   return enable_protocols(klass,P_RCPT_REJ)
 
 ## Milter leading space on headers. A class decorator that calls
-# enable_protocols() with the P_HEAD_LEADSPC flag.  By default,
+# enable_protocols() with the P_HDR_LEADSPC flag.  By default,
 # header continuation lines are collected and joined before getting
 # sent to a milter.  Headers modified or added by the milter are
 # folded by the MTA as necessary according to its own standards.
@@ -125,7 +130,7 @@ def rejected_recipients(klass):
 # @param klass the %milter application class to modify
 # @return the modified %milter class
 def header_leading_space(klass):
-  return enable_protocols(klass,P_HEAD_LEADSPC)
+  return enable_protocols(klass,P_HDR_LEADSPC)
 
 ## Function decorator to disable callback methods.
 # If the MTA supports it, tells the MTA not to invoke this callback,
@@ -142,6 +147,7 @@ def nocallback(func):
   except KeyError:
     raise ValueError(
       '@nocallback applied to non-optional method: '+func.__name__)
+  @wraps(func)
   def wrapper(self,*args):
     if func(self,*args) != CONTINUE:
       raise RuntimeError('%s return code must be CONTINUE with @nocallback'
@@ -174,6 +180,21 @@ def noreply(func):
   wrapper.milter_protocol = nr_mask
   return wrapper
 
+## Function decorator to set macros used in a callback.
+# By default, the MTA sends all macros defined for a callback.
+# If some or all of these are unused, the bandwidth can be saved
+# by listing the ones that are used.
+# @since 1.0.2
+def symlist(*syms):
+  if len(syms) > 5:
+    raise ValueError('@symlist limited to 5 macros by MTA: '+func.__name__)
+  def setsyms(func):
+    if func.__name__ not in MACRO_CALLBACKS:
+      raise ValueError('@symlist applied to non-symlist method: '+func.__name__)
+    func._symlist = syms
+    return func
+  return setsyms
+
 ## Disabled action exception.
 # set_flags() can tell the MTA that this application will not use certain
 # features (such as CHGFROM).  This can also be negotiated for each
@@ -263,7 +284,7 @@ class Base(object):
   ## Defined by subclasses to write log messages.
   def log(self,*msg): pass
   ## Called for each connection to the MTA.  Called by the
-  # <a href="https://www.milter.org/developers/api/xxfi_connect">
+  # <a href="milter_api/xxfi_connect.html">
   # xxfi_connect</a> callback.  
   # The <code>hostname</code> provided by the local MTA is either
   # the PTR name or the IP in the form "[1.2.3.4]" if no PTR is available.
@@ -300,7 +321,7 @@ class Base(object):
   @nocallback
   def hello(self,hostname): return CONTINUE
   ## Called when the SMTP client says MAIL FROM. Called by the
-  # <a href="https://www.milter.org/developers/api/xxfi_envfrom">
+  # <a href="milter_api/xxfi_envfrom.html">
   # xxfi_envfrom</a> callback.  
   # Returning REJECT rejects the message, but not the connection.
   # The sender is the "envelope" from as defined by
@@ -311,7 +332,7 @@ class Base(object):
   @nocallback
   def envfrom(self,f,*str): return CONTINUE
   ## Called when the SMTP client says RCPT TO. Called by the
-  # <a href="https://www.milter.org/developers/api/xxfi_envrcpt">
+  # <a href="milter_api/xxfi_envrcpt.html">
   # xxfi_envrcpt</a> callback.
   # Returning REJECT rejects the current recipient, not the entire message.
   # The recipient is the "envelope" recipient as defined by 
@@ -371,13 +392,13 @@ class Base(object):
       for func,(nr,nc) in OPTIONAL_CALLBACKS.items():
         func = getattr(klass,func)
         ca = getattr(func,'milter_protocol',0)
-        #print func,hex(nr),hex(nc),hex(ca)
+        #print(func,hex(nr),hex(nc),hex(ca))
         p |= (nr|nc) & ~ca
       klass._protocol_mask = p
       return p
     
   ## Negotiate milter protocol options.  Called by the
-  # <a href="https://www.milter.org/developers/api/xxfi_negotiate">
+  # <a href="milter_api/xxfi_negotiate.html">
   # xffi_negotiate</a> callback.  This is an advanced callback,
   # do not override unless you know what you are doing.  Most
   # negotiation can be done simply by using the supplied 
@@ -394,6 +415,11 @@ class Base(object):
   def negotiate(self,opts):
     try:
       self._actions,p,f1,f2 = opts
+      for func,stage in MACRO_CALLBACKS.items():
+        func = getattr(self,func)
+        syms = getattr(func,'_symlist',None)
+        if syms is not None:
+          self.setsymlist(stage,syms)
       opts[1] = self._protocol = p & ~self.protocol_mask()
       opts[2] = 0
       opts[3] = 0
@@ -408,7 +434,7 @@ class Base(object):
   ## Return the value of an MTA macro.  Sendmail macro names
   # are either single chars (e.g. "j") or multiple chars enclosed
   # in braces (e.g. "{auth_type}").  Macro names are MTA dependent.
-  # See <a href="https://www.milter.org/developers/api/smfi_getsymval">
+  # See <a href="milter_api/smfi_getsymval.html">
   # smfi_getsymval</a> for default sendmail macros.
   # @param sym the macro name
   def getsymval(self,sym):
@@ -422,7 +448,7 @@ class Base(object):
   # head scratching.  What will <i>really</i> irritate you, however,
   # is that if you carefully double any '%%', your message will be
   # sent - but with the '%%' still doubled!
-  # See <a href="https://www.milter.org/developers/api/smfi_setreply">
+  # See <a href="milter_api/smfi_setreply.html">
   # smfi_setreply</a> for more information.
   # @param rcode The three-digit (RFC 821/2821) SMTP reply code as a string. 
   # rcode cannot be None, and <b>must be a valid 4XX or 5XX reply code</b>.
@@ -444,28 +470,32 @@ class Base(object):
   # set.  The protocol stages are M_CONNECT, M_HELO, M_ENVFROM, M_ENVRCPT,
   # M_DATA, M_EOM, M_EOH.
   #
-  # May only be called from negotiate callback.
+  # May only be called from negotiate callback.  Hence, this is an advanced
+  # feature.  Use the @@symlist function decorator to conviently set
+  # the macros used by a callback.
   # @since 0.9.8, previous version was misspelled!
   # @param stage the protocol stage to set to macro list for, 
   # one of the M_* constants defined in Milter
   # @param macros space separated and/or lists of strings
   def setsymlist(self,stage,*macros):
     if not self._actions & SETSYMLIST: raise DisabledAction("SETSYMLIST")
+    if len(macros) > 5:
+      raise ValueError('setsymlist limited to 5 macros by MTA')
     a = []
     for m in macros:
       try:
         m = m.encode('utf8')
       except: pass
       try:
-        m = m.split(' ')
+        m = m.split(b' ')
-      except: pass
         a += m
-    return self._ctx.setsmlist(stage,' '.join(a))
+      except: pass
+    return self._ctx.setsymlist(stage,b' '.join(a))
 
   # Milter methods which can only be called from eom callback.
 
   ## Add a mail header field.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_addheader">
+  # Calls <a href="milter_api/smfi_addheader.html">
   # smfi_addheader</a>.  
   # The <code>Milter.ADDHDRS</code> action flag must be set.
   #
@@ -479,7 +509,7 @@ class Base(object):
     return self._ctx.addheader(field,value,idx)
 
   ## Change the value of a mail header field.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_chgheader">
+  # Calls <a href="milter_api/smfi_chgheader.html">
   # smfi_chgheader</a>.  
   # The <code>Milter.CHGHDRS</code> action flag must be set.
   #
@@ -493,7 +523,7 @@ class Base(object):
     return self._ctx.chgheader(field,idx,value)
 
   ## Add a recipient to the message.  
-  # Calls <a href="https://www.milter.org/developers/api/smfi_addrcpt">
+  # Calls <a href="milter_api/smfi_addrcpt.html">
   # smfi_addrcpt</a>.  
   # If no corresponding mail header is added, this is like a Bcc.
   # The syntax of the recipient is the same as used in the SMTP
@@ -513,7 +543,7 @@ class Base(object):
         raise DisabledAction("ADDRCPT_PAR")
     return self._ctx.addrcpt(rcpt,params)
   ## Delete a recipient from the message.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_delrcpt">
+  # Calls <a href="milter_api/smfi_delrcpt.html">
   # smfi_delrcpt</a>.  
   # The recipient should match one passed to the envrcpt callback.
   # The <code>Milter.DELRCPT</code> action flag must be set.
@@ -526,7 +556,7 @@ class Base(object):
     return self._ctx.delrcpt(rcpt)
 
   ## Replace the message body.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_replacebody">
+  # Calls <a href="milter_api/smfi_replacebody.html">
   # smfi_replacebody</a>.  
   # The entire message body must be replaced.  
   # Call repeatedly with blocks of data until the entire body is transferred.
@@ -540,7 +570,7 @@ class Base(object):
     return self._ctx.replacebody(body)
 
   ## Change the SMTP envelope sender address.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_chgfrom">
+  # Calls <a href="milter_api/smfi_chgfrom.html">
   # smfi_chgfrom</a>.  
   # The syntax of the sender is that same as used in the SMTP
   # MAIL FROM command (and as delivered to the envfrom callback),
@@ -557,7 +587,7 @@ class Base(object):
     return self._ctx.chgfrom(sender,params)
 
   ## Quarantine the message.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_quarantine">
+  # Calls <a href="milter_api/smfi_quarantine.html">
   # smfi_quarantine</a>.  
   # When quarantined, a message goes into the mailq as if to be delivered,
   # but delivery is deferred until the message is unquarantined.
@@ -571,7 +601,7 @@ class Base(object):
     return self._ctx.quarantine(reason)
 
   ## Tell the MTA to wait a bit longer.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_progress">
+  # Calls <a href="milter_api/smfi_progress.html">
   # smfi_progress</a>.  
   # Resets timeouts in the MTA that detect a "hung" milter.
   def progress(self):
@@ -585,9 +615,9 @@ class Milter(Base):
 
   ## Provide simple logging to sys.stdout
   def log(self,*msg):
-    print 'Milter:',
+    print('Milter:',end=None)
-    for i in msg: print i,
+    for i in msg: print(i,end=None)
-    print
+    print()
 
   @noreply
   def connect(self,hostname,family,hostaddr):
@@ -724,28 +754,7 @@ def envcallback(c,args):
 # @param socketname the socket to be passed to milter.setconn()
 # @param timeout the time in secs the MTA should wait for a response before 
 #	considering this %milter dead
-def runmilter(name,socketname,timeout = 0):
+def runmilter(name,socketname,timeout = 0,rmsock=True):
-  # This bit is here on the assumption that you will be starting this filter
-  # before sendmail.  If sendmail is not running and the socket already exists,
-  # libmilter will throw a warning.  If sendmail is running, this is still
-  # safe if there are no messages currently being processed.  It's safer to
-  # shutdown sendmail, kill the filter process, restart the filter, and then
-  # restart sendmail.
-  pos = socketname.find(':')
-  if pos > 1:
-    s = socketname[:pos]
-    fname = socketname[pos+1:]
-  else:
-    s = "unix"
-    fname = socketname
-  if s == "unix" or s == "local":
-    print "Removing %s" % fname
-    try:
-      os.unlink(fname)
-    except os.error, x:
-      import errno
-      if x.errno != errno.ENOENT:
-        raise milter.error(x)
 
   # The default flags set include everything
   # milter.set_flags(milter.ADDHDRS)
@@ -776,6 +785,14 @@ def runmilter(name,socketname,timeout = 0):
         unknown=lambda ctx,cmd: ctx.getpriv().unknown(cmd),
         negotiate=ncb
   )
+
+  # We remove the socket here by default on the assumption that you will be
+  # starting this filter before sendmail.  If sendmail is not running and the
+  # socket already exists, libmilter will throw a warning.  If sendmail is
+  # running, this is still safe if there are no messages currently being
+  # processed.  It's safer to shutdown sendmail, kill the filter process,
+  # restart the filter, and then restart sendmail.
+  milter.opensocket(rmsock)
   start_seq = _seq
   try:
     milter.main()

Milter/cache.py

@@ -46,8 +46,9 @@
 # Copyright 2001,2002,2003,2004,2005 Business Management Systems, Inc.
 # This code is under the GNU General Public License.  See COPYING for details.
 
+from __future__ import print_function
 import time
-from plock import PLock
+from Milter.plock import PLock
 
 class AddrCache(object):
   time_format = '%Y%b%d %H:%M:%S %Z'
@@ -131,8 +132,8 @@ class AddrCache(object):
       if not ts or ts > too_old:
         return res
       del self.cache[lsender]
-      raise KeyError, sender
+      raise KeyError(sender)
-    except KeyError,x:
+    except KeyError as x:
       try:
         user,host = sender.split('@',1)
         return self.__getitem__(host)
@@ -147,7 +148,8 @@ class AddrCache(object):
       if not ts: return		# already permanent
     self.cache[lsender] = (None,res)
     if not res:
-      print >>open(self.fname,'a'),sender
+      with open(self.fname,'a') as fp:
+        print(sender,file=fp)
     
   def __setitem__(self,sender,res):
     lsender = sender.lower()
@@ -155,7 +157,8 @@ class AddrCache(object):
     self.cache[lsender] = (now,res)
     if not res and self.fname:
       s = time.strftime(AddrCache.time_format,time.localtime(now))
-      print >>open(self.fname,'a'),sender,s # log refreshed senders
+      with open(self.fname,'a') as fp:
+        print(sender,s,file=fp) # log refreshed senders
 
   def __len__(self):
     return len(self.cache)

Milter/dns.py

@@ -1,6 +1,7 @@
 ## @package Milter.dns
 # Provide a higher level interface to pydns.
 
+from __future__ import print_function
 import DNS
 from DNS import DNSError
 
@@ -73,6 +74,7 @@ class Session(object):
     if name.endswith('.'): name = name[:-1]
     if not reduce(lambda x,y:x and 0 < len(y) < 64, name.split('.'),True):
         return []   # invalid DNS name (too long or empty)
+    name = name.lower()
     result = self.cache.get( (name, qtype) )
     cname = None
     if result: return result
@@ -96,7 +98,7 @@ class Session(object):
             #return result    # if too many == NX_DOMAIN
             raise DNSError('Length of CNAME chain exceeds %d' % MAX_CNAME)
         cnames[name] = cname
-        if cname in cnames:
+        if cname.lower().rstrip('.') in cnames:
             raise DNSError('CNAME loop')
         result = self.dns(cname, qtype, cnames=cnames)
         if result:
@@ -119,5 +121,5 @@ if __name__ == '__main__':
   import sys
   s = Session()
   for n,t in zip(*[iter(sys.argv[1:])]*2):
-    print n,t
+    print(n,t)
-    print s.dns(n,t)
+    print(s.dns(n,t))

Milter/dsn.py

@@ -69,6 +69,7 @@
 # a DSN or use a null MAIL FROM with an email address obtained from
 # anywhere else.
 #
+from __future__ import print_function
 import smtplib
 import socket
 from email.Message import Message
@@ -230,6 +231,6 @@ Subject: Test
 Test DSN template
 """
   )
-  print msg.as_string()
+  print(msg.as_string())
-  # print send_dsn(f,msg.as_string())
+  # print(send_dsn(f,msg.as_string()))
-  # print send_dsn(q.s,'mail.example.com',msg.as_string())
+  # print(send_dsn(q.s,'mail.example.com',msg.as_string()))

Milter/dynip.py

@@ -9,6 +9,7 @@
 # wiley-268-8196.roadrunner.nf.net at ('205.251.174.46', 4810)
 # cbl-sd-02-79.aster.com.do at ('200.88.62.79', 4153)
 
+from __future__ import print_function
 import re
 
 ip3 = re.compile('[0-9]{1,3}')
@@ -53,11 +54,11 @@ def is_dynip(host,addr):
     if host.find(addr) >= 0: return True
     if addr.find(':') >= 0: return False        # IP6
     a = addr.split('.')
-    ia = map(int,a)
+    ia = list(map(int,a))
     h = host
     m = ip3.findall(host)
     if m:
-      g = map(int,m)[:4]
+      g = list(map(int,m))[:4]
       ia3 = (ia[1:],ia[:3])
       if g[-3:] in ia3: return True
       if g[0] == ia[3] and g[1:3] == ia[:2]: return True
@@ -91,6 +92,6 @@ if __name__ == '__main__':
       if ip in seen: continue
       seen.add(ip)
       if is_dynip(host,ip):
-        print '%s\t%s DYN' % (ip,host)
+        print('%s\t%s DYN' % (ip,host))
       else:
-        print '%s\t%s' % (ip,host)
+        print('%s\t%s' % (ip,host))

Milter/greylist.py

@@ -1,3 +1,4 @@
+from __future__ import print_function
 import time
 import shelve
 import thread
@@ -41,13 +42,24 @@ class Greylist(object):
     self.dbp = shelve.open(dbname,'c',protocol=2)
     self.lock = thread.allocate_lock()
 
+  def export_csv(self,fp,timeinc=0):
+    "Export records to csv."
+    import csv
+    dbp = self.dbp
+    w = csv.writer(fp)
+    now = time.time() + timeinc
+    for key, r in dbp.iteritems():
+      if now > r.lastseen + self.greylist_retain: continue
+      ip,sender,recipient = key.rsplit(':',2)
+      w.writerow([ip,sender,recipient,r.firstseen,r.lastseen,r.cnt,r.umis])
+
   def clean(self,timeinc=0):
     "Delete records past the retention limit."
     now = time.time() + timeinc
     cnt = 0
     dbp = self.dbp
     for key, r in dbp.iteritems():
-      #print key,r,time.ctime(now)
+      #print(key,r,time.ctime(now))
       if now > r.lastseen + self.greylist_retain:
         self.lock.acquire()
         try:
@@ -100,3 +112,10 @@ class Greylist(object):
 
   def close(self):
     self.dbp.close()
+
+if __name__ == '__main__':
+  import sys
+  g = Greylist(sys.argv[1],5,24,36)
+  try:
+    g.export_csv(sys.stdout)
+  finally: g.close()

Milter/greysql.py

@@ -2,7 +2,10 @@ import time
 import logging
 import urllib
 import sqlite3
-import thread
+try:
+  import thread
+except:
+  import _thread as thread
 from datetime import datetime
 
 log = logging.getLogger('milter.greylist')
@@ -25,6 +28,19 @@ class Greylist(object):
         primary key (ip,sender,recipient))''')
     except: pass
 
+  def import_csv(self,fp):
+    import csv
+    rdr = csv.reader(fp)
+    cur = self.conn.execute('begin immediate')
+    try:
+      for r in rdr:
+        cur.execute('''insert into 
+          greylist(ip,sender,recipient,firstseen,lastseen,cnt,umis)
+          values(?,?,?,?,?,?,?)''', r)
+      self.conn.commit()
+    finally:
+      cur.close();
+
   def clean(self,timeinc=0):
     "Delete records past the retention limit."
     now = time.time() + timeinc - self.greylist_retain
@@ -84,3 +100,10 @@ class Greylist(object):
 
   def close(self):
     self.conn.close()
+
+if __name__ == '__main__':
+  import sys
+  g = Greylist(sys.argv[1])
+  try:
+    g.import_csv(sys.stdin)
+  finally: g.close()

Milter/plock.py

@@ -11,7 +11,7 @@ class PLock(object):
     self.basename = basename
     self.fp = None
 
-  def lock(self,lockname=None,mode=0660,strict_perms=False):
+  def lock(self,lockname=None,mode=0o660,strict_perms=False):
     "Start an update transaction.  Return FILE to write new version."
     self.unlock()
     if not lockname:
@@ -21,7 +21,7 @@ class PLock(object):
       st = os.stat(self.basename)
       mode |= st.st_mode
     except OSError: pass
-    u = os.umask(0002)
+    u = os.umask(0o2)
     try:
       fd = os.open(lockname,os.O_WRONLY+os.O_CREAT+os.O_EXCL,mode)
     finally:
@@ -46,7 +46,7 @@ class PLock(object):
   def commit(self,backname=None):
     "Commit update transaction with optional backup file."
     if not self.fp:
-      raise IOError,"File not locked"
+      raise IOError("File not locked")
     self.fp.close()
     self.fp = None
     if backname:

Milter/pyip6.py

@@ -6,6 +6,7 @@ This module is free software, and you may redistribute it and/or modify
 it under the same terms as Python itself, so long as this copyright message
 and disclaimer are retained in their original form.
 """
+from __future__ import print_function
 import struct
 #from spf import RE_IP4 
 import re
@@ -80,11 +81,11 @@ def inet_pton(p):
   (0, 0, 0, 0, 0, 65535, 258, 772)
 
   >>> try: inet_pton('::1.2.3.4.5')
-  ... except ValueError,x: print x
+  ... except ValueError as x: print(x)
   ::1.2.3.4.5
   """
   if p == '::':
-    return '\0'*16
+    return b'\0'*16
   s = p
   m = RE_IP4.search(s)
   try:
@@ -114,4 +115,4 @@ def inet_pton(p):
         return struct.pack('!HHHHHHHH',
 	    *[int(s,16) for s in a[0].split(':')])
   except ValueError: pass
-  raise ValueError,p
+  raise ValueError(p)

Milter/test.py

@@ -1,13 +1,17 @@
 ## @package Milter.test
 # A test framework for milters
 
-import rfc822
+from __future__ import print_function
-import StringIO
+import mime
+try:
+  from io import BytesIO
+except:
+  from StringIO import StringIO as BytesIO
 import Milter
 
 Milter.NOREPLY = Milter.CONTINUE
 
-## Test mixin for unit testing milter applications.
+## Test mixin for unit testing %milter applications.
 # This mixin overrides many Milter.MilterBase methods
 # with stub versions that simply record what was done.
 # @since 0.9.8
@@ -16,6 +20,8 @@ class TestBase(object):
   def __init__(self,logfile='test/milter.log'):
     self._protocol = 0
     self.logfp = open(logfile,"a")
+    ## The MAIL FROM for the current email being fed to the %milter
+    self._sender = None
     ## List of recipients deleted
     self._delrcpt = []
     ## List of recipients added
@@ -24,43 +30,67 @@ class TestBase(object):
     self._macros = { }
     ## The message body.
     self._body = None
-    ## True if the milter replaced the message body.
+    ## True if the %milter replaced the message body.
     self._bodyreplaced = False
-    ## True if the milter changed any headers.
+    ## True if the %milter changed any headers.
     self._headerschanged = False
-    ## Reply codes and messages set by milter
+    ## True if the %milter changed the envelope from.
+    self._envfromchanged = False
+    ## Reply codes and messages set by the %milter
     self._reply = None
-    ## The rfc822 message object for the current email being fed to the milter.
+    ## The rfc822 message object for the current email being fed to the %milter.
     self._msg = None
+    ## The protocol stage for macros returned
+    self._stage = None
+    ## The macros returned by protocol stage
     self._symlist = [ None, None, None, None, None, None, None ]
 
   def log(self,*msg):
-    for i in msg: print >>self.logfp, i,
+    for i in msg: print(i,file=self.logfp,end=None)
-    print >>self.logfp
+    print(file=self.logfp)
 
   ## Set a macro value.
-  # These are retrieved by the milter with getsymval.
+  # These are retrieved by the %milter with getsymval.
   # @param name the macro name, as passed to getsymval
   # @param val the macro value
   def setsymval(self,name,val):
     self._macros[name] = val
     
   def getsymval(self,name):
-    # FIXME: track stage, and use _symlist
+    stage = self._stage
-    return self._macros.get(name,'')
+    if stage >= 0:
+      syms = self._symlist[stage]
+      if syms is not None and name not in syms:
+        return None
+    return self._macros.get(name,None)
 
   def replacebody(self,chunk):
     if self._body:
       self._body.write(chunk)
       self._bodyreplaced = True
     else:
-      raise IOError,"replacebody not called from eom()"
+      raise IOError("replacebody not called from eom()")
+
+  def chgfrom(self,sender,params=None):
+    if not self._body:
+      raise IOError("chgfrom not called from eom()")
+    self.log('chgfrom: sender=%s' % (sender))
+    self._envfromchanged = True
+    self._sender = sender
+
+  # TODO: write implement quarantine()
+  def quarantine(self,reason):
+    raise NotImplemented
+
+  # TODO: measure time between milter calls
+  def progress(self):
+    pass
 
   # FIXME: rfc822 indexing does not really reflect the way chg/add header
-  # work for a milter
+  # work for a %milter
   def chgheader(self,field,idx,value):
     if not self._body:
-      raise IOError,"chgheader not called from eom()"
+      raise IOError("chgheader not called from eom()")
     self.log('chgheader: %s[%d]=%s' % (field,idx,value))
     if value == '':
       del self._msg[field]
@@ -70,19 +100,19 @@ class TestBase(object):
 
   def addheader(self,field,value,idx=-1):
     if not self._body:
-      raise IOError,"addheader not called from eom()"
+      raise IOError("addheader not called from eom()")
     self.log('addheader: %s=%s' % (field,value))
     self._msg[field] = value
     self._headerschanged = True
 
   def delrcpt(self,rcpt):
     if not self._body:
-      raise IOError,"delrcpt not called from eom()"
+      raise IOError("delrcpt not called from eom()")
     self._delrcpt.append(rcpt)
 
   def addrcpt(self,rcpt):
     if not self._body:
-      raise IOError,"addrcpt not called from eom()"
+      raise IOError("addrcpt not called from eom()")
     self._addrcpt.append(rcpt)
 
   ## Save the reply codes and messages in self._reply.
@@ -90,7 +120,10 @@ class TestBase(object):
     self._reply = (rcode,xcode) + msg
 
   def setsymlist(self,stage,macros):
-    if not self._actions & SETSYMLIST: raise DisabledAction("SETSYMLIST")
+    if not self._actions & Milter.SETSYMLIST:
+      raise DisabledAction("SETSYMLIST")
+    if self._stage != -1:
+      raise RuntimeError("setsymlist may only be called from negotiate")
     # not used yet, but just for grins we save the data
     a = []
     for m in macros:
@@ -98,14 +131,18 @@ class TestBase(object):
         m = m.encode('utf8')
       except: pass
       try:
-        m = m.split(' ')
+        m = m.split(b' ')
       except: pass
       a += m
+    if len(a) > 5:
+      raise ValueError('setsymlist limited to 5 macros by MTA')
+    if self._symlist[stage] is not None:
+      raise ValueError('setsymlist already called for stage:'+stage)
     self._symlist[stage] = set(a)
 
-  ## Feed a file like object to the milter.  Calls envfrom, envrcpt for
+  ## Feed a file like object to the %milter.  Calls envfrom, envrcpt for
   # each recipient, header for each header field, body for each body 
-  # block, and finally eom.  A return code from the milter other than
+  # block, and finally eom.  A return code from the %milter other than
   # CONTINUE returns immediately with that return code.
   #
   # This is a convenience method, a test could invoke the callbacks
@@ -119,57 +156,60 @@ class TestBase(object):
     self._bodyreplaced = False
     self._headerschanged = False
     self._reply = None
-    msg = rfc822.Message(fp)
+    self._sender = '<%s>'%sender
-    rc = self.envfrom('<%s>'%sender)
+    msg = mime.message_from_file(fp)
+    # envfrom
+    self._stage = Milter.M_ENVFROM
+    rc = self.envfrom(self._sender)
+    self._stage = None
     if rc != Milter.CONTINUE: return rc
+    # envrcpt
     for rcpt in (rcpt,) + rcpts:
+      self._stage = Milter.M_ENVRCPT
       rc = self.envrcpt('<%s>'%rcpt)
+      self._stage = None
       if rc != Milter.CONTINUE: return rc
-    line = None
+    # data
-    for h in msg.headers:
+    self._stage = Milter.M_DATA
-      if h[:1].isspace():
+    rc = self.data()
-        line = line + h
+    self._stage = None
-        continue
-      if not line:
-        line = h
-        continue
-      s = line.split(': ',1)
-      if len(s) > 1: val = s[1].strip()
-      else: val = ''
-      rc = self.header(s[0],val)
     if rc != Milter.CONTINUE: return rc
-      line = h
+    # header
-    if line:
+    for h,val in msg.items():
-      s = line.split(': ',1)
+      rc = self.header(h,val)
-      rc = self.header(s[0],s[1])
       if rc != Milter.CONTINUE: return rc
+    # eoh
+    self._stage = Milter.M_EOH
     rc = self.eoh()
+    self._stage = None
     if rc != Milter.CONTINUE: return rc
+    # body
+    header,body = msg.as_bytes().split(b'\n\n',1)
+    bfp = BytesIO(body)
     while 1:
-      buf = fp.read(8192)
+      buf = bfp.read(8192)
       if len(buf) == 0: break
       rc = self.body(buf)
       if rc != Milter.CONTINUE: return rc
     self._msg = msg
-    self._body = StringIO.StringIO()
+    self._body = BytesIO()
+    self._stage = Milter.M_EOM
     rc = self.eom()
+    self._stage = None
     if self._bodyreplaced:
       body = self._body.getvalue()
-    else:
+    self._body = BytesIO()
-      msg.rewindbody()
+    self._body.write(header)
-      body = msg.fp.read()
+    self._body.write(b'\n\n')
-    self._body = StringIO.StringIO()
-    self._body.writelines(msg.headers)
-    self._body.write('\n')
     self._body.write(body)
     return rc
 
-  ## Feed an email contained in a file to the milter.
+  ## Feed an email contained in a file to the %milter.
   # This is a convenience method that invokes @link #feedFile feedFile @endlink.
   # @param sender MAIL FROM
   # @param rcpts RCPT TO, multiple recipients may be supplied
   def feedMsg(self,fname,sender="spam@adv.com",*rcpts):
-    with open('test/'+fname,'r') as fp:
+    with open('test/'+fname,'rb') as fp:
       return self.feedFile(fp,sender,*rcpts)
 
   ## Call the connect and helo callbacks.
@@ -180,13 +220,19 @@ class TestBase(object):
   def connect(self,host='localhost',helo='spamrelay',ip='1.2.3.4'):
     self._body = None
     self._bodyreplaced = False
+    self._setctx(None)
     opts = [ Milter.CURR_ACTS,~0,0,0 ]
+    self._stage = -1
     rc = self.negotiate(opts)
+    self._stage = Milter.M_CONNECT
     rc =  super(TestBase,self).connect(host,1,(ip,1234)) 
     if rc != Milter.CONTINUE:
+      self._stage = None
       self.close()
       return rc
+    self._stage = Milter.M_HELO
     rc = self.hello(helo)
+    self._stage = None
     if rc != Milter.CONTINUE:
       self.close()
     return rc

Milter/utils.py

@@ -5,12 +5,13 @@
 import re
 import struct
 import socket
-import email.Errors
+import email.errors
+from email.header import decode_header
+import email.base64mime
 from fnmatch import fnmatchcase
-from email.Header import decode_header
+from binascii import a2b_base64
-#import email.Utils
-import rfc822
 
+dnsre = re.compile(r'^[a-z][-a-z\d.]+$', re.IGNORECASE)
 PAT_IP4 = r'\.'.join([r'(?:\d|[1-9]\d|1\d\d|2[0-4]\d|25[0-5])']*4)
 ip4re = re.compile(PAT_IP4+'$')
 ip6re = re.compile(                 '(?:%(hex4)s:){6}%(ls32)s$'
@@ -53,8 +54,8 @@ if hasattr(socket,'has_ipv6') and socket.has_ipv6:
 else:
     from pyip6 import inet_ntop, inet_pton
 
-MASK = 0xFFFFFFFFL
+MASK = 0xFFFFFFFF
-MASK6 = 0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFL
+MASK6 = 0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
 
 def cidr(i,n,mask=MASK):
   return ~(mask >> n) & mask & i
@@ -67,6 +68,12 @@ def iniplist(ipaddr,iplist):
   True
   >>> iniplist('192.168.0.45',['192.168.0.*'])
   True
+  >>> iniplist('4.2.2.2',['b.resolvers.Level3.net'])
+  True
+  >>> iniplist('2606:2800:220:1::',['example.com/40'])
+  True
+  >>> iniplist('4.2.2.2',['nothing.example.com'])
+  False
   >>> iniplist('2001:610:779:0:223:6cff:fe9a:9cf3',['127.0.0.1','172.20.1.0/24','2001:610:779::/48'])
   True
   >>> iniplist('2G01:610:779:0:223:6cff:fe9a:9cf3',['127.0.0.1','172.20.1.0/24','2001:610:779::/48'])
@@ -75,8 +82,10 @@ def iniplist(ipaddr,iplist):
   ValueError: Invalid ip syntax:2G01:610:779:0:223:6cff:fe9a:9cf3
   """
   if ip4re.match(ipaddr):
+    fam = socket.AF_INET
     ipnum = addr2bin(ipaddr)
   elif ip6re.match(ipaddr):
+    fam = socket.AF_INET6
     ipnum = bin2long6(inet_pton(ipaddr))
   else:
     raise ValueError('Invalid ip syntax:'+ipaddr)
@@ -96,13 +105,21 @@ def iniplist(ipaddr,iplist):
         n = 128
       if cidr(bin2long6(inet_pton(p[0])),n,MASK6) == cidr(ipnum,n,MASK6):
         return True
+    elif dnsre.match(p[0]):
+      try:
+        sfx = '/'.join(['']+p[1:])
+        addrlist = [r[4][0]+sfx for r in socket.getaddrinfo(p[0],25,fam)]
+        if iniplist(ipaddr,addrlist):
+          return True
+      except socket.gaierror: pass
     elif fnmatchcase(ipaddr,pat):
       return True
   return False
 
 ## Split email into Fullname and address.
-# This replaces <code>email.Utils.parseaddr</code> but fixes
+# This replaces <code>email.utils.parseaddr</code> but fixes
 # some <a href="http://bugs.python.org/issue1025395">tricky test cases</a>.
+# Additional tricky cases are still broken.  Patches welcome.
 #
 def parseaddr(t):
   """Split email into Fullname and address.
@@ -116,12 +133,10 @@ def parseaddr(t):
   >>> parseaddr('God@heaven <@hop1.org,@hop2.net:jeff@spec.org>')
   ('God@heaven', 'jeff@spec.org')
   >>> parseaddr('Real Name ((comment)) <addr...@example.com>')
-  ('Real Name', 'addr...@example.com')
+  ('Real Name (comment)', 'addr...@example.com')
-  >>> parseaddr('a(WRONG)@b')
-  ('WRONG', 'a@b')
   """
-  #return email.Utils.parseaddr(t)
+  #return email.utils.parseaddr(t)
-  res = rfc822.parseaddr(t)
+  res = email.utils.parseaddr(t)
   # dirty fix for some broken cases
   if not res[0]:
     pos = t.find('<')
@@ -130,7 +145,7 @@ def parseaddr(t):
       pos1 = addrspec.rfind(':')
       if pos1 > 0:
         addrspec = addrspec[pos1+1:]
-      return rfc822.parseaddr('"%s" <%s>' % (t[:pos].strip(),addrspec))
+      return email.utils.parseaddr('"%s" <%s>' % (t[:pos].strip(),addrspec))
   if not res[1]:
     pos = t.find('<')
     if pos > 0 and t[-1] == '>':
@@ -138,9 +153,19 @@ def parseaddr(t):
       pos1 = addrspec.rfind(':')
       if pos1 > 0:
         addrspec = addrspec[pos1+1:]
-      return rfc822.parseaddr('%s<%s>' % (t[:pos].strip(),addrspec))
+      return email.utils.parseaddr('%s<%s>' % (t[:pos].strip(),addrspec))
   return res
 
+## Fix email.base64mime.decode to add any missing padding
+def decode(s, convert_eols=None):
+  if not s: return s
+  while len(s) % 4: s += '='	# add missing padding
+  dec = a2b_base64(s)
+  if convert_eols:
+      return dec.replace(CRLF, convert_eols)
+  return dec
+    
+email.base64mime.decode = decode
 
 def parse_addr(t):
   """Split email into user,domain.
@@ -185,18 +210,18 @@ def parse_header(val):
     for s,enc in h:
       if enc:
         try:
-          u.append(unicode(s,enc,'replace'))
+          u.append(s.decode(enc,'replace'))
         except LookupError:
-          u.append(unicode(s))
+          u.append(s.decode())
       else:
-        u.append(unicode(s))
+        u.append(s.decode())
-    u = ''.join(u)
+    u = u''.join(u)
-    for enc in ('us-ascii','iso-8859-1','utf8'):
+    for enc in ('us-ascii','iso-8859-1','utf-8'):
       try:
         return u.encode(enc)
       except UnicodeError: continue
   except UnicodeDecodeError: pass
   except LookupError: pass
   except ValueError: pass
-  except email.Errors.HeaderParseError: pass
+  except email.errors.HeaderParseError: pass
   return val

README

@@ -11,25 +11,24 @@ any point, tell Sendmail to reject, discard, or accept the message.
 Requirements
 ------------
 
-This python milter extension: http://www.bmsi.com/python/milter.html
+Python milter extension: http://https://pypi.python.org/pypi/pymilter/
 Python: http://www.python.org
 Sendmail: http://www.sendmail.org
+
 NB: From Sendmail's libmilter/README:
 
 libmilter requires pthread support in the operating system.  Moreover, it
 requires that the library functions it uses are thread safe; which is true
 for the operating systems libmilter has been developed and tested on.  On
 some operating systems this requires special compile time options (e.g.,
-not just -pthread).  libmilter is currently known to work on (modulo
+not just -pthread).  libmilter is currently known to work on (modulo problems
-problems in the pthread support of some specific versions):
+in the pthread support of some specific versions):
 
 FreeBSD 3.x, 4.x
 SunOS 5.x (x >= 5)
 AIX 4.3.x
 HP UX 11.x
 Linux (recent versions/distributions)
-OpenBSD
-AIX 4.1.5
 
 libmilter is currently not supported on:
 
@@ -110,28 +109,6 @@ _FFR_MILTER for the cf macros.  For example,
 
 m4 -D_FFR_MILTER ../m4/cf.m4 myconfig.mc > myconfig.cf
 
-
-RedHat 6.2 Notes
-----------------
-
-The Redhat 6.2 sendmail RPM does not enable milter.  You can obtain a
-modified spec file at
-
-http://www.bmsi.com/linux/rh62/sendmail-rhmilter.spec
-
-use it to rebuild the Redhat 7.2 SRPM.  The RH6.2 SRPM does not have
-recent sendmail security patches.
-
-RedHat 7.2 Notes
-----------------
-
-The Redhat 7.2 sendmail RPM enables milter in sendmail - but does not include
-the headers needed for compiling a milter.  You can obtain a modified spec
-file with a sendmail-devel package that includes the needed static libraries
-and headers at
-
-http://www.bmsi.com/linux/sendmail-rh72.spec
-
 IPv6 Notes
 ----------
 
@@ -192,7 +169,7 @@ Authors
 -------
 
 Jim Niemira (urmane@urmane.org) wrote the original C module and some quick
-and dirty python to use it.  Stuart D. Gathman (stuart@bmsi.com) took that
+and dirty python to use it.  Stuart D. Gathman (stuart@gathman.org) took that
 kludge and added threading and context objects to it, wrote a proper OO
 wrapper (Milter.py) that handles attachments, did lots of testing, packaged
 it with distutils, and generally transformed it from a quick hack to a

TODO

@@ -1,6 +1,2 @@
-Support smfi_negotiate and auto negotiate only those callbacks for which 
-Milter.Milter methods have been overridden.  (Python should be able to
-do that.)
-
 Lookup exact RFC syntax of real name / email and make 
 Milter.utils.parse_addr() pass all unit tests.

doc/mainpage.py

@@ -1,46 +1,48 @@
 ## @mainpage Writing Milters in Python
 #
 # At the lowest level, the <code>milter</code> module provides a thin wrapper
-# around the <a href="https://www.milter.org/developers/api/index"> sendmail
+# around the <a href="milter_api/index.html"> sendmail
 # libmilter API</a>.  This API lets you register callbacks for a number of
 # events in the process of sendmail receiving a message via SMTP.  These
 # events include the initial connection from a MTA, the envelope sender and
 # recipients, the top level mail headers, and the message body.  There are
 # options to mangle all of these components of the message as it passes through
-# the milter.
+# the %milter.
 # 
 # At the next level, the <code>Milter</code> module (note the case difference)
 # provides a Python friendly object oriented wrapper for the low level API.  To
 # use the Milter module, an application registers a 'factory' to create an
 # object for each connection from a MTA to sendmail.  These connection objects
-# must provide methods corresponding to the libmilter callback events.
+# must provide methods corresponding to the libmilter event callbacks.
 # 
-# Each event method returns a code to tell sendmail whether to proceed with
+# Each callback method returns a code to tell sendmail whether to proceed with
 # processing the message.  This is a big advantage of milters over other mail
 # filtering systems.  Unwanted mail can be stopped in its tracks at the
-# earliest possible point.
+# earliest possible point.  The callback return codes are
+# milter.CONTINUE, milter.REJECT, milter.DISCARD, milter.ACCEPT, 
+# milter.TEMPFAIL, milter.SKIP, milter.NOREPLY.
 # 
-# The <code>Milter.Base</code> class provides default implementations for
+# The Milter.Base class provides default implementations for
 # event methods that do nothing, and also provides wrappers for the libmilter
 # methods to mutate the message.  It automatically negotiates with MTA
-# which protocol steps need to be processed by the milter, based on
+# which protocol steps need to be processed by the %milter, based on
 # which callback methods are overridden.
 #
-# The <code>Milter.Milter</code> class provides an alternate default
+# The Milter.Milter class provides an alternate default
-# implementation that logs the main milter events, but otherwise does nothing.
+# implementation that logs the main milter callbacks, but otherwise does
-# It is provided for compatibility.
+# nothing.  It is provided for compatibility.
 # 
-# The <code>mime</code> module provides a wrapper for the Python email package
+# The mime module provides a wrapper for the Python email package
 # that fixes some bugs, and simplifies modifying selected parts of a MIME
 # message.
 #
 # @section threading
 #
 # The libmilter library which pymilter wraps 
-# <a href="https://www.milter.org/developers/overview#SignalHandling">handles
+# <a href="milter_overview#SignalHandling">handles
 # all signals</a> itself, and expects to be called from a single main thread.
 # It handles SIGTERM, SIGHUP, and SIGINT, mapping the first two to 
-# <a href="https://www.milter.org/developers/api/smfi_stop">smfi_stop</a>
+# <a href="milter_api/smfi_stop.html">smfi_stop</a>
 # and the last to an internal ABORT.
 #
 # If you use python threads or threading modules, then signal handling gets
@@ -50,4 +52,37 @@
 # You may find the
 # <a href="http://docs.python.org/release/2.6.6/library/multiprocessing.html">
 # multiprocessing</a> module useful.  It can be a drop-in
-# replacement for threading as illustrated in @ref milter-template.py.
+# replacement for threading as illustrated in 
+# <a href="milter-template_8py-example.html">milter-template.py</a>.
+#
+# @section Useful python packages for milters
+#
+# <a href="https://pypi.python.org/pypi/pyspf">pyspf</a> checks the
+# SMTP envelope sender (MAIL FROM, passed to the Milter.Base.envfrom callback)
+# against a Sender Policy published in DNS by the sending domain.  This
+# can prevent forgery of the MAIL FROM.  SPF is Sender Policy Framework.
+#
+# <a href="https://launchpad.net/dkimpy">pydkim</a> checks a DKIM signature
+# of the email body and headers against a public key published in DNS by
+# the signing domain.  DKIM is DomainKeys Identified Mail.
+#
+# The <a href="https://pypi.python.org/pypi/authres/">authres</a> module 
+# parses and formats the Authentication-Results email header, providing
+# a standard place to summarize the results from DKIM, SPF, rDNS, SMTP AUTH,
+# and other email authentication methods.
+#
+# <a href="https://pypi.python.org/pypi/pydspam/">pydspam</a> wraps 
+# the libdspam API of the <a href="http://dspam.sourceforge.net/">DSPAM</a>
+# project.
+#
+# @section Milters written with pymilter
+#
+# <a href="https://github.com/croessner/vrfydmn">Verify Domain</a> is a
+# Postfix milter that rejects/fixes manipulated From: header
+# on a mail host with multiple virtual domains.
+#
+# <a href="https://pypi.python.org/pypi/milter/">BMS Milter</a> has several
+# milters, a big complicated spam filter that integrates multiple
+# authentication protocols with pydpsm, and two simple ones: spfmilter.py and
+# dkim-milter.py.
+#

doc/milter.py

@@ -3,10 +3,65 @@
 
 ## @package milter
 #
-# A thin wrapper around libmilter.
+# A thin wrapper around libmilter.  Most users will not import
+# milter directly, but will instead import Milter and subclass
+# Milter.Base.  This module gives you ultimate low level control
+# from python.
 #
 
-## Hold context for a milter connection.
+## Continue processing the current connection, message, or recipient.
+CONTINUE  = 0
+##  For a connection-oriented routine, reject this connection; 
+# call Milter.Base.close(). For a message-oriented routine, except
+# Milter.Base.eom() or Milter.Base.abort(), reject this message.  For a
+# recipient-oriented routine, reject the current recipient (but continue
+# processing the current message).
+REJECT    = 1
+
+## For a message- or recipient-oriented routine, accept this message, but
+# silently discard it.  SMFIS_DISCARD should not be returned by a
+# connection-oriented routine.
+DISCARD   = 2
+
+## For a connection-oriented routine, accept this connection without further
+# filter processing; call Milter.Base.close().   For a message- or
+# recipient-oriented routine, accept this message without further filtering.
+ACCEPT    = 3
+
+## Return a temporary failure, i.e., the corresponding SMTP command will return
+# an appropriate 4xx status code. For a message-oriented routine, except
+# Milter.Base.envfrom(), fail for this message.  For a connection-oriented
+# routine, fail for this connection; call Milter.Base.close().  For a recipient-oriented
+# routine, only
+# fail for the current recipient; continue message processing.
+TEMPFAIL  = 4
+
+## Skip further callbacks of the same type in this transaction. 
+# Currently this return value is only allowed in Milter.Base.body(). It can be
+# used if a %milter has received sufficiently many body chunks to make a
+# decision, but still wants to invoke message modification functions that are
+# only allowed to be called from Milter.Base.eom(). Note: the %milter must
+# negotiate this behavior with the MTA, i.e., it must check whether the
+# protocol action SMFIP_SKIP is available and if so, the %milter must request
+# it.
+SKIP      = 5
+
+## Do not send a reply back to the MTA. 
+# The %milter must negotiate this behavior with the MTA, i.e., it must check
+# whether the appropriate protocol action P_NR_* is available and if so,
+# the %milter must request it. If you set the P_NR_* protocol action for a
+# callback, that callback must always reply with NOREPLY. Using any other
+# reply code is a violation of the API. If in some cases your callback may
+# return another value (e.g., due to some resource shortages), then you must
+# not set P_NR_* and you must use CONTINUE as the default return
+# code. (Alternatively you can try to delay reporting the problem to a later
+# callback for which P_NR_* is not set.)
+#
+# This is negotiated and returned automatically by the Milter.noreply 
+# function decorator.
+NOREPLY   = 6
+
+## Hold context for a %milter connection.
 # Each connection to sendmail creates a new <code>SMFICTX</code> struct within
 # libmilter.  The milter module in turn creates a milterContext
 # tied to the <code>SMFICTX</code> struct via <code>smfi_setpriv</code>
@@ -20,52 +75,52 @@
 # and converts function callbacks to instance method invocations.
 #
 class milterContext(object):
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_getsymval">smfi_getsymval</a>.
+  ## Calls <a href="milter_api/smfi_getsymval.html">smfi_getsymval</a>.
   def getsymval(self,sym): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_setreply">
+  ## Calls <a href="milter_api/smfi_setreply.html">
   # smfi_setreply</a> or
-  # <a href="https://www.milter.org/developers/api/smfi_setmlreply">
+  # <a href="milter_api/smfi_setmlreply.html">
   # smfi_setmlreply</a>.
   # @param rcode SMTP response code
   # @param xcode extended SMTP response code
   # @param msg one or more message lines.  If the MTA does not support 
   #     multiline messages, only the first is used.
   def setreply(self,rcode,xcode,*msg): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_addheader">smfi_addheader</a>.
+  ## Calls <a href="milter_api/smfi_addheader.html">smfi_addheader</a>.
   def addheader(self,name,value,idx=-1): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_chgheader">smfi_chgheader</a>.
+  ## Calls <a href="milter_api/smfi_chgheader.html">smfi_chgheader</a>.
   def chgheader(self,name,idx,value): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_addrcpt">smfi_addrcpt</a>.
+  ## Calls <a href="milter_api/smfi_addrcpt.html">smfi_addrcpt</a>.
   def addrcpt(self,rcpt,params=None): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_delrcpt">smfi_delrcpt</a>.
+  ## Calls <a href="milter_api/smfi_delrcpt.html">smfi_delrcpt</a>.
   def delrcpt(self,rcpt): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_replacebody">smfi_replacebody</a>.
+  ## Calls <a href="milter_api/smfi_replacebody.html">smfi_replacebody</a>.
   def replacebody(self,data): pass
   ## Attach a Python object to this connection context.
   # @return the old value or None
   def setpriv(self,priv): pass
   ## Return the Python object attached to this connection context.
   def getpriv(self): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_quarantine">smfi_quarantine</a>.
+  ## Calls <a href="milter_api/smfi_quarantine.html">smfi_quarantine</a>.
   def quarantine(self,reason): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_progress">smfi_progress</a>.
+  ## Calls <a href="milter_api/smfi_progress.html">smfi_progress</a>.
   def progress(self): pass
-  ## Calls <a href="https://www.milter.org/developers/api/smfi_chgfrom">smfi_chgfrom</a>.
+  ## Calls <a href="milter_api/smfi_chgfrom.html">smfi_chgfrom</a>.
   def chgfrom(self,sender,param=None): pass
   ## Tell the MTA which macro values we are interested in for a given stage.
   # Of interest only when you need to squeeze a few more bytes of bandwidth.
   # It may only be called from the negotiate callback.
   # The protocol stages are 
   # M_CONNECT, M_HELO, M_ENVFROM, M_ENVRCPT, M_DATA, M_EOM, M_EOH.
-  # Calls <a href="https://www.milter.org/developers/api/smfi_setsymlist">smfi_setsymlist</a>.
+  # Calls <a href="milter_api/smfi_setsymlist.html">smfi_setsymlist</a>.
   # @param stage protocol stage in which the macro list should be used
   # @param macrolist a space separated list of macro names
   def setsymlist(self,stage,macrolist): pass
 
 class error(Exception): pass
 
-## Enable optional milter actions.
+## Enable optional %milter actions.
-# Certain milter actions need to be enabled before calling main()
+# Certain %milter actions need to be enabled before calling main()
 # or they throw an exception.  Pymilter enables them all by
 # default (since 0.9.2), but you may wish to disable unneeded
 # actions as an optimization.
@@ -83,59 +138,72 @@ def set_abort_callback(cb): pass
 def set_close_callback(cb): pass
 
 ## Sets the return code for untrapped Python exceptions during a callback.
-# Must be one of TEMPFAIL,REJECT,CONTINUE.  The default is TEMPFAIL.
+# The default is TEMPFAIL.  You should not depend on this handler.  Your
-# You should not depend on this handler.  Your application should
+# application should have its own top level exception handler for each
-# have its own top level exception handler for each callback.  You can
+# callback.  You can then choose your own reply message, log the stack track
-# then choose your own reply message, log the stack track were you please,
+# were you please, and so on.  However, if you miss one, this last ditch
-# and so on.  However, if you miss one, this last ditch handler will
+# handler will print a standard stack trace to sys.stderr, and return to
-# print a standard stack trace to sys.stderr, and return to sendmail.
+# sendmail.  
+# @param code one of #TEMPFAIL,#REJECT,#CONTINUE, or since 1.0, #ACCEPT
 def set_exception_policy(code): pass
 
-## Register python milter with libmilter.
+## Register python %milter with libmilter.
-# The name we pass is used to identify the milter in the MTA configuration.
+# The name we pass is used to identify the %milter in the MTA configuration.
 # Callback functions must be set using the set_*_callback() functions before
-# registering the milter.
+# registering the %milter.
 # Three additional callbacks are specified as keyword parameters.  These
 # were added by recent versions of libmilter.  The keyword parameters is
 # a nicer way to do it, I think, since it makes clear that you have to do
-# it before registering.  I may move all the callbacks 
+# it before registering.  I may move all the callbacks in the future (perhaps
-# in the future (perhaps keeping the set functions for compatibility).
+# keeping the set functions for compatibility).  Note that Milter.Base
-# @param name the milter name by which the MTA finds us
+# automatically maps all callbacks to member functions, and negotiates which
+# member functions are actually overridden by an application class.
+# @param name the %milter name by which the MTA finds us
 # @param negotiate the
-#       <a href="https://www.milter.org/developers/api/xxfi_negotiate">
+#       <a href="milter_api/xxfi_negotiate.html">
 #       xxfi_negotiate</a> callback, called to negotiate supported
 #       actions, callbacks, and protocol steps.
 # @param unknown the
-#       <a href="https://www.milter.org/developers/api/xxfi_unknown">
+#       <a href="milter_api/xxfi_unknown.html">
 #       xxfi_unknown</a> callback, called when for SMTP commands
 #       not recognized by the MTA. (Extend SMTP in your milter!)
 # @param data the
-#       <a href="https://www.milter.org/developers/api/xxfi_data">
+#       <a href="milter_api/xxfi_data.html">
 #       xxfi_data</a> callback, called when the DATA
 #       SMTP command is received.
 def register(name,negotiate=None,unknown=None,data=None): pass
+
+## Attempt to create the socket used to communicate with the MTA.
+# milter.opensocket() attempts to create the socket specified previously by a
+# call to milter.setconn() which will be the interface between MTAs and the
+# %milter.  This allows the calling application to ensure that the socket can be
+# created.  If this is not called, milter.main() will do so implicitly.
+# Calls <a href="milter_api/smfi_opensocket.html">
+# smfi_opensocket</a>.  While not documented for libmilter, my experiments
+# indicate that you must call register() before calling opensocket().
+# @param rmsock Try to remove an existing unix domain socket if true.
 def opensocket(rmsock): pass
 
 ## Transfer control to libmilter.
-# Calls <a href="https://www.milter.org/developers/api/smfi_main">
+# Calls <a href="milter_api/smfi_main.html">
 #   smfi_main</a>.
 def main(): pass
 
 ## Set the libmilter debugging level.
-# <a href="https://www.milter.org/developers/api/smfi_setdbg">smfi_setdbg</a>
+# <a href="milter_api/smfi_setdbg.html">smfi_setdbg</a>
-# sets the milter library's internal debugging level to a new level
+# sets the %milter library's internal debugging level to a new level
 # so that code details may be traced. A level of zero turns off debugging. The
 # greater (more positive) the level the more detailed the debugging. Six is the
 # current, highest, useful value.  Must be called before calling main().
 def setdbg(lev): pass
 
 ## Set timeout for MTA communication.
-# Calls <a href="https://www.milter.org/developers/api/smfi_settimeout">
+# Calls <a href="milter_api/smfi_settimeout.html">
 # smfi_settimeout</a>.  Must be called before calling main().
 def settimeout(secs): pass
 
 ## Set socket backlog.
-# Calls <a href="https://www.milter.org/developers/api/smfi_setbacklog">
+# Calls <a href="milter_api/smfi_setbacklog.html">
 # smfi_setbacklog</a>.  Must be called before calling main().
 def setbacklog(n): pass
 
@@ -144,16 +212,21 @@ def setbacklog(n): pass
 # unix, inet, or inet6 socket. By default, a unix domain socket
 # is used.  It must not exist,
 # and sendmail will throw warnings if, eg, the file is under a
-# group or world writable directory.
+# group or world writable directory.  milter.setconn() will not fail with
+# an invalid socket - this will be detected only when calling milter.main()
+# or milter.opensocket().
+# @param s the socket address in proto:address format
 # <pre>
-# setconn('unix:/var/run/pythonfilter')
+# milter.setconn('unix:/var/run/pythonfilter')  # a named pipe
-# setconn('inet:8800') 			# listen on ANY interface
+# milter.setconn('local:/var/run/pythonfilter') # a named pipe
-# setconn('inet:7871@@publichost')	# listen on a specific interface
+# milter.setconn('inet:8800') 			# listen on ANY interface
-# setconn('inet6:8020')
+# milter.setconn('inet:7871@@publichost')	# listen on a specific interface
+# milter.setconn('inet6:8020')
+# milter.setconn('inet6:8020@[2001:db8:1234::1]')      # listen on specific IP
 # </pre>
 def setconn(s): pass
 
-## Stop the milter gracefully.
+## Stop the %milter gracefully.
 def stop(): pass
 
 ## Retrieve diagnostic info.

makefile

@@ -1,16 +1,14 @@
 web:
 	doxygen
+	test -L doc/html/milter_api || ln -sf /usr/share/doc/sendmail-devel-* doc/html/milter_api
+	rsync -ravKk doc/html/ spidey2.bmsi.com:/Public/pymilter
 	cd doc/html; zip -r ../../doc .
-	rsync -ravK doc/html/ spidey2.bmsi.com:/Public/pymilter
 
-VERSION=0.9.8
+VERSION=1.0.2
-CVSTAG=pymilter-0_9_8
 PKG=pymilter-$(VERSION)
 SRCTAR=$(PKG).tar.gz
 
 $(SRCTAR):
-	cvs export -r$(CVSTAG) -d $(PKG) pymilter
+	git archive --format=tar.gz --prefix=$(PKG)/ -o $(SRCTAR) $(PKG)
-	tar cvfz $(PKG).tar.gz $(PKG)
-	rm -r $(PKG)
 
-cvstar: $(SRCTAR)
+gittar: $(SRCTAR)

milter-nomix.py

@@ -2,6 +2,7 @@
 # Internal is defined as using one of a list of internal top level domains.
 #  This code is open-source on the same terms as Python.
 
+from __future__ import print_function
 import Milter
 import time
 import sys
@@ -13,7 +14,7 @@ internal_tlds = ["corp", "personal"]
 # True if internal, False otherwise
 def is_internal(hostname):
     components = hostname.split(".")
-    return components.pop() in internal_tlds:
+    return components.pop() in internal_tlds
 
 # Determine if internal and external hosts are mixed based on a list
 # of hostnames
@@ -68,12 +69,12 @@ def main():
   timeout = 600
   # Register to have the Milter factory create instances of your class:
   Milter.factory = NoMixMilter
-  print "%s milter startup" % time.strftime('%Y%b%d %H:%M:%S')
+  print("%s milter startup" % time.strftime('%Y%b%d %H:%M:%S'))
   sys.stdout.flush()
   Milter.runmilter("nomixfilter",socketname,timeout)
   logq.put(None)
   bt.join()
-  print "%s nomix milter shutdown" % time.strftime('%Y%b%d %H:%M:%S')
+  print("%s nomix milter shutdown" % time.strftime('%Y%b%d %H:%M:%S'))
 
 if __name__ == "__main__":
   main()

milter-template.py

@@ -1,14 +1,18 @@
 ## To roll your own milter, create a class that extends Milter.  
 #  See the pymilter project at http://bmsi.com/python/milter.html
-#  based on Sendmail's milter API http://www.milter.org/milter_api/api.html
+#  based on Sendmail's milter API 
 #  This code is open-source on the same terms as Python.
 
 ## Milter calls methods of your class at milter events.
 ## Return REJECT,TEMPFAIL,ACCEPT to short circuit processing for a message.
 ## You can also add/del recipients, replacebody, add/del headers, etc.
 
+from __future__ import print_function
 import Milter
-import StringIO
+try:
+  from StringIO import StringIO
+except:
+  from io import StringIO
 import time
 import email
 import sys
@@ -71,7 +75,10 @@ class myMilter(Milter.Base):
     self.fromparms = Milter.dictfromlist(str)	# ESMTP parms
     self.user = self.getsymval('{auth_authen}')	# authenticated user
     self.log("mail from:", mailfrom, *str)
-    self.fp = StringIO.StringIO()
+    # NOTE: self.fp is only an *internal* copy of message data.  You
+    # must use addheader, chgheader, replacebody to change the message
+    # on the MTA.
+    self.fp = StringIO()
     self.canon_from = '@'.join(parse_addr(mailfrom))
     self.fp.write('From %s %s\n' % (self.canon_from,time.ctime()))
     return Milter.CONTINUE
@@ -104,7 +111,6 @@ class myMilter(Milter.Base):
   def eom(self):
     self.fp.seek(0)
     msg = email.message_from_file(self.fp)
-    self.setreply('250','2.5.1','Grokked by pymilter')
     # many milter functions can only be called from eom()
     # example of adding a Bcc:
     self.addrcpt('<%s>' % 'spy@example.com')
@@ -129,10 +135,12 @@ def background():
     t = logq.get()
     if not t: break
     msg,id,ts = t
-    print "%s [%d]" % (time.strftime('%Y%b%d %H:%M:%S',time.localtime(ts)),id),
+    print("%s [%d]" % (time.strftime('%Y%b%d %H:%M:%S',time.localtime(ts)),id),
+        end=None)
     # 2005Oct13 02:34:11 [1] msg1 msg2 msg3 ...
-    for i in msg: print i,
+    for i in msg: print(i,end=None)
-    print
+    print()
+    sys.stdout.flush()
 
 ## ===
     
@@ -147,12 +155,12 @@ def main():
   flags += Milter.ADDRCPT
   flags += Milter.DELRCPT
   Milter.set_flags(flags)       # tell Sendmail which features we use
-  print "%s milter startup" % time.strftime('%Y%b%d %H:%M:%S')
+  print("%s milter startup" % time.strftime('%Y%b%d %H:%M:%S'))
   sys.stdout.flush()
   Milter.runmilter("pythonfilter",socketname,timeout)
   logq.put(None)
   bt.join()
-  print "%s bms milter shutdown" % time.strftime('%Y%b%d %H:%M:%S')
+  print("%s bms milter shutdown" % time.strftime('%Y%b%d %H:%M:%S'))
 
 if __name__ == "__main__":
   main()

milter.patch

@@ -0,0 +1,174 @@
+diff --git a/miltermodule.c b/miltermodule.c
+index aa10a08..4d5a93d 100644
+--- a/miltermodule.c
++++ b/miltermodule.c
+@@ -343,7 +343,7 @@ static struct MilterCallback {
+       { NULL , NULL }
+     };
+ 
+-staticforward struct smfiDesc description; /* forward declaration */
++static struct smfiDesc description; /* forward declaration */
+ 
+ static PyObject *MilterError;
+ /* The interpreter instance that called milter.main */
+@@ -355,7 +355,7 @@ typedef struct {
+ 
+ static milter_Diag diag;
+ 
+-staticforward PyTypeObject milter_ContextType;
++static PyTypeObject milter_ContextType;
+ 
+ typedef struct {
+   PyObject_HEAD
+@@ -700,7 +700,7 @@ _generic_wrapper(milter_ContextObject *self, PyObject *cb, PyObject *arglist) {
+   result = PyEval_CallObject(cb, arglist);
+   Py_DECREF(arglist);
+   if (result == NULL) return _report_exception(self);
+-  if (!PyInt_Check(result)) {
++  if (!PyLong_Check(result)) {
+     const struct MilterCallback *p;
+     const char *cbname = "milter";
+     char buf[40];
+@@ -715,7 +715,7 @@ _generic_wrapper(milter_ContextObject *self, PyObject *cb, PyObject *arglist) {
+     PyErr_SetString(MilterError,buf);
+     return _report_exception(self);
+   }
+-  retval = PyInt_AS_LONG(result);
++  retval = PyLong_AS_LONG(result);
+   Py_DECREF(result);
+   _release_thread(self->t);
+   return retval;
+@@ -732,7 +732,7 @@ makeipaddr(struct sockaddr_in *addr) {
+ 	sprintf(buf, "%d.%d.%d.%d",
+ 		(int) (x>>24) & 0xff, (int) (x>>16) & 0xff,
+ 		(int) (x>> 8) & 0xff, (int) (x>> 0) & 0xff);
+-	return PyString_FromString(buf);
++	return PyUnicode_FromString(buf);
+ }
+ 
+ #ifdef HAVE_IPV6_SUPPORT
+@@ -740,8 +740,8 @@ static PyObject *
+ makeip6addr(struct sockaddr_in6 *addr) {
+ 	char buf[100]; /* must be at least INET6_ADDRSTRLEN + 1 */
+ 	const char *s = inet_ntop(AF_INET6, &addr->sin6_addr, buf, sizeof buf);
+-	if (s) return PyString_FromString(s);
+-	return PyString_FromString("inet6:unknown");
++	if (s) return PyUnicode_FromString(s);
++	return PyUnicode_FromString("inet6:unknown");
+ }
+ #endif
+ 
+@@ -832,7 +832,7 @@ generic_env_wrapper(SMFICTX *ctx, PyObject*cb, char **argv) {
+    for (i=0;i<count;i++) {
+      /* There's some error checking performed in do_mkvalue() for a string */
+      /* that's not currently done here - it probably should be */
+-     PyObject *o = PyString_FromStringAndSize(argv[i], strlen(argv[i]));
++     PyObject *o = PyUnicode_FromStringAndSize(argv[i], strlen(argv[i]));
+      if (o == NULL) {	/* out of memory */
+        Py_DECREF(arglist);
+        return _report_exception(self);
+@@ -889,7 +889,7 @@ milter_wrap_body(SMFICTX *ctx, u_char *bodyp, size_t bodylen) {
+    c = _get_context(ctx);
+    if (!c) return SMFIS_TEMPFAIL;
+    /* Unclear whether this should be s#, z#, or t# */
+-   arglist = Py_BuildValue("(Os#)", c, bodyp, bodylen);
++   arglist = Py_BuildValue("(Oy#)", c, bodyp, bodylen);
+    return _generic_wrapper(c, body_callback, arglist);
+ }
+ 
+@@ -963,7 +963,7 @@ milter_wrap_negotiate(SMFICTX *ctx,
+     int i;
+     for (i = 0; i < 4; ++i) {
+       *pa[i] = (i <= len)
+-      	? PyInt_AsUnsignedLongMask(PyList_GET_ITEM(optlist,i))
++      	? PyLong_AsUnsignedLongMask(PyList_GET_ITEM(optlist,i))
+ 	: fa[i];
+     }
+     if (PyErr_Occurred()) {
+@@ -1551,11 +1551,6 @@ static PyMethodDef context_methods[] = {
+   { NULL, NULL }
+ };
+ 
+-static PyObject *
+-milter_Context_getattr(PyObject *self, char *name) {
+-  return Py_FindMethod(context_methods, self, name);
+-}
+-
+ static struct smfiDesc description = {  /* Set some reasonable defaults */
+   "pythonfilter",
+   SMFI_VERSION,
+@@ -1604,14 +1599,13 @@ static PyMethodDef milter_methods[] = {
+ };
+ 
+ static PyTypeObject milter_ContextType = {
+-  PyObject_HEAD_INIT(&PyType_Type)
+-  0,
+-  "milterContext",
++  PyVarObject_HEAD_INIT(&PyType_Type,0)
++  "milter.Context",
+   sizeof(milter_ContextObject),
+   0,
+         milter_Context_dealloc,            /* tp_dealloc */
+         0,               /* tp_print */
+-        milter_Context_getattr,           /* tp_getattr */
++        0,           /* tp_getattr */
+         0,			/* tp_setattr */
+         0,                                      /* tp_compare */
+         0,                 /* tp_repr */
+@@ -1625,6 +1619,13 @@ static PyTypeObject milter_ContextType = {
+         0,                                      /* tp_setattro */
+         0,                                      /* tp_as_buffer */
+         Py_TPFLAGS_DEFAULT,                     /* tp_flags */
++	NULL,	/* Documentation string */
++	0, 	/* call function for all accessible objects */
++    	0,	/* delete references to contained objects */
++    	0,	/* rich comparisons */
++    	0,	/* weak reference enabler */
++    	0, 0,	/* Iterators */
++    	context_methods, /* Attribute descriptor and subclassing stuff */
+ };
+ 
+ static const char milter_documentation[] =
+@@ -1634,17 +1635,31 @@ Libmilter is currently marked FFR, and needs to be explicitly installed.\n\
+ See <sendmailsource>/libmilter/README for details on setting it up.\n";
+ 
+ static void setitem(PyObject *d,const char *name,long val) {
+-  PyObject *v = PyInt_FromLong(val);
++  PyObject *v = PyLong_FromLong(val);
+   PyDict_SetItemString(d,name,v);
+   Py_DECREF(v);
+ }
+ 
+-void
+-initmilter(void) {
++static struct PyModuleDef moduledef = {
++    PyModuleDef_HEAD_INIT,
++    "milter",     	 /* m_name */
++    milter_documentation,/* m_doc */
++    -1,                  /* m_size */
++    milter_methods,      /* m_methods */
++    NULL,                /* m_reload */
++    NULL,                /* m_traverse */
++    NULL,                /* m_clear */
++    NULL,                /* m_free */
++};
++
++PyMODINIT_FUNC PyInit_milter(void) {
+    PyObject *m, *d;
+ 
+-   m = Py_InitModule4("milter", milter_methods, milter_documentation,
+-		      (PyObject*)NULL, PYTHON_API_VERSION);
++   if (PyType_Ready(&milter_ContextType) < 0)
++          return NULL;
++
++   m = PyModule_Create(&moduledef);
++   if (m == NULL) return NULL;
+    d = PyModule_GetDict(m);
+    MilterError = PyErr_NewException("milter.error", NULL, NULL);
+    PyDict_SetItemString(d,"error", MilterError);
+@@ -1710,4 +1725,5 @@ initmilter(void) {
+    setitem(d,"DISCARD",  SMFIS_DISCARD);
+    setitem(d,"ACCEPT",  SMFIS_ACCEPT);
+    setitem(d,"TEMPFAIL",  SMFIS_TEMPFAIL);
++   return m;
+ }

miltermodule.c

@@ -1,6 +1,6 @@
 /* Copyright (C) 2001  James Niemira (niemira@colltech.com, urmane@urmane.org)
  * Portions Copyright (C) 2001,2002,2003,2004,2005,2006,2007
- *   Stuart Gathman (stuart@bmsi.com)
+ *   Stuart Gathman (stuart@gathman.org)
  *
  * This program is free software: you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
@@ -282,7 +282,7 @@ $ python setup.py help
  * published.  Unfortunately I know of no good way to do this
  * other than with OS-specific tests.
  */
-#if defined(__FreeBSD_kernel__) || defined(__linux__)
+#if defined(__FreeBSD__) || defined(__linux__)
 #define HAVE_IPV6_RFC2553
 #include <arpa/inet.h>
 #endif

mime.py

@@ -93,26 +93,33 @@
 # Copyright 2001,2002,2003,2004,2005 Business Management Systems, Inc.
 # This code is under the GNU General Public License.  See COPYING for details.
 
-import StringIO
+from __future__ import print_function
+try:
+  from io import BytesIO, StringIO
+except:
+  from StringIO import StringIO 
+  BytesIO = StringIO
 import socket
 import Milter
 import zipfile
 
 import email
-import email.Message
+from email.message import Message
-from email.Message import Message
+try:
-from email.Generator import Generator
+  from email.generator import BytesGenerator
-from email.Utils import quote
+  from email import message_from_binary_file
-from email import Utils
+except:
-from email.Parser import Parser
+  from email.generator import Generator as BytesGenerator
-from email import Errors
+  from email import message_from_file as message_from_binary_file
+from email.utils import quote
 
-from types import ListType,StringType
+if not getattr(Message,'as_bytes',None):
+  Message.as_bytes = Message.as_string
 
 ## Return a list of filenames in a zip file.
 # Embedded zip files are recursively expanded.
 def zipnames(txt):
-  fp =  StringIO.StringIO(txt)
+  fp =  BytesIO(txt)
   zipf = zipfile.ZipFile(fp,'r')
   names = []
   for nm in zipf.namelist():
@@ -123,7 +130,7 @@ def zipnames(txt):
 
 ## Fix multipart handling in email.Generator.
 #
-class MimeGenerator(Generator):
+class MimeGenerator(BytesGenerator):
     def _dispatch(self, msg):
         # Get the Content-Type: for the message, then try to dispatch to
         # self._handle_<maintype>_<subtype>().  If there's no handler for the
@@ -133,7 +140,7 @@ class MimeGenerator(Generator):
         if msg.is_multipart() and main.lower() != 'multipart':
           self._handle_multipart(msg)
         else:
-          Generator._dispatch(self,msg)
+          BytesGenerator._dispatch(self,msg)
 
 def unquote(s):
     """Remove quotes from a string."""
@@ -150,19 +157,17 @@ def unquote(s):
           return s[1:-1]
     return s
 
-from types import TupleType
-
 def _unquotevalue(value):
-  if isinstance(value, TupleType):
+  if isinstance(value, tuple):
       return value[0], value[1], unquote(value[2])
   else:
       return unquote(value)
 
 #email.Message._unquotevalue = _unquotevalue
 
-from email.Message import _parseparam
+from email.message import _parseparam
 
-## Enhance email.Message 
+## Enhance email.message.Message
 #
 # Tracks modifications to headers of body or any part independently.
 
@@ -203,7 +208,7 @@ class MimeMessage(Message):
        interpret as a name - and hence decide to execute this message."""
     names = []
     for attr,val in self._get_params_preserve([],'content-type'):
-      if isinstance(val, TupleType):
+      if isinstance(val, tuple):
 	  # It's an RFC 2231 encoded parameter
           newvalue = _unquotevalue(val)
           if val[0]:
@@ -238,9 +243,9 @@ class MimeMessage(Message):
     g = MimeGenerator(file)
     g.flatten(self,unixfrom=unixfrom)
 
-  def as_string(self, unixfrom=False):
+  def as_bytes(self, unixfrom=False):
       "Return the entire formatted message as a string."
-      fp = StringIO.StringIO()
+      fp = BytesIO()
       self.dump(fp,unixfrom=unixfrom)
       return fp.getvalue()
 
@@ -301,7 +306,7 @@ class MimeMessage(Message):
     return None
 
 def message_from_file(fp):
-  msg = email.message_from_file(fp,MimeMessage)
+  msg = message_from_binary_file(fp,MimeMessage)
   for part in msg.walk():
     part.modified = False
   assert not msg.ismodified()
@@ -312,7 +317,7 @@ ade,adp,asd,asx,asp,bas,bat,chm,cmd,com,cpl,crt,dll,exe,hlp,hta,inf,ins,isp,js,
 jse,lnk,mdb,mde,msc,msi,msp,mst,ocx,pcd,pif,reg,scr,sct,shs,url,vb,vbe,vbs,wsc,
 wsf,wsh 
 """.split())
-bad_extensions = map(lambda x:'.' + x,extlist.split(','))
+bad_extensions = ['.' + x for x in extlist.split(',')]
 
 def check_ext(name):
   "Check a name for dangerous Winblows extensions."
@@ -351,8 +356,6 @@ def check_name(msg,savname=None,ckname=check_ext,scan_zip=False):
   msg["Content-Type"] = "text/plain; name="+name
   return Milter.CONTINUE
 
-import email.Iterators
-
 def check_attachments(msg,check):
   """Scan attachments.
 msg	MimeMessage
@@ -398,18 +401,18 @@ class _defang:
 # emulate old defang function
 defang = _defang()
 
-import sgmllib
+from sgmllib import SGMLParser as HTMLParser
 
 import re
 declname = re.compile(r'[a-zA-Z][-_.a-zA-Z0-9]*\s*')
 declstringlit = re.compile(r'(\'[^\']*\'|"[^"]*")\s*')
 
-class SGMLFilter(sgmllib.SGMLParser):
+class SGMLFilter(HTMLParser):
   """Parse HTML and pass through all constructs unchanged.  It is intended for
      derived classes to implement exceptional processing for selected cases.
   """
   def __init__(self,out):
-    sgmllib.SGMLParser.__init__(self)
+    HTMLParser.__init__(self)
     self.out = out
 
   def handle_comment(self,comment):
@@ -440,7 +443,7 @@ class SGMLFilter(sgmllib.SGMLParser):
     self.out.write("<!%s>" % data)
 
   def write(self,buf):
-    "Act like a writer.  Why doesn't SGMLParser do this by default?"
+    "Act like a writer.  Why doesn't HTMLParser do this by default?"
     self.feed(buf)
 
   # Python-2.1 sgmllib rejects illegal declarations.  Since various Microsoft
@@ -483,11 +486,14 @@ class HTMLScriptFilter(SGMLFilter):
     self.modified = False
     self.msg = "<!-- WARNING: embedded script removed -->"
   def start_script(self,unused):
+    #print('beg script',unused)
     self.ignoring += 1
     self.modified = True
-    self.out.write(self.msg)
   def end_script(self):
+    #print('end script')
     self.ignoring -= 1
+    if not self.ignoring:
+      self.out.write(self.msg)
   def handle_data(self,data):
     if not self.ignoring: SGMLFilter.handle_data(self,data)
   def handle_comment(self,comment):
@@ -502,14 +508,14 @@ def check_html(msg,savname=None):
       if name and name.lower().endswith(".htm"):
         msgtype = 'text/html'
   if msgtype == 'text/html':
-    out = StringIO.StringIO()
+    out = StringIO()
     htmlfilter = HTMLScriptFilter(out)
     try:
-      htmlfilter.write(msg.get_payload(decode=True))
+      htmlfilter.write(msg.get_payload(decode=True).decode())
       htmlfilter.close()
     #except sgmllib.SGMLParseError:
     except:
-      #mimetools.copyliteral(msg.get_payload(),open('debug.out','w')
+      mimetools.copyliteral(msg.get_payload(),open('debug.out','wb'))
       htmlfilter.close()
       hostname = socket.gethostname()
       msg.set_payload(
@@ -532,14 +538,14 @@ if __name__ == '__main__':
   def _list_attach(msg):
     t = msg.get_content_type()
     p = msg.get_payload(decode=True)
-    print msg.get_filename(),msg.get_content_type(),type(p)
+    print(msg.get_filename(),msg.get_content_type(),type(p))
     msg = msg.get_submsg()
     if isinstance(msg,Message):
       return check_attachments(msg,_list_attach)
     return Milter.CONTINUE
 
   for fname in sys.argv[1:]:
-    fp = open(fname)
+    fp = open(fname,'rb')
     msg = message_from_file(fp)
-    email.Iterators._structure(msg)
+    email.iterators._structure(msg)
     check_attachments(msg,_list_attach)

pep8.dat

@@ -0,0 +1,6 @@
+Check	Description		Justification
+E111	req indent 4		Creates more continuation lines
+E114	req indent 4 cmnt	Same
+E231	req space after ,	makes calls like print() harder to read
+E266	no ## 			Required by Doxygen
+W291	trailing spaces in cmnt	Needed for space preserving para reformat

pep8.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+ignore=`awk -F\\\\t '{ print $1 }' pep8.dat | tail -n +2`
+a=(${ignore})
+list=$(echo "${a[@]}"|tr '[ ]' '[,]')
+echo python3 -m pep8 --ignore="$list" $@

py3milter.spec

@@ -0,0 +1,197 @@
+%if 0%{?rhel} == 7
+%define pythonbase python34
+%else
+%define pythonbase python3
+%endif
+%define __python python3
+
+%define libdir %{_libdir}/pymilter
+%{!?python_sitearch: %define python_sitearch %(%{__python} -c "from distutils.sysconfig import get_python_lib; print get_python_lib(1)")}
+
+Summary: Python interface to sendmail milter API
+Name: %{pythonbase}-pymilter
+Version: 1.0.2
+Release: 1%{dist}
+Source: https://github.com/sdgathman/pymilter/archive/pymilter-%{version}.tar.gz
+Source1: pymilter.te
+# Patch miltermodule to python3
+# FIXME: replace with reverse patch at some point (make py3 the default)
+Patch: milter.patch
+License: GPLv2+
+Group: Development/Libraries
+BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root
+Url: http://www.bmsi.com/python/milter.html
+# python-2.6.4 gets RuntimeError: not holding the import lock
+Requires: %{pythonbase} >= 2.6.5, sendmail-milter >= 8.13
+%if 0%{?fedora} >= 23
+# Need python2.6 specific pydns, not the version for system python
+Recommends: %{pythonbase}-pydns
+%endif
+# Needed for callbacks, not a core function but highly useful for milters
+BuildRequires: ed, %{pythonbase}-devel, sendmail-devel >= 8.13
+
+%description
+This is a python extension module to enable python scripts to
+attach to sendmail's libmilter functionality.  Additional python
+modules provide for navigating and modifying MIME parts, sending
+DSNs, and doing CBV.
+
+%package selinux
+Summary: SELinux policy module for pymilter
+Group: System Environment/Base
+Requires: policycoreutils, selinux-policy, %{name}
+BuildRequires: policycoreutils, checkpolicy
+%if 0%{?epel} >= 6
+BuildRequires: policycoreutils-python
+%else
+BuildRequires: policycoreutils-python-utils
+%endif
+
+%description selinux
+SELinux policy module for using pymilter with sendmail with selinux enforcing
+
+%prep
+%setup -q -n pymilter-%{version}
+%patch -p1 -b .py3
+cp %{SOURCE1} pymilter.te
+
+%build
+env CFLAGS="$RPM_OPT_FLAGS" %{__python} setup.py build
+checkmodule -m -M -o pymilter.mod pymilter.te
+semodule_package -o pymilter.pp -m pymilter.mod
+
+%install
+rm -rf $RPM_BUILD_ROOT
+%{__python} setup.py install --root=$RPM_BUILD_ROOT
+mkdir -p $RPM_BUILD_ROOT%{_localstatedir}/run/milter
+mkdir -p $RPM_BUILD_ROOT%{_localstatedir}/log/milter
+mkdir -p $RPM_BUILD_ROOT%{libdir}
+
+# install selinux modules
+mkdir -p %{buildroot}%{_datadir}/selinux/targeted
+cp -p pymilter.pp %{buildroot}%{_datadir}/selinux/targeted
+
+%files
+%defattr(-,root,root,-)
+%doc README ChangeLog NEWS TODO CREDITS sample.py milter-template.py
+%{python_sitearch}/*
+%{libdir}
+%dir %attr(0755,mail,mail) %{_localstatedir}/run/milter
+%dir %attr(0755,mail,mail) %{_localstatedir}/log/milter
+
+%files selinux
+%doc pymilter.te
+%{_datadir}/selinux/targeted/*
+
+%clean
+rm -rf $RPM_BUILD_ROOT
+
+%post selinux
+/usr/sbin/semodule -s targeted -i %{_datadir}/selinux/targeted/pymilter.pp \
+	&>/dev/null || :
+
+%postun selinux
+if [ $1 -eq 0 ] ; then
+/usr/sbin/semodule -s targeted -r pymilter &> /dev/null || :
+fi
+
+%changelog
+* Tue Dec 13 2016 Stuart Gathman <stuart@gathman.org> 1.0.2-1
+- Fix the last setsymlist misspelling.  Support in test framework and tests.
+- Add @symlist decorator.
+- Change body callback and a few other APIs to use bytes instead of str.
+
+* Tue Sep 20 2016 Stuart Gathman <stuart@gathman.org> 1.0.1-1
+- Support python3
+
+* Sat Mar  1 2014 Stuart Gathman <stuart@gathman.org> 1.0-2
+- Remove start.sh to track EPEL repository, suggest daemonize as replacement
+- Selinux subpackage should not care about pymilter version
+
+* Wed Jun 26 2013 Stuart Gathman <stuart@gathman.org> 1.0-1
+- Allow ACCEPT as untrapped exception policy
+- Optional dir for getaddrset and getaddrdict in Milter.config
+- Show registered milter name in untrapped exception message.
+- Include selinux subpackage
+- Provide Milter.greylist export and Milter.greylist import to migrate data
+
+* Sat Mar  9 2013 Stuart Gathman <stuart@bmsi.com> 0.9.8-1
+- Add Milter.test module for unit testing milters.
+- Fix typo that prevented setsymlist from being active.
+- Change untrapped exception message to:
+- "pymilter: untrapped exception in milter app"
+
+* Thu Apr 12 2012 Stuart Gathman <stuart@bmsi.com> 0.9.7-1
+- Raise RuntimeError when result != CONTINUE for @noreply and @nocallback
+- Remove redundant table in miltermodule
+- Fix CNAME chain duplicating TXT records in Milter.dns (from pyspf).
+
+* Sat Feb 25 2012 Stuart Gathman <stuart@bmsi.com> 0.9.6-1
+- Raise ValueError on unescaped '%' passed to setreply
+- Grace time at end of Greylist window
+
+* Fri Aug 19 2011 Stuart Gathman <stuart@bmsi.com> 0.9.5-1
+- Print milter.error for invalid callback return type.
+  (Since stacktrace is empty, the TypeError exception is confusing.)
+- Fix milter-template.py
+- Tweak Milter.utils.addr2bin and Milter.dynip to handle IP6
+
+* Tue Mar 02 2010 Stuart Gathman <stuart@bmsi.com> 0.9.4-1
+- Handle IP6 in Milter.utils.iniplist()
+- python-2.6
+
+* Thu Jul 02 2009 Stuart Gathman <stuart@bmsi.com> 0.9.3-1
+- Handle source route in Milter.utils.parse_addr()
+- Fix default arg in chgfrom.
+- Disable negotiate callback for libmilter < 8.14.3 (1,0,1)
+
+* Tue Jun 02 2009 Stuart Gathman <stuart@bmsi.com> 0.9.2-3
+- Change result of @noreply callbacks to NOREPLY when so negotiated.
+
+* Tue Jun 02 2009 Stuart Gathman <stuart@bmsi.com> 0.9.2-2
+- Cache callback negotiation
+
+* Thu May 28 2009 Stuart Gathman <stuart@bmsi.com> 0.9.2-1
+- Add new callback support: data,negotiate,unknown
+- Auto-negotiate protocol steps 
+
+* Thu Feb 05 2009 Stuart Gathman <stuart@bmsi.com> 0.9.1-1
+- Fix missing address of optional param to addrcpt
+
+* Wed Jan 07 2009 Stuart Gathman <stuart@bmsi.com> 0.9.0-4
+- Stop using INSTALLED_FILES to make Fedora happy
+- Remove config flag from start.sh glue
+- Own /var/log/milter
+- Use _localstatedir
+
+* Wed Jan 07 2009 Stuart Gathman <stuart@bmsi.com> 0.9.0-2
+- Changes to meet Fedora standards
+
+* Mon Nov 24 2008 Stuart Gathman <stuart@bmsi.com> 0.9.0-1
+- Split pymilter into its own CVS module
+- Support chgfrom and addrcpt_par
+- Support NS records in Milter.dns
+
+* Mon Aug 25 2008 Stuart Gathman <stuart@bmsi.com> 0.8.10-2
+- /var/run/milter directory must be owned by mail
+
+* Mon Aug 25 2008 Stuart Gathman <stuart@bmsi.com> 0.8.10-1
+- improved parsing into email and fullname (still 2 self test failures)
+- implement no-DSN CBV, reduce full DSNs
+
+* Mon Sep 24 2007 Stuart Gathman <stuart@bmsi.com> 0.8.9-1
+- Use ifarch hack to build milter and milter-spf packages as noarch
+- Remove spf dependency from dsn.py, add dns.py
+
+* Fri Jan 05 2007 Stuart Gathman <stuart@bmsi.com> 0.8.8-1
+- move AddrCache, parse_addr, iniplist to Milter package
+- move parse_header to Milter.utils
+- fix plock for missing source and can't change owner/group
+- split out pymilter and pymilter-spf packages
+- move milter apps to /usr/lib/pymilter
+
+* Sat Nov 04 2006 Stuart Gathman <stuart@bmsi.com> 0.8.7-1
+- SPF moved to pyspf RPM
+
+* Tue May 23 2006 Stuart Gathman <stuart@bmsi.com> 0.8.6-2
+- Support CBV timeout

pymilter.spec

@@ -1,22 +1,32 @@
-%define __python python2.6
+%define __python python2
+%if 0%{?rhel} == 6
 %define pythonbase python
+%else
+%define pythonbase python2
+%endif
 
 %define libdir %{_libdir}/pymilter
 %{!?python_sitearch: %define python_sitearch %(%{__python} -c "from distutils.sysconfig import get_python_lib; print get_python_lib(1)")}
 
 Summary: Python interface to sendmail milter API
 Name: %{pythonbase}-pymilter
-Version: 0.9.8
+Version: 1.0.2
 Release: 1%{dist}
-Source: http://downloads.sourceforge.net/pymilter/pymilter-%{version}.tar.gz
+Source: https://github.com/sdgathman/pymilter/archive/pymilter-%{version}.tar.gz
+Source1: pymilter.te
+# Patch miltermodule to python3
+# FIXME: replace with reverse patch at some point (make py3 the default)
+Patch: milter.patch
 License: GPLv2+
 Group: Development/Libraries
 BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root
 Url: http://www.bmsi.com/python/milter.html
 # python-2.6.4 gets RuntimeError: not holding the import lock
-Requires: %{pythonbase} >= 2.6.5, sendmail >= 8.13
+Requires: %{pythonbase} >= 2.6.5, sendmail-milter >= 8.13
+%if 0%{?fedora} >= 23
 # Need python2.6 specific pydns, not the version for system python
-Requires: %{pythonbase}-pydns
+Recommends: %{pythonbase}-pydns
+%endif
 # Needed for callbacks, not a core function but highly useful for milters
 BuildRequires: ed, %{pythonbase}-devel, sendmail-devel >= 8.13
 
@@ -26,11 +36,28 @@ attach to sendmail's libmilter functionality.  Additional python
 modules provide for navigating and modifying MIME parts, sending
 DSNs, and doing CBV.
 
+%package selinux
+Summary: SELinux policy module for pymilter
+Group: System Environment/Base
+Requires: policycoreutils, selinux-policy, %{name}
+BuildRequires: policycoreutils, checkpolicy
+%if 0%{?epel} >= 6
+BuildRequires: policycoreutils-python
+%else
+BuildRequires: policycoreutils-python-utils
+%endif
+
+%description selinux
+SELinux policy module for using pymilter with sendmail with selinux enforcing
+
 %prep
 %setup -q -n pymilter-%{version}
+cp %{SOURCE1} pymilter.te
 
 %build
 env CFLAGS="$RPM_OPT_FLAGS" %{__python} setup.py build
+checkmodule -m -M -o pymilter.mod pymilter.te
+semodule_package -o pymilter.pp -m pymilter.mod
 
 %install
 rm -rf $RPM_BUILD_ROOT
@@ -38,31 +65,11 @@ rm -rf $RPM_BUILD_ROOT
 mkdir -p $RPM_BUILD_ROOT%{_localstatedir}/run/milter
 mkdir -p $RPM_BUILD_ROOT%{_localstatedir}/log/milter
 mkdir -p $RPM_BUILD_ROOT%{libdir}
-cp start.sh $RPM_BUILD_ROOT%{libdir}
-ed $RPM_BUILD_ROOT%{libdir}/start.sh <<'EOF'
-/^datadir=/
-c
-datadir="%{_localstatedir}/log/milter"
-.
-/^piddir=/
-c
-piddir="%{_localstatedir}/run/milter"
-.
-/^libdir=/
-c
-libdir="%{libdir}"
-.
-/^python=/
-c
-python="%{__python}"
-.
-w
-q
-EOF
-chmod a+x $RPM_BUILD_ROOT%{libdir}/start.sh
 
-# start.sh is used by spfmilter, srsmilter, and milter, and could be used by
+# install selinux modules
-# other milters using pymilter.
+mkdir -p %{buildroot}%{_datadir}/selinux/targeted
+cp -p pymilter.pp %{buildroot}%{_datadir}/selinux/targeted
+
 %files
 %defattr(-,root,root,-)
 %doc README ChangeLog NEWS TODO CREDITS sample.py milter-template.py
@@ -71,17 +78,49 @@ chmod a+x $RPM_BUILD_ROOT%{libdir}/start.sh
 %dir %attr(0755,mail,mail) %{_localstatedir}/run/milter
 %dir %attr(0755,mail,mail) %{_localstatedir}/log/milter
 
+%files selinux
+%doc pymilter.te
+%{_datadir}/selinux/targeted/*
+
 %clean
 rm -rf $RPM_BUILD_ROOT
 
+%post selinux
+/usr/sbin/semodule -s targeted -i %{_datadir}/selinux/targeted/pymilter.pp \
+	&>/dev/null || :
+
+%postun selinux
+if [ $1 -eq 0 ] ; then
+/usr/sbin/semodule -s targeted -r pymilter &> /dev/null || :
+fi
+
 %changelog
+* Tue Dec 13 2016 Stuart Gathman <stuart@gathman.org> 1.0.2-1
+- Fix the last setsymlist misspelling.  Support in test framework and tests.
+- Add @symlist decorator.
+- Change body callback and a few other APIs to use bytes instead of str.
+
+* Tue Sep 20 2016 Stuart Gathman <stuart@gathman.org> 1.0.1-1
+- Support python3
+
+* Sat Mar  1 2014 Stuart Gathman <stuart@gathman.org> 1.0-2
+- Remove start.sh to track EPEL repository, suggest daemonize as replacement
+- Selinux subpackage should not care about pymilter version
+
+* Wed Jun 26 2013 Stuart Gathman <stuart@gathman.org> 1.0-1
+- Allow ACCEPT as untrapped exception policy
+- Optional dir for getaddrset and getaddrdict in Milter.config
+- Show registered milter name in untrapped exception message.
+- Include selinux subpackage
+- Provide Milter.greylist export and Milter.greylist import to migrate data
+
 * Sat Mar  9 2013 Stuart Gathman <stuart@bmsi.com> 0.9.8-1
 - Add Milter.test module for unit testing milters.
 - Fix typo that prevented setsymlist from being active.
 - Change untrapped exception message to:
 - "pymilter: untrapped exception in milter app"
 
-* Sat Feb 25 2012 Stuart Gathman <stuart@bmsi.com> 0.9.7-1
+* Thu Apr 12 2012 Stuart Gathman <stuart@bmsi.com> 0.9.7-1
 - Raise RuntimeError when result != CONTINUE for @noreply and @nocallback
 - Remove redundant table in miltermodule
 - Fix CNAME chain duplicating TXT records in Milter.dns (from pyspf).
@@ -96,7 +135,7 @@ rm -rf $RPM_BUILD_ROOT
 - Fix milter-template.py
 - Tweak Milter.utils.addr2bin and Milter.dynip to handle IP6
 
-* Wed Mar 02 2010 Stuart Gathman <stuart@bmsi.com> 0.9.4-1
+* Tue Mar 02 2010 Stuart Gathman <stuart@bmsi.com> 0.9.4-1
 - Handle IP6 in Milter.utils.iniplist()
 - python-2.6
 

pymilter.te

@@ -0,0 +1,13 @@
+module pymilter 1.0;
+
+require {
+	type sendmail_t;
+	type var_run_t;
+	type initrc_t;
+	class sock_file { write getattr };
+	class unix_stream_socket connectto;
+}
+
+#============= sendmail_t ==============
+allow sendmail_t initrc_t:unix_stream_socket connectto;
+allow sendmail_t var_run_t:sock_file { write getattr };

sample.py

@@ -1,4 +1,4 @@
-
+from __future__ import print_function
 # A simple milter.
 
 # Author: Stuart D. Gathman <stuart@bmsi.com>
@@ -7,8 +7,10 @@
 
 import sys
 import os
-import StringIO
+try:
-import rfc822
+  from io import BytesIO
+except:
+  from StringIO import StringIO as BytesIO
 import mime
 import Milter
 import tempfile
@@ -21,9 +23,9 @@ class sampleMilter(Milter.Milter):
   "Milter to replace attachments poisonous to Windows with a WARNING message."
 
   def log(self,*msg):
-    print "%s [%d]" % (strftime('%Y%b%d %H:%M:%S'),self.id),
+    print("%s [%d]" % (strftime('%Y%b%d %H:%M:%S'),self.id),end=None)
-    for i in msg: print i,
+    for i in msg: print(i,end=None)
-    print
+    print()
 
   def __init__(self):
     self.tempname = None
@@ -31,18 +33,25 @@ class sampleMilter(Milter.Milter):
     self.fp = None
     self.bodysize = 0
     self.id = Milter.uniqueID()
+    self.user = None
 
   # multiple messages can be received on a single connection
   # envfrom (MAIL FROM in the SMTP protocol) seems to mark the start
   # of each message.
+  @Milter.symlist('{auth_authen}')
   @Milter.noreply
   def envfrom(self,f,*str):
     "start of MAIL transaction"
-    self.log("mail from",f,str)
+    self.fp = BytesIO()
-    self.fp = StringIO.StringIO()
     self.tempname = None
     self.mailfrom = f
     self.bodysize = 0
+    self.user = self.getsymval('{auth_authen}')
+    self.auth_type = self.getsymval('{auth_type}')
+    if self.user:
+      self.log("user",self.user,"sent mail from",f,str)
+    else:
+      self.log("mail from",f,str)
     return Milter.CONTINUE
 
   def envrcpt(self,to,*str):
@@ -95,12 +104,12 @@ class sampleMilter(Milter.Milter):
     if lname in ('subject','x-mailer'):
       self.log('%s: %s' % (name,val))
     if self.fp:
-      self.fp.write("%s: %s\n" % (name,val))	# add header to buffer
+      self.fp.write(("%s: %s\n" % (name,val)).encode())	# add header to buffer
     return Milter.CONTINUE
 
   def eoh(self):
     if not self.fp: return Milter.TEMPFAIL	# not seen by envfrom
-    self.fp.write("\n")
+    self.fp.write(b'\n')
     self.fp.seek(0)
     # copy headers to a temp file for scanning the body
     headers = self.fp.getvalue()
@@ -138,19 +147,16 @@ class sampleMilter(Milter.Milter):
     self.log("Temp file:",self.tempname)
     self.tempname = None	# prevent removal of original message copy
     # copy defanged message to a temp file 
-    out = tempfile.TemporaryFile()
+    with tempfile.TemporaryFile() as out:
-    try:
       msg.dump(out)
       out.seek(0)
-      msg = rfc822.Message(out)
+      msg = mime.message_from_file(out)
-      msg.rewindbody()
+      fp = BytesIO(msg.as_bytes().split(b'\n\n',1)[1])
       while 1:
-        buf = out.read(8192)
+        buf = fp.read(8192)
         if len(buf) == 0: break
         self.replacebody(buf)	# feed modified message to sendmail
       return Milter.ACCEPT	# ACCEPT modified message
-    finally:
-      out.close()
     return Milter.TEMPFAIL
 
   def close(self):
@@ -171,13 +177,13 @@ if __name__ == "__main__":
   socketname = os.getenv("HOME") + "/pythonsock"
   Milter.factory = sampleMilter
   Milter.set_flags(Milter.CHGBODY + Milter.CHGHDRS + Milter.ADDHDRS)
-  print """To use this with sendmail, add the following to sendmail.cf:
+  print("""To use this with sendmail, add the following to sendmail.cf:
 
 O InputMailFilters=pythonfilter
 Xpythonfilter,        S=local:%s
 
 See the sendmail README for libmilter.
-sample  milter startup""" % socketname
+sample  milter startup""" % socketname)
   sys.stdout.flush()
   Milter.runmilter("pythonfilter",socketname,240)
-  print "sample milter shutdown"
+  print("sample milter shutdown")

setup.cfg

@@ -1,5 +1,5 @@
 [bdist_rpm]
-python=python2.6
+python=python3
-doc_files=README NEWS TODO
+doc_files=README NEWS TODO COPYING CREDITS
-packager=Stuart D. Gathman <stuart@bmsi.com>
+packager=Stuart D. Gathman <stuart@gathman.org>
 release=1

setup.py

@@ -11,9 +11,13 @@ if sys.version < '2.6.5':
 #libs = ["milter", "smutil"]
 libs = ["milter"]
 libdirs = ["/usr/lib/libmilter"]    # needed for Debian
+modules = ["mime"]
+if sys.version >= '3':
+  modules.append("sgmllib")
+  print("modules=",modules)
 
 # NOTE: importing Milter to obtain version fails when milter.so not built
-setup(name = "pymilter", version = '0.9.8',
+setup(name = "pymilter", version = '1.0.2',
 	description="Python interface to sendmail milter API",
 	long_description="""\
 This is a python extension module to enable python scripts to
@@ -24,10 +28,10 @@ sending DSNs or doing CBVs.
 	author="Jim Niemira",
 	author_email="urmane@urmane.org",
 	maintainer="Stuart D. Gathman",
-	maintainer_email="stuart@bmsi.com",
+	maintainer_email="stuart@gathman.org",
 	license="GPL",
-	url="http://www.bmsi.com/python/milter.html",
+	url="https://pythonhosted.org/milter/",
-	py_modules=["mime"],
+	py_modules=modules,
 	packages = ['Milter'],
 	ext_modules=[
 	  Extension("milter", ["miltermodule.c"],

sgmllib.py

@@ -0,0 +1,553 @@
+
+"""A parser for SGML, using the derived class as a static DTD."""
+
+# XXX This only supports those SGML features used by HTML.
+
+# XXX There should be a way to distinguish between PCDATA (parsed
+# character data -- the normal case), RCDATA (replaceable character
+# data -- only char and entity references and end tags are special)
+# and CDATA (character data -- only end tags are special).  RCDATA is
+# not supported at all.
+
+from __future__ import print_function
+
+try:
+  import _markupbase
+except:
+  import markupbase as _markupbase
+import re
+
+__all__ = ["SGMLParser", "SGMLParseError"]
+
+# Regular expressions used for parsing
+
+interesting = re.compile('[&<]')
+incomplete = re.compile('&([a-zA-Z][a-zA-Z0-9]*|#[0-9]*)?|'
+                           '<([a-zA-Z][^<>]*|'
+                              '/([a-zA-Z][^<>]*)?|'
+                              '![^<>]*)?')
+
+entityref = re.compile('&([a-zA-Z][-.a-zA-Z0-9]*)[^a-zA-Z0-9]')
+charref = re.compile('&#([0-9]+)[^0-9]')
+
+starttagopen = re.compile('<[>a-zA-Z]')
+shorttagopen = re.compile('<[a-zA-Z][-.a-zA-Z0-9]*/')
+shorttag = re.compile('<([a-zA-Z][-.a-zA-Z0-9]*)/([^/]*)/')
+piclose = re.compile('>')
+endbracket = re.compile('[<>]')
+tagfind = re.compile('[a-zA-Z][-_.a-zA-Z0-9]*')
+attrfind = re.compile(
+    r'\s*([a-zA-Z_][-:.a-zA-Z_0-9]*)(\s*=\s*'
+    r'(\'[^\']*\'|"[^"]*"|[][\-a-zA-Z0-9./,:;+*%?!&$\(\)_#=~\'"@]*))?')
+
+
+class SGMLParseError(RuntimeError):
+    """Exception raised for all parse errors."""
+    pass
+
+
+# SGML parser base class -- find tags and call handler functions.
+# Usage: p = SGMLParser(); p.feed(data); ...; p.close().
+# The dtd is defined by deriving a class which defines methods
+# with special names to handle tags: start_foo and end_foo to handle
+# <foo> and </foo>, respectively, or do_foo to handle <foo> by itself.
+# (Tags are converted to lower case for this purpose.)  The data
+# between tags is passed to the parser by calling self.handle_data()
+# with some data as argument (the data may be split up in arbitrary
+# chunks).  Entity references are passed by calling
+# self.handle_entityref() with the entity reference as argument.
+
+class SGMLParser(_markupbase.ParserBase):
+    # Definition of entities -- derived classes may override
+    entity_or_charref = re.compile('&(?:'
+      '([a-zA-Z][-.a-zA-Z0-9]*)|#([0-9]+)'
+      ')(;?)')
+
+    def __init__(self, verbose=0):
+        """Initialize and reset this instance."""
+        self.verbose = verbose
+        self.reset()
+
+    def reset(self):
+        """Reset this instance. Loses all unprocessed data."""
+        self.__starttag_text = None
+        self.rawdata = ''
+        self.stack = []
+        self.lasttag = '???'
+        self.nomoretags = 0
+        self.literal = 0
+        _markupbase.ParserBase.reset(self)
+
+    def setnomoretags(self):
+        """Enter literal mode (CDATA) till EOF.
+
+        Intended for derived classes only.
+        """
+        self.nomoretags = self.literal = 1
+
+    def setliteral(self, *args):
+        """Enter literal mode (CDATA).
+
+        Intended for derived classes only.
+        """
+        self.literal = 1
+
+    def feed(self, data):
+        """Feed some data to the parser.
+
+        Call this as often as you want, with as little or as much text
+        as you want (may include '\n').  (This just saves the text,
+        all the processing is done by goahead().)
+        """
+
+        self.rawdata = self.rawdata + data
+        self.goahead(0)
+
+    def close(self):
+        """Handle the remaining data."""
+        self.goahead(1)
+
+    def error(self, message):
+        raise SGMLParseError(message)
+
+    # Internal -- handle data as far as reasonable.  May leave state
+    # and data to be processed by a subsequent call.  If 'end' is
+    # true, force handling all data as if followed by EOF marker.
+    def goahead(self, end):
+        rawdata = self.rawdata
+        i = 0
+        n = len(rawdata)
+        while i < n:
+            if self.nomoretags:
+                self.handle_data(rawdata[i:n])
+                i = n
+                break
+            match = interesting.search(rawdata, i)
+            if match: j = match.start()
+            else: j = n
+            if i < j:
+                self.handle_data(rawdata[i:j])
+            i = j
+            if i == n: break
+            if rawdata[i] == '<':
+                if starttagopen.match(rawdata, i):
+                    if self.literal:
+                        self.handle_data(rawdata[i])
+                        i = i+1
+                        continue
+                    k = self.parse_starttag(i)
+                    if k < 0: break
+                    i = k
+                    continue
+                if rawdata.startswith("</", i):
+                    k = self.parse_endtag(i)
+                    if k < 0: break
+                    i = k
+                    self.literal = 0
+                    continue
+                if self.literal:
+                    if n > (i + 1):
+                        self.handle_data("<")
+                        i = i+1
+                    else:
+                        # incomplete
+                        break
+                    continue
+                if rawdata.startswith("<!--", i):
+                        # Strictly speaking, a comment is --.*--
+                        # within a declaration tag <!...>.
+                        # This should be removed,
+                        # and comments handled only in parse_declaration.
+                    k = self.parse_comment(i)
+                    if k < 0: break
+                    i = k
+                    continue
+                if rawdata.startswith("<?", i):
+                    k = self.parse_pi(i)
+                    if k < 0: break
+                    i = i+k
+                    continue
+                if rawdata.startswith("<!", i):
+                    # This is some sort of declaration; in "HTML as
+                    # deployed," this should only be the document type
+                    # declaration ("<!DOCTYPE html...>").
+                    k = self.parse_declaration(i)
+                    if k < 0: break
+                    i = k
+                    continue
+            elif rawdata[i] == '&':
+                if self.literal:
+                    self.handle_data(rawdata[i])
+                    i = i+1
+                    continue
+                match = charref.match(rawdata, i)
+                if match:
+                    name = match.group(1)
+                    self.handle_charref(name)
+                    i = match.end(0)
+                    if rawdata[i-1] != ';': i = i-1
+                    continue
+                match = entityref.match(rawdata, i)
+                if match:
+                    name = match.group(1)
+                    self.handle_entityref(name)
+                    i = match.end(0)
+                    if rawdata[i-1] != ';': i = i-1
+                    continue
+            else:
+                self.error('neither < nor & ??')
+            # We get here only if incomplete matches but
+            # nothing else
+            match = incomplete.match(rawdata, i)
+            if not match:
+                self.handle_data(rawdata[i])
+                i = i+1
+                continue
+            j = match.end(0)
+            if j == n:
+                break # Really incomplete
+            self.handle_data(rawdata[i:j])
+            i = j
+        # end while
+        if end and i < n:
+            self.handle_data(rawdata[i:n])
+            i = n
+        self.rawdata = rawdata[i:]
+        # XXX if end: check for empty stack
+
+    # Extensions for the DOCTYPE scanner:
+    _decl_otherchars = '='
+
+    # Internal -- parse processing instr, return length or -1 if not terminated
+    def parse_pi(self, i):
+        rawdata = self.rawdata
+        if rawdata[i:i+2] != '<?':
+            self.error('unexpected call to parse_pi()')
+        match = piclose.search(rawdata, i+2)
+        if not match:
+            return -1
+        j = match.start(0)
+        self.handle_pi(rawdata[i+2: j])
+        j = match.end(0)
+        return j-i
+
+    def get_starttag_text(self):
+        return self.__starttag_text
+
+    # Internal -- handle starttag, return length or -1 if not terminated
+    def parse_starttag(self, i):
+        self.__starttag_text = None
+        start_pos = i
+        rawdata = self.rawdata
+        if shorttagopen.match(rawdata, i):
+            # SGML shorthand: <tag/data/ == <tag>data</tag>
+            # XXX Can data contain &... (entity or char refs)?
+            # XXX Can data contain < or > (tag characters)?
+            # XXX Can there be whitespace before the first /?
+            match = shorttag.match(rawdata, i)
+            if not match:
+                return -1
+            tag, data = match.group(1, 2)
+            self.__starttag_text = '<%s/' % tag
+            tag = tag.lower()
+            k = match.end(0)
+            self.finish_shorttag(tag, data)
+            self.__starttag_text = rawdata[start_pos:match.end(1) + 1]
+            return k
+        # XXX The following should skip matching quotes (' or ")
+        # As a shortcut way to exit, this isn't so bad, but shouldn't
+        # be used to locate the actual end of the start tag since the
+        # < or > characters may be embedded in an attribute value.
+        match = endbracket.search(rawdata, i+1)
+        if not match:
+            return -1
+        j = match.start(0)
+        # Now parse the data between i+1 and j into a tag and attrs
+        attrs = []
+        if rawdata[i:i+2] == '<>':
+            # SGML shorthand: <> == <last open tag seen>
+            k = j
+            tag = self.lasttag
+        else:
+            match = tagfind.match(rawdata, i+1)
+            if not match:
+                self.error('unexpected call to parse_starttag')
+            k = match.end(0)
+            tag = rawdata[i+1:k].lower()
+            self.lasttag = tag
+        while k < j:
+            match = attrfind.match(rawdata, k)
+            if not match: break
+            attrname, rest, attrvalue = match.group(1, 2, 3)
+            if not rest:
+                attrvalue = attrname
+            else:
+                if (attrvalue[:1] == "'" == attrvalue[-1:] or
+                    attrvalue[:1] == '"' == attrvalue[-1:]):
+                    # strip quotes
+                    attrvalue = attrvalue[1:-1]
+                attrvalue = self.entity_or_charref.sub(
+                    self._convert_ref, attrvalue)
+            attrs.append((attrname.lower(), attrvalue))
+            k = match.end(0)
+        if rawdata[j] == '>':
+            j = j+1
+        self.__starttag_text = rawdata[start_pos:j]
+        self.finish_starttag(tag, attrs)
+        return j
+
+    # Internal -- convert entity or character reference
+    def _convert_ref(self, match):
+        if match.group(2):
+            return self.convert_charref(match.group(2)) or \
+                '&#%s%s' % match.groups()[1:]
+        elif match.group(3):
+            return self.convert_entityref(match.group(1)) or \
+                '&%s;' % match.group(1)
+        else:
+            return '&%s' % match.group(1)
+
+    # Internal -- parse endtag
+    def parse_endtag(self, i):
+        rawdata = self.rawdata
+        match = endbracket.search(rawdata, i+1)
+        if not match:
+            return -1
+        j = match.start(0)
+        tag = rawdata[i+2:j].strip().lower()
+        if rawdata[j] == '>':
+            j = j+1
+        self.finish_endtag(tag)
+        return j
+
+    # Internal -- finish parsing of <tag/data/ (same as <tag>data</tag>)
+    def finish_shorttag(self, tag, data):
+        self.finish_starttag(tag, [])
+        self.handle_data(data)
+        self.finish_endtag(tag)
+
+    # Internal -- finish processing of start tag
+    # Return -1 for unknown tag, 0 for open-only tag, 1 for balanced tag
+    def finish_starttag(self, tag, attrs):
+        try:
+            method = getattr(self, 'start_' + tag)
+        except AttributeError:
+            try:
+                method = getattr(self, 'do_' + tag)
+            except AttributeError:
+                self.unknown_starttag(tag, attrs)
+                return -1
+            else:
+                self.handle_starttag(tag, method, attrs)
+                return 0
+        else:
+            self.stack.append(tag)
+            self.handle_starttag(tag, method, attrs)
+            return 1
+
+    # Internal -- finish processing of end tag
+    def finish_endtag(self, tag):
+        if not tag:
+            found = len(self.stack) - 1
+            if found < 0:
+                self.unknown_endtag(tag)
+                return
+        else:
+            if tag not in self.stack:
+                try:
+                    method = getattr(self, 'end_' + tag)
+                except AttributeError:
+                    self.unknown_endtag(tag)
+                else:
+                    self.report_unbalanced(tag)
+                return
+            found = len(self.stack)
+            for i in range(found):
+                if self.stack[i] == tag: found = i
+        while len(self.stack) > found:
+            tag = self.stack[-1]
+            try:
+                method = getattr(self, 'end_' + tag)
+            except AttributeError:
+                method = None
+            if method:
+                self.handle_endtag(tag, method)
+            else:
+                self.unknown_endtag(tag)
+            del self.stack[-1]
+
+    # Overridable -- handle start tag
+    def handle_starttag(self, tag, method, attrs):
+        method(attrs)
+
+    # Overridable -- handle end tag
+    def handle_endtag(self, tag, method):
+        method()
+
+    # Example -- report an unbalanced </...> tag.
+    def report_unbalanced(self, tag):
+        if self.verbose:
+            print('*** Unbalanced </' + tag + '>')
+            print('*** Stack:', self.stack)
+
+    def convert_charref(self, name):
+        """Convert character reference, may be overridden."""
+        try:
+            n = int(name)
+        except ValueError:
+            return
+        if not 0 <= n <= 127:
+            return
+        return self.convert_codepoint(n)
+
+    def convert_codepoint(self, codepoint):
+        return chr(codepoint)
+
+    def handle_charref(self, name):
+        """Handle character reference, no need to override."""
+        replacement = self.convert_charref(name)
+        if replacement is None:
+            self.unknown_charref(name)
+        else:
+            self.handle_data(replacement)
+
+    # Definition of entities -- derived classes may override
+    entitydefs = \
+            {'lt': '<', 'gt': '>', 'amp': '&', 'quot': '"', 'apos': '\''}
+
+    def convert_entityref(self, name):
+        """Convert entity references.
+
+        As an alternative to overriding this method; one can tailor the
+        results by setting up the self.entitydefs mapping appropriately.
+        """
+        table = self.entitydefs
+        if name in table:
+            return table[name]
+        else:
+            return
+
+    def handle_entityref(self, name):
+        """Handle entity references, no need to override."""
+        replacement = self.convert_entityref(name)
+        if replacement is None:
+            self.unknown_entityref(name)
+        else:
+            self.handle_data(replacement)
+
+    # Example -- handle data, should be overridden
+    def handle_data(self, data):
+        pass
+
+    # Example -- handle comment, could be overridden
+    def handle_comment(self, data):
+        pass
+
+    # Example -- handle declaration, could be overridden
+    def handle_decl(self, decl):
+        pass
+
+    # Example -- handle processing instruction, could be overridden
+    def handle_pi(self, data):
+        pass
+
+    # To be overridden -- handlers for unknown objects
+    def unknown_starttag(self, tag, attrs): pass
+    def unknown_endtag(self, tag): pass
+    def unknown_charref(self, ref): pass
+    def unknown_entityref(self, ref): pass
+
+
+class TestSGMLParser(SGMLParser):
+
+    def __init__(self, verbose=0):
+        self.testdata = ""
+        SGMLParser.__init__(self, verbose)
+
+    def handle_data(self, data):
+        self.testdata = self.testdata + data
+        if len(repr(self.testdata)) >= 70:
+            self.flush()
+
+    def flush(self):
+        data = self.testdata
+        if data:
+            self.testdata = ""
+            print('data:', repr(data))
+
+    def handle_comment(self, data):
+        self.flush()
+        r = repr(data)
+        if len(r) > 68:
+            r = r[:32] + '...' + r[-32:]
+        print('comment:', r)
+
+    def unknown_starttag(self, tag, attrs):
+        self.flush()
+        if not attrs:
+            print('start tag: <' + tag + '>')
+        else:
+            print('start tag: <' + tag, end=' ')
+            for name, value in attrs:
+                print(name + '=' + '"' + value + '"', end=' ')
+            print('>')
+
+    def unknown_endtag(self, tag):
+        self.flush()
+        print('end tag: </' + tag + '>')
+
+    def unknown_entityref(self, ref):
+        self.flush()
+        print('*** unknown entity ref: &' + ref + ';')
+
+    def unknown_charref(self, ref):
+        self.flush()
+        print('*** unknown char ref: &#' + ref + ';')
+
+    def unknown_decl(self, data):
+        self.flush()
+        print('*** unknown decl: [' + data + ']')
+
+    def close(self):
+        SGMLParser.close(self)
+        self.flush()
+
+
+def test(args = None):
+    import sys
+
+    if args is None:
+        args = sys.argv[1:]
+
+    if args and args[0] == '-s':
+        args = args[1:]
+        klass = SGMLParser
+    else:
+        klass = TestSGMLParser
+
+    if args:
+        file = args[0]
+    else:
+        file = 'test.html'
+
+    if file == '-':
+        f = sys.stdin
+    else:
+        try:
+            f = open(file, 'r')
+        except IOError as msg:
+            print(file, ":", msg)
+            sys.exit(1)
+
+    data = f.read()
+    if f is not sys.stdin:
+        f.close()
+
+    x = klass()
+    for c in data:
+        x.feed(c)
+    x.close()
+
+
+if __name__ == '__main__':
+    test()

testgrey.py

@@ -8,6 +8,7 @@ class GreylistTestCase(unittest.TestCase):
 
   def setUp(self):
     self.fname = 'test.db'
+    if os.path.isfile(self.fname):
       os.remove(self.fname)
 
   def tearDown(self):

testmime.py

@@ -26,14 +26,21 @@
 # Revision 1.20  2004/11/20 16:38:17  stuart
 # Add rcs log
 #
+from __future__ import print_function
 import unittest
 import mime
 import socket
-import StringIO
+try:
+  from StringIO import StringIO
+except:
+  from io import StringIO
 import email
 import sys
 import Milter
-from email import Errors
+try:
+  from email import Errors as errors
+except:
+  from email import errors
 
 samp1_txt1 = """Dear Agent 1
 I hope you can read this.  Whenever you write label it  P.B.S kids.
@@ -46,48 +53,56 @@ class MimeTestCase(unittest.TestCase):
 
   # test mime parameter parsing
   def testParam(self):
-    plist = mime._parseparam(
+    plist = mime._parseparam('; boundary="----=_NextPart_000_4e56_490d_48e3"')
-      '; boundary="----=_NextPart_000_4e56_490d_48e3"')
+    plist = [ x for x in plist if x ] # py2 doesn't include empty params
-    self.failUnless(len(plist)==1)
+    self.assertEqual(1,len(plist))
-    self.failUnless(plist[0] == 'boundary="----=_NextPart_000_4e56_490d_48e3"')
+    self.assertTrue(plist[0] == 'boundary="----=_NextPart_000_4e56_490d_48e3"')
     plist = mime._parseparam('; name="Jim&Girlz.jpg"')
-    self.failUnless(len(plist)==1)
+    plist = [ x for x in plist if x ] # py2 doesn't include empty params
-    self.failUnless(plist[0] == 'name="Jim&Girlz.jpg"')
+    self.assertEqual(1,len(plist))
+    self.assertTrue(plist[0] == 'name="Jim&Girlz.jpg"')
 
   def testParse(self,fname='samp1'):
-    msg = mime.message_from_file(open('test/'+fname,"r"))
+    with open('test/'+fname,"rb") as fp:
-    self.failUnless(msg.ismultipart())
+      msg = mime.message_from_file(fp)
+    self.assertTrue(msg.ismultipart())
     parts = msg.get_payload()
-    self.failUnless(len(parts) == 2)
+    self.assertTrue(len(parts) == 2)
     txt1 = parts[0].get_payload()
-    self.failUnless(txt1.rstrip() == samp1_txt1,txt1)
+    self.assertTrue(txt1.rstrip() == samp1_txt1,txt1)
-    msg = mime.message_from_file(open('test/missingboundary',"r"))
+    with open('test/missingboundary',"rb") as fp:
+      msg = mime.message_from_file(fp)
     # should get no exception as long as we don't try to parse
     # message attachments
     mime.defang(msg,scan_rfc822=False)
-    msg.dump(open('test/missingboundary.out','w'))
+    with open('test/missingboundary.out','wb') as fp:
-    msg = mime.message_from_file(open('test/missingboundary',"r"))
+      msg.dump(fp)
+    with open('test/missingboundary',"rb") as fp:
+      msg = mime.message_from_file(fp)
     try:
       mime.defang(msg)
       # python 2.4 doesn't get exceptions on missing boundaries, and
       # if message is modified, output is readable by mail clients
       if sys.hexversion < 0x02040000:
         self.fail('should get boundary error parsing bad rfc822 attachment')
-    except Errors.BoundaryError:
+    except errors.BoundaryError:
       pass
   
   def testDefang(self,vname='virus1',part=1,
 	fname='LOVE-LETTER-FOR-YOU.TXT.vbs'):
-    msg = mime.message_from_file(open('test/'+vname,"r"))
+    with open('test/'+vname,"rb") as fp:
+      msg = mime.message_from_file(fp)
     mime.defang(msg,scan_zip=True)
-    self.failUnless(msg.ismodified(),"virus not removed")
+    self.assertTrue(msg.ismodified(),"virus not removed")
     oname = vname + '.out'
-    msg.dump(open('test/'+oname,"w"))
+    with open('test/'+oname,"wb") as fp:
-    msg = mime.message_from_file(open('test/'+oname,"r"))
+      msg.dump(fp)
+    with open('test/'+oname,"rb") as fp:
+      msg = mime.message_from_file(fp)
     txt2 = msg.get_payload()
     if type(txt2) == list:
       txt2 = txt2[part].get_payload()
-    self.failUnless(
+    self.assertTrue(
       txt2.rstrip()+'\n' == mime.virus_msg % (fname,hostname,None),txt2)
 
   def testDefang3(self):
@@ -103,51 +118,60 @@ class MimeTestCase(unittest.TestCase):
 
   # virus6 has no parts - the virus is directly inline
   def testDefang6(self,vname="virus6",fname='FAX20.exe'):
-    msg = mime.message_from_file(open('test/'+vname,"r"))
+    with open('test/'+vname,"rb") as fp:
+      msg = mime.message_from_file(fp)
     mime.defang(msg)
     oname = vname + '.out'
-    msg.dump(open('test/'+oname,"w"))
+    with open('test/'+oname,"wb") as fp:
-    msg = mime.message_from_file(open('test/'+oname,"r"))
+      msg.dump(fp)
-    self.failIf(msg.ismultipart())
+    with open('test/'+oname,"rb") as fp:
+      msg = mime.message_from_file(fp)
+    self.assertFalse(msg.ismultipart())
     txt2 = msg.get_payload()
-    self.failUnless(txt2 == mime.virus_msg % \
+    self.assertTrue(txt2 == mime.virus_msg % \
 	(fname,hostname,None),txt2)
 
   # honey virus has a sneaky ASP payload which is parsed correctly
   # by email package in python-2.2.2, but not by mime.MimeMessage or 2.2.1
   def testDefang7(self,vname="honey",fname='story[1].scr'):
-    msg = mime.message_from_file(open('test/'+vname,"r"))
+    with open('test/'+vname,"rb") as fp:
+      msg = mime.message_from_file(fp)
     mime.defang(msg)
     oname = vname + '.out'
-    msg.dump(open('test/'+oname,"w"))
+    with open('test/'+oname,"wb") as fp:
-    msg = mime.message_from_file(open('test/'+oname,"r"))
+      msg.dump(fp)
+    with open('test/'+oname,"rb") as fp:
+      msg = mime.message_from_file(fp)
     parts = msg.get_payload()
     txt2 = parts[1].get_payload()
     txt3 = parts[2].get_payload()
-    self.failUnless(txt2.rstrip()+'\n' == mime.virus_msg % \
+    self.assertTrue(txt2.rstrip()+'\n' == mime.virus_msg % \
 	(fname,hostname,None),txt2)
     if txt3 != '':
-      self.failUnless(txt3.rstrip()+'\n' == mime.virus_msg % \
+      self.assertTrue(txt3.rstrip()+'\n' == mime.virus_msg % \
 	  ('story[1].asp',hostname,None),txt3)
 
   def testParse2(self,fname="spam7"):
-    msg = mime.message_from_file(open('test/'+fname,"r"))
+    with open('test/'+fname,"rb") as fp:
-    self.failUnless(msg.ismultipart())
+      msg = mime.message_from_file(fp)
+    self.assertTrue(msg.ismultipart())
     parts = msg.get_payload()
-    self.failUnless(len(parts) == 2)
+    self.assertTrue(len(parts) == 2)
     name = parts[1].getname()
-    self.failUnless(name == "Jim&amp;amp;Girlz.jpg","name=%s"%name)
+    self.assertTrue(name == "Jim&amp;amp;Girlz.jpg","name=%s"%name)
 
   def testZip(self,vname="zip1",fname='zip.zip'):
     self.testDefang(vname,1,'zip.zip')
     # test scan_zip flag
-    msg = mime.message_from_file(open('test/'+vname,"r"))
+    with open('test/'+vname,"rb") as fp:
+      msg = mime.message_from_file(fp)
     mime.defang(msg,scan_zip=False)
-    self.failIf(msg.ismodified())
+    self.assertFalse(msg.ismodified())
     # test ignoring empty zip (often found in DSNs)
-    msg = mime.message_from_file(open('test/zip2','r'))
+    with open('test/zip2','rb') as fp:
+      msg = mime.message_from_file(fp)
     mime.defang(msg,scan_zip=True)
-    self.failIf(msg.ismodified())
+    self.assertFalse(msg.ismodified())
     # test corrupt zip (often an EXE named as a ZIP)
     self.testDefang('zip3',1,'zip.zip')
     # test zip within zip
@@ -164,22 +188,24 @@ class MimeTestCase(unittest.TestCase):
     mime.check_html(msg)
     # don't let a tricky virus slip one past us
     msg = msg.get_submsg()
-    if isinstance(msg,email.Message.Message):
+    if isinstance(msg,email.message.Message):
       return mime.check_attachments(msg,self._chk_attach)
     return Milter.CONTINUE
 
   def testCheckAttach(self,fname="test1"):
     # test1 contains a very long filename
-    msg = mime.message_from_file(open('test/'+fname,'r'))
+    with open('test/'+fname,'rb') as fp:
+      msg = mime.message_from_file(fp)
     mime.defang(msg,scan_zip=True)
-    self.failIf(msg.ismodified())
+    self.assertFalse(msg.ismodified())
-    msg = mime.message_from_file(open('test/test2','r'))
+    with open('test/test2','rb') as fp:
+      msg = mime.message_from_file(fp)
     rc = mime.check_attachments(msg,self._chk_attach)
-    self.assertEquals(self.filename,"7501'S FOR TWO GOLDEN SOURCES SHIPMENTS FOR TAX & DUTY PURPOSES ONLY.PDF")
+    self.assertEqual(self.filename,"7501'S FOR TWO GOLDEN SOURCES SHIPMENTS FOR TAX & DUTY PURPOSES ONLY.PDF")
-    self.assertEquals(rc,Milter.CONTINUE)
+    self.assertEqual(rc,Milter.CONTINUE)
 
   def testHTML(self,fname=""):
-    result = StringIO.StringIO()
+    result = StringIO()
     filter = mime.HTMLScriptFilter(result)
     msg = """<! Illegal declaration used as comment>
       <![if conditional]> Optional SGML <![endif]>
@@ -188,8 +214,10 @@ class MimeTestCase(unittest.TestCase):
     script = "<script lang=javascript> Dangerous script </script>"
     filter.feed(msg + script)
     filter.close()
-    #print result.getvalue()
+    #print(result.getvalue())
-    self.failUnless(result.getvalue() == msg + filter.msg)
+    #print('---')
+    #print(msg + filter.msg)
+    self.assertTrue(result.getvalue() == msg + filter.msg)
 
 def suite(): return unittest.makeSuite(MimeTestCase,'test')
 
@@ -198,7 +226,7 @@ if __name__ == '__main__':
     unittest.main()
   else:
     for fname in sys.argv[1:]:
-      fp = open(fname,'r')
+      with open(fname,'rb') as fp:
         msg = mime.message_from_file(fp)
       mime.defang(msg,scan_zip=True)
-      print msg.as_string()
+      print(msg.as_string())

testsample.py

@@ -2,8 +2,6 @@ import unittest
 import Milter
 import sample
 import mime
-import rfc822
-import StringIO
 from Milter.test import TestBase
 
 class TestMilter(TestBase,sample.sampleMilter):
@@ -15,13 +13,18 @@ class BMSMilterTestCase(unittest.TestCase):
 
   def testDefang(self,fname='virus1'):
     milter = TestMilter()
+    milter.setsymval('{auth_authen}','batman')
+    milter.setsymval('{auth_type}','batcomputer')
+    milter.setsymval('j','mailhost')
     rc = milter.connect()
     self.failUnless(rc == Milter.CONTINUE)
     rc = milter.feedMsg(fname)
+    self.failUnless(milter.user == 'batman',"getsymval failed")
+    self.failUnless(milter.auth_type != 'batcomputer',"setsymlist failed")
     self.failUnless(rc == Milter.ACCEPT)
     self.failUnless(milter._bodyreplaced,"Message body not replaced")
     fp = milter._body
-    open('test/'+fname+".tstout","w").write(fp.getvalue())
+    open('test/'+fname+".tstout","wb").write(fp.getvalue())
     #self.failUnless(fp.getvalue() == open("test/virus1.out","r").read())
     fp.seek(0)
     msg = mime.message_from_file(fp)
@@ -36,7 +39,7 @@ class BMSMilterTestCase(unittest.TestCase):
     self.failUnless(rc == Milter.ACCEPT)
     self.failIf(milter._bodyreplaced,"Milter needlessly replaced body.")
     fp = milter._body
-    open('test/'+fname+".tstout","w").write(fp.getvalue())
+    open('test/'+fname+".tstout","wb").write(fp.getvalue())
     milter.close()
 
   def testDefang2(self):
@@ -49,14 +52,14 @@ class BMSMilterTestCase(unittest.TestCase):
     self.failUnless(rc == Milter.ACCEPT)
     self.failUnless(milter._bodyreplaced,"Message body not replaced")
     fp = milter._body
-    open("test/virus3.tstout","w").write(fp.getvalue())
+    open("test/virus3.tstout","wb").write(fp.getvalue())
     #self.failUnless(fp.getvalue() == open("test/virus3.out","r").read())
     rc = milter.feedMsg("virus6")
     self.failUnless(rc == Milter.ACCEPT)
     self.failUnless(milter._bodyreplaced,"Message body not replaced")
     self.failUnless(milter._headerschanged,"Message headers not adjusted")
     fp = milter._body
-    open("test/virus6.tstout","w").write(fp.getvalue())
+    open("test/virus6.tstout","wb").write(fp.getvalue())
     milter.close()
 
 def suite(): return unittest.makeSuite(BMSMilterTestCase,'test')

testutils.py

@@ -1,9 +1,11 @@
+from __future__ import print_function
 import unittest
 import doctest
 import os
 import Milter.utils
 from Milter.cache import AddrCache
 from Milter.dynip import is_dynip
+from Milter.pyip6 import inet_ntop
 
 class AddrCacheTestCase(unittest.TestCase):
 
@@ -11,6 +13,7 @@ class AddrCacheTestCase(unittest.TestCase):
     self.fname = 'test.dat'
 
   def tearDown(self):
+    if os.path.exists(self.fname):
       os.remove(self.fname)
 
   def testAdd(self):
@@ -31,17 +34,27 @@ class AddrCacheTestCase(unittest.TestCase):
     self.failUnless(not cache['temp@bar.com'])
 
   def testDomain(self):
-    fp = open(self.fname,'w')
+    with open(self.fname,'w') as fp:
-    print >>fp,'spammer.com'
+      print('spammer.com',file=fp)
-    fp.close()
     cache = AddrCache(fname=self.fname)
     cache.load(self.fname,30)
     self.failUnless('spammer.com' in cache)
 
+  def testParseHeader(self):
+    s='=?UTF-8?B?TGFzdCBGZXcgQ29sZHBsYXkgQWxidW0gQXJ0d29ya3MgQXZhaWxhYmxlAA?='
+    h = Milter.utils.parse_header(s)
+    self.assertEqual(h,b'Last Few Coldplay Album Artworks Available\x00')
+
+  @unittest.expectedFailure
+  def testParseAddress(self):
+    s = Milter.utils.parseaddr('a(WRONG)@b')
+    self.assertEqual(s,('WRONG', 'a@b'))
+
 def suite(): 
   s = unittest.makeSuite(AddrCacheTestCase,'test')
   s.addTest(doctest.DocTestSuite(Milter.utils))
   s.addTest(doctest.DocTestSuite(Milter.dynip))
+  s.addTest(doctest.DocTestSuite(Milter.pyip6))
   return s
 
 if __name__ == '__main__':