commonPaths(paths)

Returns the common component and the stripped paths

It expects that directories do always end with a trailing slash and paths never begin with a slash (except root).

Parameters:

paths - The list of paths ([str, str, ...])
           (type=list)

Returns:

The common component (always a directory) and the stripped paths ((str, [str, str, ...]))
           (type=tuple)

extractX509User(author)

Parameters:

author - The author string
           (type=basestring)

Returns:

user name, mail address (user name maybe None)
           (type=tuple or None)

filterForXml(value)

Parameters:

value - The value to filter
           (type=unicode)

Returns:

The filtered value
           (type=unicode)

getGlobValue(globs, path)

Parameters:

globs - The glob list ([(glob, associated value)])
           (type=sequence)

path - The path to match
           (type=str)

Returns:

The matched value or None
           (type=any)

getParentDirList(path)

Parameters:

path - The path to process
           (type=str)

Returns:

The directory list
           (type=list)

inherit(cls, *bases)

Parameters:

cls - The class to inherit from bases
           (type=class)

bases - The base class(es)
           (type=tuple)

Note: cls needs a __dict__, so __slots__ is a nono

loadDotted(name)

Loads a dotted name

The dotted name can be anything, which is passively resolvable (i.e. without the invocation of a class to get their attributes or the like). For example, name could be 'svnmailer.util.loadDotted' and would return this very function. It's assumed that the first part of the name is always is a module.

If a dotted name was loaded successfully, the object will be cached and delivered from there the next time.

Parameters:

name - The dotted name to load
           (type=str)

Returns:

The loaded object
           (type=any)

Raises:

ImportError - A module in the path could not be loaded

splitCommand(command)

Split a command string with respect to quotes and such

The command string consists of several tokens:

• whitespace: Those are separators except inside quoted items
• unquoted items: every token that doesn't start with a double quote (")
• quoted items: every token that starts with a double quote ("). Those items must be closed with a double quote and may contain whitespaces. The enclosing quotes are stripped. To put a double quote character inside such a token, it has to be escaped with a backslash (\). Therefore - backslashes themselves have to be escaped as well. The escapes are also stripped from the result.

Here's an example: r'foo bar "baz" "zo\"" "\\nk"' resolves to ['foo', 'bar', 'baz', 'zo"', r'\nk']

Parameters:

command - The command string
           (type=str)

Returns:

The splitted command
           (type=list)

Raises:

ValueError - The command string is not valid (unclosed quote or the like)

substitute(template, subst)

Returns a filled template

If the template is None, this function returns None as well.

Parameters:

template - The temlate to fill
           (type=basestring)

subst - The substitution parameters
           (type=dict)

Returns:

The filled template (The return type depends on the template and the parameters)
           (type=basestring)

parseContentType(value)

Parses a content type

(the email module unfortunately doesn't provide a public interface for this)

Parameters:

value - The value to parse - must be ascii compatible
           (type=basestring)

Returns:

The parsed header ((value, {key, [value, value, ...]})) or None
           (type=tuple)

Warning: comments are not recognized (yet?)