Package epydoc :: Package docwriter :: Module xlink :: Class DocUrlGenerator

[frames] | no frames]

Class DocUrlGenerator

source code

Read a documentation index and generate URL's for it.

Nested Classes

[hide private]

Inherited from UrlGenerator: IndexAmbiguous

Instance Methods

[hide private]

__init__(self)

source code

str

get_url(self, name)
Look for a name and return the matching URL documentation.

source code

Inherited from UrlGenerator: get_canonical_name

Content loading

clear(self)
Clear the current class content.

source code

load_index(self, f)
Read the content of an index file.

source code

_iter_tuples(self, f)
Iterate on a file returning 2-tuples.

source code

load_records(self, records)
Read a sequence of pairs name -> url and populate the internal maps.

source code

Class Variables

[hide private]

Inherited from UrlGenerator (private): _SEP_RE

Instance Variables

[hide private]

_exact_matches
A map from an object fully qualified name to its URL.

_partial_names
A map from partial names to the fully qualified names they may refer.

prefix
Prefix portion for the URL's returned by get_url().

_filename
Not very important: only for logging.

Method Details

[hide private]

get_url(self, name)

source code

Look for a name and return the matching URL documentation.

First look for a fully qualified name. If not found, try with partial name.

If no url exists for the given object, return None.

Parameters:

name - the name to look for

Returns: str

the URL that can be used to reach the name documentation. None if no such URL exists.

Raises:

IndexError - no object found with name
DocUrlGenerator.IndexAmbiguous - more than one object found with a non-fully qualified name; notice that this is an IndexError subclass

Overrides: UrlGenerator.get_url

(inherited documentation)

load_index(self, f)

source code

Read the content of an index file.

Populate the internal maps with the file content using load_records().

Parameters:

f (str or file) - a file name or file-like object fron which read the index.

load_records(self, records)

source code

Read a sequence of pairs name -> url and populate the internal maps.

Parameters:

records (iterable) - the sequence of pairs (name, url) to add to the maps.

Instance Variable Details

[hide private]

_exact_matches

A map from an object fully qualified name to its URL.

Values are both the name as tuple of fragments and as read from the records (see load_records()), mostly to help _partial_names to perform lookup for unambiguous names.

_partial_names

A map from partial names to the fully qualified names they may refer.

The keys are the possible left sub-tuples of fully qualified names, the values are list of strings as provided by the index.

If the list for a given tuple contains a single item, the partial match is not ambuguous. In this case the string can be looked up in _exact_matches.

If the name fragment is ambiguous, a warning may be issued to the user. The items can be used to provide an informative message to the user, to help him qualifying the name in a unambiguous manner.