epydoc
NAME
epydoc - generate API documentation from Python docstrings
SYNOPSIS
epydoc
[action]
[options]
B names...
DESCRIPTION
epydoc
generates API documentation for Python modules and packages, based on
their docstrings. A lightweight markup language called
epytext
can be used to format docstrings, and to add information about
specific fields, such as parameters and instance variables. Epydoc
also understands docstrings written in ReStructuredText, Javadoc, and
plaintext. Currently, epydoc supports two basic output formats: HTML
and LaTeX.
The HTML API documentation produced by
epydoc
consists of a set of HTML files, including: an API documentation page
for each class and module; a syntax-highlighted source code page for
each module; an identifier index page; a help page; and a frames-based
table of contents. When appropriate,
epydoc
will also generate
index pages for bugs, defined terms, and to-do items;
a class hierarchy page; and a package hierarchy page.
The LaTeX API documentation produced by
epydoc
consists of a main LaTeX file, and a LaTeX file for each module. If
you use
R --dvi ,
R --ps ,
or
--pdf ,
then
epydoc
will invoke external commands to convert the LaTeX output to the
requested format. Note that the LaTeX files containing the
documentation for individual modules can be included as chapters or
sections of other LaTeX documents, using the LaTeX
\\include
command. If you wish to include individual classes in other LaTeX
documents, then use the
--separate-classes
option to produce a separate LaTeX file for each class.
epydoc
can also be used to check the completeness of the API documentation.
By default, it checks that every public package, module, class,
method, and function has a docstring description. The
--tests
option can be used to specify additional tests to perform.
OPTIONS
Epydoc's options are divided into six categories: basic options,
actions, generation options, output options, graph options, and return
value options.
BASIC OPTIONS
R names ...
The list of objects that should be documented. Objects can be
specified using Python dotted names (such as
R os.path ),
filenames (such as
R epydoc/epytext.py ),
or directory names (such as
R epydoc/ ).
Directory names specify packages, and are expanded to include
all sub-modules and sub-packages. If you wish to exclude
certain sub-modules or sub-packages, use the
--exclude
option (described below).
I --config file
A configuration file, specifying additional
R options and/or names .
This option may be repeated.
--q, --quiet, --v, --verbose
Produce quite (or verbose) output. If used multiple times, this
option produces successively more quiet (or verbose) output.
--debug
Show full tracebacks for internal errors.
--simple-term
Do not try to use color or cursor control when displaying the progress
bar, warnings, or errors.
ACTIONS
--html
Write HTML output.
[default]
--latex
Write LaTeX output.
--dvi
Write DVI output.
--ps
Write Postscript output.
--pdf
Write Adobe Acrobat (pdf) output.
--check
Perform completeness checks on the documentation.
--pickle
Write the documentation to a pickle file.
GENERATION OPTIONS
I --docformat format
Set the default value for
__docformat__
to
R format .
__docformat__
is a module variable that specifies the markup language for the
docstrings in a module. Its value consists of the name of a markup
language, optionally followed by a language code (such as
en
for English). For a list of the markup languages currently recognized
by epydoc, run
R epydoc --help docformat .
I --parse-only
Gather all information about the documented objects by parsing the
relevant Python source code; in particular, do
not
use introspection to gather information about the documented objects.
This option should be used when epydoc is run on untrusted code; or on
code that can not be introspected because of missing dependencies, or
because importing it would cause undesired side-effects.
I --introspect-only
Gather all information about the documented objects by introspection;
in particular, do
not
gather information by parsing the object's Python source code.
I --exclude PATTERN
Do not document any object whose name matches the given regular
expression pattern.
I --exclude-introspect PATTERN
Do not use introspection to gather information about any object whose
name matches the given regular expression.
I --exclude-parse PATTERN
Do not use Python source code parsing to gather information about any
object whose name matches the given regular expression.
I --inheritance format
The format that should be used to display inherited methods,
variables, and properties in the generated "summary" tables.
If
format
is "grouped," then inherited objects are gathered into groups, based
on which class that they are inherited from. If
format
is "listed," then inherited objects are listed in a short list at the
end of the summary table. If
format
is "included," then inherited objects are mixed in with non-inherited
objects. The default format for HTML output is "grouped."
--show-private, --no-private
These options control whether documentation is generated for private
objects. By default, the generated documentation includes private
objects, and users can choose whether to view private objects or not,
by clicking on "show private" and "hide private" links. But if you
want to discourage users from directly accessing private objects, then
you may prefer not to generate documentation for private objects.
--show-imports, --no-imports
These options control whether module imports are included in the
generated documentation. By default, imports are not included.
--show-sourcecode, --no-sourcecode
These options control whether or not epydoc should generate
syntax-highlighted pages containing the souce code of each module in
the HTML output. By default, the sourcecode pages are generated.
--include-log
Generate an HTML page
epydoc-log.html
containing all error and warning messages that are generated by
epydoc, and include it in the generated output.
OUTPUT OPTIONS
I -o dir , --output dir
The output directory. If
dir
does not exist, then it will be created. If no output directory is
specified, then the action name (e.g.,
R html or pdf ).
html
I -c sheet , --css sheet
CSS stylesheet for HTML output files. If
sheet
is a file, then the stylesheet is copied from that file; otherwise,
sheet
is taken to be the name of a built-in stylesheet. For a list of
the built-in stylesheets, run
R epydoc --help css .
If a CSS stylesheet is not specified, then the default stylesheet is
used.
I --name name
The name of the project whose documentation is being generated.
I -u url , --url url
The URL of the project's homepage.
I --navlink html
HTML code for the homepage link on the HTML navigation bar. If this
HTML code contains any hyperlinks
(<a href=...>),
then it will be inserted verbatim. If
it does not contain any hyperlinks, and a project url is specified
(with
R --url ),
then a hyperlink to the specified URL is added to the link.
I --help-file file
An alternate help file.
file
should contain the body of an HTML file -- navigation bars will be
added to it.
--show-frames, --no-frames
These options control whether HMTL output will include a frames-base
table of contents page. By default, the frames-based table of
contents is included.
--separate-classes
In the LaTeX output, describe each class in a separate section of the
documentation, instead of including them in the documentation for
their modules. This creates a separate LaTeX file for each class, so
it can also be useful if you want to include the documentation for one
or two classes as sections of your own LaTeX document.
GRAPH OPTIONS
I --graph graphtype
Include graphs of type
graphtype
in the generated output. Graphs are generated using the Graphviz dot
executable. If this executable is not on the path, then use
--dotpath
to specify its location. This option may be repeated to include
multiple graph types in the output.
graphtype
should be one of:
R all , classtree , callgraph , or umlclasstree .
I --dotpath path
The path to the Graphviz
R dot
executable.
I --graph-font font
The name of the font used to generate Graphviz graphs. (e.g.,
helvetica or times).
I --graph-font-size size
The size of the font used to generate Graphviz graphs, in points.
I --pstat file
A pstat output file, to be used in generating call graphs.
RETURN VALUE OPTIONS
--fail-on-error
Return a non-zero exit status, indicating failure, if any errors are
encountered.
--fail-on-warning
Return a non-zero exit status, indicating failure, if any errors or
warnings are encountered (not including docstring warnings).
--fail-on-docstring-warning
Return a non-zero exit status, indicating failure, if any errors or
warnings are encountered (including docstring warnings).
HTML FILES
The HTML API documentation produced by
epydoc
consists of the following files:
OBJECT DOCUMENTATION PAGES
index.html
The standard entry point for the documentation. Normally,
index.html
is a copy of the frames file
(frames.html).
But if the
--no-frames
option is used, then
index.html
is a copy of the API documentation home page, which is normally the
documentation page for the top-level package or module (or the trees
page if there is no top-level package or module).
B module -module.html
The API documentation for a module.
module
is the complete dotted name of the module, such as
sys
or
R epydoc.epytext .
B class -class.html
The API documentation for a class, exception, or type.
class
is the complete dotted name of the class, such as
epydoc.epytext.Token
or
R array.ArrayType .
B module -pysrc.html
A syntax-highlighted page containing the Python source code for
R module .
This page includes links back to the API documentation pages.
module-tree.html
The module hierarchy.
class-tree.html
The class hierarchy. This page is only generated if at least one
class is documented.
INDICES
identifier-index.html
An index of all documented identifiers. If the identifier index
contains more than 3,000 entries, then it will be split into separate
pages for each letter, named
R identifier-index-a.html ,
R identifier-index-b.html , etc.
term-index.html
An index of all explicitly marked definitional terms. This page is
only generated if at least one definition term is marked in a
formatted docstring.
bug-index.html
An index of all explicitly marked
@bug
fields. This page is only
generated if at least one
@bug
field is listed in a formatted docstring.
todo-index.html
An index of all explicitly marked
@todo
fields. This page is only
generated if at least one
@todo
field is listed in a formatted docstring.
changed-index.html
An index of all explicitly marked
@changed
fields. This page is only
generated if at least one
@changed
field is listed in a formatted docstring.
deprecated-index.html
An index of all explicitly marked
@deprecated
fields. This page is only
generated if at least one
@deprecated
field is listed in a formatted docstring.
since-index.html
An index of all explicitly marked
@since
fields. This page is only
generated if at least one
@since
field is listed in a formatted docstring.
FRAMES-BASED TABLE OF CONTENTS
frames.html
The main frames file. Two frames on the left side of the window
contain a table of contents, and the main frame on the right side of
the window contains API documentation pages.
toc.html
The top-level table of contents page. This page is displayed in the
upper-left frame of
R frames.html ,
and provides links to the
toc-everything.html
and
I toc- module -module.html
pages.
toc-everything.html
The table of contents for the entire project. This page is displayed
in the lower-left frame of
R frames.html ,
and provides links to every class, type, exception, function, and
variable defined by the project.
I toc- module -module.html
The table of contents for a module. This page is displayed in the
lower-left frame of
R frames.html ,
and provides links to every class, type, exception, function, and
variable defined by the module.
module
is the complete dotted name of the module, such as
sys
or
R epydoc.epytext .
OTHER PAGES
help.html
The help page for the project. This page explains how to use and
navigate the webpage produced by epydoc.
redirect.html
This page uses javascript to translate dotted names to their
corresponding URLs. For example, in epydoc's documentation,
loading the page
<redirect.html#epydoc.apidoc.DottedName>
will automatically redirect the browser to
R <epydoc.apidoc-module.html#DottedName> .
epydoc.css
The CSS stylesheet used to display all HTML pages.
epydoc.js
A javascript file used to define javascript functions used by epydoc.
epydoc-log.html
A page containing a log of all warnings and errors that were generated
by epydoc, along with a table listing all of the options that were
used.
LATEX FILES
The LaTeX API documentation produced by
epydoc
consists of the following files:
api.pdf
An Adobe Acrobat (pdf) file containing the complete API documentation.
This file is only generated if you use the
--pdf
option.
api.tex
The top-level LaTeX file. This file imports the other LaTeX files, to
create a single unified document.
api.dvi
A dvi file containing the complete API documentation. This file is
only generated if you use the
--dvi
option, the
--ps
option, or the
--pdf
option.
api.ps
A postscript file containing the complete API documentation. This
file is only generated if you use the
--ps
option or the
--pdf
option.
B module -module.tex
The API documentation for a module.
module
is the complete dotted name of the module, such as
sys or
R epydoc.epytext .
B class -class.tex
The API documentation for a class, exception, or type.
class
is the complete dotted name of the class, such as
epydoc.epytext.Token
or array.ArrayType. These class documentation files are only created
if the
--separate-classes
option is used; otherwise, the documentation for each class is
included in its module's documentation file.
DIAGNOSTICS
EPYTEXT MARKUP WARNING MESSAGES
Epytext errors are caused by epytext docstrings that contain invalid
markup. Whenever an epytext error is detected, the docstring in
question is treated as a plaintext docstring. Epydoc can generate the
following epytext errors:
Bad link target.
The target specified for an inline link contruction
(L{...})
is not well-formed. Link targets must be valid python identifiers.
Bad uri target.
The target specified for an inline uri contruction
(U{...})
is not well-formed. This typically occurs if inline markup is nested
inside the URI target.
Fields must be at the top level.
The list of fields
(@param, etc.)
is contained by some other
block structure (such as a list or a section).
Fields must be the final elements.
The list of fields
(@param, etc.)
is not at the end of a docstring.
Headings must occur at top level.
The heading is contianed in some other block structure (such as a
list).
Improper doctest block indentation.
The doctest block dedents past the indentation of its initial prompt
line.
Improper heading indentation.
The heading for a section is not left-aligned with the paragraphs in
the section that contains it.
Improper paragraph indentation.
The paragraphs within a block are not left-aligned. This error is
often generated when plaintext docstrings are parsed using epytext.
Invalid escape.
An unknown escape sequence was used with the inline escape construction
(E{...}).
Lists must be indented.
An unindented line immediately following a paragraph starts with a
list bullet. Epydoc is not sure whether you meant to start a new list
item, or meant for a paragraph to include a word that looks like a
bullet. If you intended the former, then indent the list. If you
intended the latter, then change the word-wrapping of the paragraph,
or escape the first character of the word that looks like a bullet.
Unbalanced '{'.
The docstring contains unbalanced braces. Epytext requires that all
braces must be balanced. To include a single unbalanced brace, use
the escape sequences E{lb} (left brace) and E{rb} (right brace).
Unbalanced '}'.
The docstring contains unbalanced braces. Epytext requires that all
braces must be balanced. To include a single unbalanced brace, use
the escape sequences E{lb} (left brace) and E{rb} (right brace).
Unknown inline markup tag.
An unknown tag was used with the inline markup construction (
B x {...}
).
Wrong underline character for heading.
The underline character used for this section heading does not
indicate an appopriate section level. The "=" character should be
used to underline sections; "-" for subsections; and "~" for
subsubsections.
Possible mal-formatted field item.
Epytext detected a line that looks like a field item, but is not
correctly formatted. This typically occurs when the trailing colon
(":") is not included in the field tag.
Possible heading typo.
Epytext detected a pair of lines that looks like a heading, but the
number of underline characters does not match the number of characters
in the heading. The number of characters in these two lines must
match exactly for them to be considered a heading.
FIELD WARNINGS
Field warnings are caused by docstrings containing invalid fields.
The contents of the invalid field are generally ignored. Epydoc can
generate the following field warnings:
I @param for unknown parameter param .
A @param field was used to specify the type for a parameter that is
not included in the function's signature. This is typically caused by
a typo in the parameter name.
B tag did not expect an argument.
The field tag
tag
was used with an argument, but it does not take one.
B tag expected an argument.
The field tag
tag
was used without an argument, but it requires one.
I @type for unknown parameter param .
A @type field was used to specify the type for a parameter that is not
included in the function's signature. This is typically
caused by a typo in the parameter name.
I @type for unknown variable var .
A @type field was used to specify the type for a variable, but no
other information is known about the variable. This is typically
caused by a typo in the variable name.
I Unknown field tag tag .
A docstring contains a field with the unknown tag
R tag .
I Redefinition of field .
Multiple field tags define the value of
field
in the same docstring, but
field
can only take a single value.
EXAMPLES
R epydoc -n epydoc -u http://epydoc.sf.net epydoc/
Generate the HTML API documentation for the epydoc package and all of
its submodules, and write the output to the
html
directory. In the headers and footers, use
epydoc
as the project name, and
http://epydoc.sf.net
as the project URL.
R epydoc --pdf -n epydoc epydoc/
Generate the LaTeX API documentation for the epydoc package and all of
its submodules, and write the output to the
latex
directory.
EXIT STATUS
0
Successful program execution.
1
Usage error.
2
Epydoc generated an error or warning, and one of the options
I --fail-on-error ,
I --fail-on-warning , or
--fail-on-docstring-warning
was specified.
other
Internal error (Python exception).
AUTHOR
Epydoc was written by Edward Loper. This man page was originally
written by Moshe Zadka, and is currently maintained by Edward Loper.
Debian-specific modifications were made by Kenneth J. Pronovici.
BUGS
Report bugs to <edloper@users.sourceforge.net>.
SEE ALSO
epydocgui(1)
The epydoc webpage
<http://epydoc.sourceforge.net>
The epytext markup language manual
<http://epydoc.sourceforge.net/epytext.html>
ČLÁNKY DO MAILU