Rechercher une page de manuel

Chercher une autre page de manuel:


Langue: en

Autres versions - même langue

Version: April 2009 (ubuntu - 07/07/09)

Section: 1 (Commandes utilisateur)


html2epub - part of calibre


html2epub [options] file.html|opf


Convert a HTML file to an EPUB ebook. Recursively follows links in the HTML file. If you specify an OPF file instead of an HTML file, the list of links is takes from the <spine> element of the OPF file.

Whenever you pass arguments to html2epub that have spaces in them, enclose the arguments in quotation marks.


show program's version number and exit
-h, --help
show this help message and exit
Character encoding for HTML files. Default is to auto detect.
-o OUTPUT, --output=OUTPUT
The output EPUB file. If not specified, it is derived from the input file name.
Profile of the target device this EPUB is meant for. Set to None to create a device independent EPUB. The profile is used for device specific restrictions on the EPUB. Choices are: ['PRS505', 'None']
Either the path to a CSS stylesheet or raw CSS. This CSS will override any existing CSS declarations in the source files.


Control page layout
Set the top margin in pts. Default is 5.0
Set the bottom margin in pts. Default is 5.0
Set the left margin in pts. Default is 5.0
Set the right margin in pts. Default is 5.0
The base font size in pts. Default is 12.0pt. Set to 0 to disable rescaling of fonts.
Remove spacing between paragraphs. Also sets a indent on paragraphs of 1.5em. You can override this by adding p {text-indent: 0cm} to --override-css. Spacing removal will not work if the source file forces interparagraph spacing.
Do not force text to be justified in output.
Remove table markup, converting it into paragraphs. This is useful if your source file uses a table to manage layout.
Preserve the HTML tag structure while splitting large HTML files. This is only neccessary if the HTML files contain CSS that uses sibling selectors. Enabling this greatly slows down processing of large HTML files.


Control the following of links in HTML files.
Traverse links in HTML files breadth first. Normally, they are traversed depth first
Maximum levels of recursion when following links in HTML files. Must be non-negative. 0 implies that no links in the root HTML file are followed.


Control the automatic generation of a Table of Contents. If an OPF file is detected and it specifies a Table of Contents, then that will be used rather than trying to auto-generate a Table of Contents.
Maximum number of links to insert into the TOC. Set to 0 to disable. Default is: 50. Links are only added to the TOC if less than the --toc-threshold number of chapters were detected.
Don't add auto-detected chapters to the Table of Contents.
If fewer than this number of chapters is detected, then links are added to the Table of Contents. Default: 6
XPath expression that specifies all tags that should be added to the Table of Contents at level one. If this is specified, it takes precedence over other forms of auto-detection.
XPath expression that specifies all tags that should be added to the Table of Contents at level two. Each entry is added under the previous level one entry.
XPath expression that specifies all tags that should be added to the Table of Contents at level three. Each entry is added under the previous level two entry.
Path to a .ncx file that contains the table of contents to use for this ebook. The NCX file should contain links relative to the directory it is placed in. See 5.html#NCX for an overview of the NCX format.
Normally, if the source file already has a Table of Contents, it is used in preference to the autodetected one. With this option, the autodetected one is always used.


Options useful for debugging
-v, --verbose
Be more verbose while processing. Can be specified multiple times to increase verbosity.
Output HTML is "pretty printed" for easier parsing by humans
Print generated OPF file to stdout
Print generated NCX file to stdout
Keep intermediate files during processing by html2epub
Extract the contents of the produced EPUB file to the specified directory.


Control auto-detection of document structure.
An XPath expression to detect chapter titles. The default is to consider <h1> or <h2> tags that contain the words "chapter","book","section" or "part" as chapter titles as well as any tags that have class="chapter". The expression used must evaluate to a list of elements. To disable chapter detection, use the expression "/". See the XPath Tutorial in the calibre User Manual for further help on using this feature.
Specify how to mark detected chapters. A value of "pagebreak" will insert page breaks before chapters. A value of "rule" will insert a line before chapters. A value of "none" will disable chapter marking and a value of "both" will use both page breaks and lines to mark chapters.
Path to the cover to be used for this book
Use the cover detected from the source file in preference to the specified cover.
Remove the first image from the input ebook. Useful if the first image in the source file is a cover and you are specifying an external cover.
Turn off splitting at page breaks. Normally, input files are automatically split at every page break into two files. This gives an output ebook that can be parsed faster and with less resources. However, splitting is slow and if your source file contains a very large number of page breaks, you should turn off splitting on page breaks.
XPath expression to detect page boundaries for building a custom pagination map, as used by AdobeDE. Default is not to build an explicit pagination map.
XPath expression to find the name of each page in the pagination map relative to its boundary element. Default is to number all pages staring with 1.


Set metadata of the generated ebook
-t TITLE, --title=TITLE
Set the title. Default is to autodetect.
-a AUTHORS, --authors=AUTHORS
The author(s) of the ebook, as a & separated list.
The subject(s) of this book, as a comma separated list.
Set the publisher of this book.
A summary of this book.
Load metadata from the specified OPF file

Created by Kovid Goyal <>

|Le cadre pointillé qui entoure ce texte représ|
|ente les bords d'une grande carte à jouer sur |
|laquelle se tient une mouche hyper-intelligent|
|e ; cette dernière, en effet, est capable de s|
|e rendre immédiatement d'un point quelconque d|
|u recto de la carte à un point quelconque du v|
|erso par le chemin le plus court. Si l'on plaç|
|ait la mouche sur le dixième p de ce texte, qu|
|el serait le point de la carte le plus éloigné|
| d'elle ? Et par quelle lettre de ce texte (ou|
| signe, ou espace) serait recouvert ce point ?|
-- Angelini, Éric