weblog Simple blog publisher. It reads structured text files and generates static HTML / RSS files. Weblog aims to be simple and robust. Weblog manual ============= :Author: Henry Pr?cheur <henry@precheur.org> :Reviewers: Anis Kadri, Bastien Simondi, Eric Salama Abstract -------- Simple blog publisher. It reads structured text files and generates static HTML / RSS files. Weblog aims to be simple and robust. In this document *Weblog* is the name of the software. The *web log* concept is referred as the more common term *blog*. According to Wikipedia_: A *blog* (a portmanteau of *web log*) is a website where entries are written in chronological order and commonly displayed in reverse chronological order. .. _Wikipedia: http://en.wikipedia.org/wiki/Blog Pre-requirements ~~~~~~~~~~~~~~~~ - Python version 2.5+ - Jinja version 1.1+. Learn how to install Jinja at http://jinja.pocoo.org/documentation/installation. Installation ------------ Download Weblog's latest version at http://henry.precheur.org/weblog/. Extract it:: tar zxf weblog.tar.gz It can be used right away using the helper script ``weblog_run.py``. Or install it using the supplied ``setup.py`` script. Run ``python setup.py --help`` to learn how to use it. Alternatively if easy_install is present, simply type:: easy_install weblog It fetches the latest version of Weblog and installs it. Quick Start ----------- In the following examples ``weblog/`` represents Weblog's installation directory. If you downloaded the source tarball without installing Weblog; Use the helper script ``weblog_run.py`` instead of the ``weblog`` command:: $ python /path/to/weblog/weblog_run.py --help Create a new directory named ``my_blog``. The $ sign represents the shell prompt, do not type it!:: $ mkdir my_blog Copy from the Weblog installation directory the file ``weblog.ini`` into ``my_blog``:: $ cp weblog/examples/weblog.ini my_blog ``weblog.ini`` is the configuration file of the blog. Check the configuration file section for more information. Do not worry about it now, no modification is required to get the following examples working. Create a file named ``first_post.html`` in the ``my_blog`` directory:: title: First post author: Me date: 2007-08-25 Hello world! Actually all the post filenames must end with ``.html``. Go in the ``my_blog`` directory and run the Weblog using the publish command:: $ cd my_blog/ $ weblog publish It should create a directory named ``output`` containing the generated files. Look at the results by opening the file ``output/index.html`` in your web-browser. The first 3 lines of the file ``first_post.html`` define the post's parameters. These are standard :RFC:`2822` headers (the headers used in Emails). Only ``title`` is mandatory. ``date`` and ``author`` are optional. If you don't fill these fields, the author is the one specified in ``weblog.ini``, and the post's date is the post file's last modification date. The line ``Hello world!`` is the actual content of the post. Note that a blank line is required between the headers and the content. The content is an HTML block. Use the HTML syntax to format your post content. For example create a second file named ``second_post.html``:: title: Second post author: Me (again!) date: 2007-08-26 Second <q>test</q> post! &copy 2007 Me Regenerate the blog files:: $ weblog publish Reload the page in your browser. You should see a second post with some formating. The default post file encoding is ASCII. To use a different encoding specify it via the field ``encoding``:: title: Encoding test date: 2007-11-5 encoding: latin-1 Here you can put some ISO-8856-1 text ... Specify the default encoding in ``weblog.ini``, to avoid setting the encoding field for every file. While writing your blog post, don't bother about the ``date`` field immediately. Weblog automatically sets the date to the filename's last modification time. A good practice though is to set the date when the post gets published. By doing so the date won't get changed if the file gets copied. To set the date of a post, use the command ``date``:: $ date Mon Apr 14 00:10:44 PDT 2008 $ cat my_blog_post.html title: My blog post This is a blog post without any date. $ weblog date my_blog_post.html Setting date to 2008-04-14 00:12:22 in file my_blog_post.html $ cat my_blog_post.html title: My blog post date: 2008-04-14 00:12:22 This is a blog post without any date. $ weblog date my_blog_post 2008-5-15 Setting date to 2008-05-15 in file my_blog_post.html $ cat my_blog_post.html title: My blog post date: 2008-05-15 This is a blog post without any date. The ``date`` command accepts 3 formats as argument: - YEAR-MONTH-DAY (2008-01-31) - YEAR-MONTH-DAY HOUR:MINUTE (2008-01-31 16:45) - YEAR-MONTH-DAY HOUR:MINUTE:SECONDS (2008-01-31 16:45:14) For conciseness the ``date`` command uses aliases to specify commonly used date: - now - today (like now but only set the date, not the time) - tomorrow (now + 24 hours) - next_day (like tomorrow but only sets the date, not the time) Encoding and escaping --------------------- Weblog tries to make sure its output is always *correct*. Non-ASCII characters, are converted to HTML entities so you don't have to worry about it. The output is *never* encoded into ISO-8856-1, UTF-8 or another non-ASCII encoding. Encoding conversions are not so simple in practice. By doing only one conversion to the simplest encoding possible, a lot of problems are solved. The content of the post is not escaped. The title and the date of the post are escaped. The title ``Hello World`` is escaped. HTML tags appear, and no formating is applied to ``world``. The original text "Hello World" appears instead of "Hello *World*", It is possible to override this by specifying ``raw`` as the encoding. Using the ``raw`` encoding nothing is escaped or converted, but you must make sure all characters are ASCII characters:: title: Non-escaped title author: <q>Me</q> <me@my_weblog.org> encoding: raw If the ``raw`` encoding is used, all the characters must be ASCII characters. Otherwise an error is reported. Attaching a file to a post -------------------------- To attach files like images to a blog post, use the field ``files``:: title: Attach a file files: picture.png directory/file <img src='picture.png' alt='a picture'> <a href='directory/file'>a file</a> It will copy ``picture.png`` and ``directory/file``. If ``directory`` does not exist, it will be created. How URI's are handled --------------------- Relative links (``<a href='test.html'>``) are rewritten in the RSS file and in some HTML files. In the RSS file ``base_url`` is prepended to the link to make sure it always points to the correct URI. Absolute links (``<a href='http://example.com'>``) are not rewritten. It should always point to the correct location regardless of the context. Note that Weblog considers ``/`` as the root directory. If ``base_url`` is ``http://example.com/``; ``test.html`` and ``/test.html`` are both rewritten to ``http://example.com/test.html``. Command line parameters ----------------------- Usage: weblog [options] Options: -h, --help show this help message and exit -s DIR, --source-dir=DIR The source directory where the blog posts and the file weblog.ini are located -o DIR, --output-dir=DIR The directory where all the generated files are written. If it does not exist it is created. -q, --quiet Do not output anything except critical error messages Configuration file ------------------ All configuration options are in the ``weblog`` section. Learn more about the format of the configuration file: http://docs.python.org/lib/module-ConfigParser.html. A sample configuration file:: [weblog] title: Blog's title url: http://example.com/ description: A sample blog. source_dir: path/to/my/posts output_dir: path/to/output/directory encoding: latin-1 author: Me <me@example.com> Fields description ~~~~~~~~~~~~~~~~~~ title The blog's title. It appears at the top of the homepage and in the page's title. This field is mandatory. url The base URL of your blog. For example ``http://my-host.com/my-weblog/``. It is used to generate the absolute URL's to your blog. This field is mandatory. description A short description of your blog. Like "My favorite books reviews", or "Dr. Spock, publications about electronics". Note that it is possible to use multiple lines:: description: My blog about configuration files. The description is merged to a single line; ``My blog about configuration files.``. This field is mandatory. source_dir The directory containing the file ``weblog.ini``, the post files and possibly the ``templates`` directory. By default the current directory. output_dir The output directory. Generated files are put there. By default ``output``. encoding The default post file encoding. Default ``ASCII``. It is overridden by the ``encoding`` field in the post file. author The default author. It is overridden by the ``author`` field in the post file. post_per_page The number of post displayed per listing page. Default is 10. rss_limit The maximum number of post to be included in the RSS file. The most recent posts are the ones included. Default is 10. html_head Additional information for the ``<head>`` section. Useful to add custom CSS style sheets. Can be a string or a filename. If a file with this name exists in the source directory then it is read. Else it is considered as a string. The result is processed using Jinja. Use the variable ``top_dir`` to link to external files. It contains the path to the top directory of the blog. Examples:: html_head=<style type='text/css'>body { font-family: sans-serif; }</style> html_head={{ top_dir }}my_stylesheet.css html_header Additional content located just before the blog content. Can be a string or a filename. (See html_head above) Useful to add a logo or a search box at the top. html_footer Additional content located just after the blog content. Can be a string or a filename. (See html_head above) Useful to add ... A footer! Tips on Uploading ----------------- rsync_ is a useful tool to upload files generated by Weblog. To make sure rsync does not change the last modification time of the files that did not change, use the following:: rsync --compress --checksum --recursive path/to/blog remote_host:public/dir/ Accurate last modification time makes efficient caching possible. .. _rsync: http://samba.anu.edu.au/rsync/ .. vim:se tw=80 sw=2 ts=2 et encoding=utf-8: Henry Precheur e326036277e5bbf9085ba4bf97fbd9ea84139f42 0.9