[hargup  blog  (gh-pages) ]$tinker --build
Making output directory...
Running Sphinx v1.2
loading pickled environment... not yet created
building [html]: targets for 11 source files that are out of date
updating environment: 11 added, 0 changed, 0 removed
reading sources... [100%] master                                                
looking for now-outdated files... none found
pickling environment... done
checking consistency... done
preparing documents... done
writing output... [100%] master                                                 
writing additional files...
Exception occurred:
  File "/usr/lib/python2.7/xml/dom/expatbuilder.py", line 223, in parseString
    parser.Parse(string, True)
ExpatError: junk after document element: line 8, column 0
The full traceback has been saved in /tmp/sphinx-err-NEFQgD.log, if you want to report the issue to the developers.
Please also report this if it was a user error, so that a better error message can be provided next time.
Either send bugs to the mailing list at <http://groups.google.com/group/sphinx-users/>,
or report them in the tracker at <http://bitbucket.org/birkenfeld/sphinx/issues/>. Thanks!

Here's the log file

# Sphinx version: 1.2
# Python version: 2.7.3
# Docutils version: 0.11 release
# Jinja2 version: 2.7.1
# Loaded extensions:
#   tinkerer.ext.disqus from /usr/local/lib/python2.7/dist-packages/tinkerer/ext/disqus.pyc
#   tinkerer.ext.blog from /usr/local/lib/python2.7/dist-packages/tinkerer/ext/blog.pyc
#   sphinx.ext.oldcmarkup from /usr/local/lib/python2.7/dist-packages/sphinx/ext/oldcmarkup.pyc
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/sphinx/cmdline.py", line 246, in main
    app.build(force_all, filenames)
  File "/usr/local/lib/python2.7/dist-packages/sphinx/application.py", line 212, in build
    self.builder.build_update()
  File "/usr/local/lib/python2.7/dist-packages/sphinx/builders/__init__.py", line 214, in build_update
    'out of date' % len(to_build))
  File "/usr/local/lib/python2.7/dist-packages/sphinx/builders/__init__.py", line 273, in build
    self.finish()
  File "/usr/local/lib/python2.7/dist-packages/sphinx/builders/html.py", line 453, in finish
    for pagename, context, template in pagelist:
  File "/usr/local/lib/python2.7/dist-packages/tinkerer/ext/blog.py", line 107, in html_collect_pages
    for name, context, template in collect_additional_pages(app):
  File "/usr/local/lib/python2.7/dist-packages/tinkerer/ext/blog.py", line 86, in collect_additional_pages
    for name, context, template in rss.generate_feed(app):
  File "/usr/local/lib/python2.7/dist-packages/tinkerer/ext/rss.py", line 56, in generate_feed
    replace_read_more_link=not app.config.rss_generate_full_posts)),
  File "/usr/local/lib/python2.7/dist-packages/tinkerer/ext/patch.py", line 106, in patch_links
    doc = xml.dom.minidom.parseString(in_str)
  File "/usr/lib/python2.7/xml/dom/minidom.py", line 1930, in parseString
    return expatbuilder.parseString(string)
  File "/usr/lib/python2.7/xml/dom/expatbuilder.py", line 940, in parseString
    return builder.parseString(string)
  File "/usr/lib/python2.7/xml/dom/expatbuilder.py", line 223, in parseString
    parser.Parse(string, True)
ExpatError: junk after document element: line 8, column 0

I have added a quick and easy way to view the blog locally before posting it to the Web. Hopefully this will help with developing themes for Tinkerer, as well as, give us an opportunity to check out posts prior to uploading them. I have added unit tests to verify that the server is pulling posts as expected.

Thank you for your work on Tinkerer. It is great software and I have quite enjoyed hacking on it.

In metadata.py dates are formatted with .strftime() on date objects. This method is locale-aware and formats dates according to the current environment locale, that is, the value of $LANG or $LC_*.

Hence, formatting of dates in posts depends on the environment in which Tinkerer is run, and ignores the language setting in conf.py.

To build an English language blog on a system with, say, a German locale, you currently need to set LANG to en_US.utf8 explicitly (or a similar English-language locale) to make the post dates appear in English. Otherwise they will be German, i.e. Januar 1, 2013 instead of January 1, 2013.

I can submit a pull request to fix this issue, but it would introduce a dependency against the Babel i18n library. Is that acceptable?

IMHO, will be nice if can use tinkerer for write a site and not a blog. Do it must be "easy" since:

• a blog must have a post in master.rst and
• a site must have none post in master.rst and a page named index.rst or home.rst.

For the above part I can write a patch. But I can't figure it out how to stop tinkerer stop try to generate the blog home page, any help?

It would be really cool if we could have publication time in the post timestamp.

This is useful if you write several posts a day. It's vital for blogs used to indicate some online service status.

Hay there, i'm looking for sth. like http://django-pipeline.readthedocs.org/en/latest/compressors.html for tinkerer. Is there any option/plugin for asset compression? Actually i never worked with sphinx or tinkerer, but i can't believe there is no such extension.

Thanks in advance, Lukas

If the <div id="more"> is enclosed by <div id="some-other-div"> then doc('p.readmorewrapper').next_all().remove() will not remove everything that should be removed. Sometimes it works, sometimes not, it depends where the .. more:: gets inserted.

Example:

make_read_more_link:

def make_read_more_link(body, docpath, docname):
    """
    Create "read more" link if marker exists.
    """
    doc = pyquery.PyQuery(body)
    link_p = ('<p class="readmorewrapper"><a class="readmore" '
              'href="%s.html#more">%s</a></p>' %
              (docpath + docname, UIStr.READ_MORE))
    doc('div#more').replaceWith(link_p)
    doc('p.readmorewrapper').next_all().remove()
    return doc.html()

body:

<h1>Example</h1>
<div class="figure align-center">
</div>
<p>....</p>
<p>Why?</p>

<div id="some-other-div">

Text...

<div id="more"> </div>

</div>

<p>more text</p>

<p>...</p>

</div>

IPython session:

In [19]: body='''
<h1>Example</h1>
<div class="figure align-center">
</div>
<p>....</p>
<p>Why?</p>
<div id="some-other-div">
Text...
<div id="more"> </div>
</div>
<p>more text</p>
<p>...</p>
</div>
'''

In [20]: doc = pyquery.PyQuery(body)

In [21]: link_p = ('<p class="readmorewrapper"><a class="readmore" '
   ....:                 'href="%s.html#more">%s</a></p>' %
   ....:                 ("/index.html", "READ_MORE"))

In [22]: doc('div#more').replaceWith(link_p)
Out[22]: [<div#more>]

In [23]: doc('p.readmorewrapper').next_all().remove()
Out[23]: []

In [24]: doc.html()
Out[24]: u'<h1>Example</h1>\n
<div class="figure align-center">\n
</div>\n
<p>....</p>\n
<p>Why?</p>\n
<div id="some-other-div">\n
Text...\n
<p class="readmorewrapper"><a class="readmore" href="/index.html.html#more">READ_MORE</a></p>\n
</div>\n
<p>more text</p>\n
<p>...</p>\n\n'

In [25]: 

<p>more text</p>... should have been removed.

I'm currently investigating solutions... But maybe @dhellmann has a good idea?

I'd like to enable the html_use_index option found here:

https://github.com/vladris/tinkerer/blob/master/tinkerer/__templates/conf.py#L93

Out of the box, tinkerer errors as it can't find the genindex.html template, but I copied the one over from sphinx and dropped it in the themes/boilerplate directory and it worked fine.

Would the project be opposed to me submitting a PR to allow this option?

Not all HTML can be parsed with a strict XML parser. Some Sphinx extensions insert HTML from sources outside of the local author's control (sphinxcontrib-embedly, for example). In order to support RSS feeds with valid links and the embedded content from these extensions, use a more forgiving HTML parser when patching the links.

Change-Id: I045e338289b63832315210b31a7188d5c6ab6570

A word_sep keyword argument is added to several functions and is ultimately used in utils.name_from_title(). I also ensure that only one instance of the word sep char appears between words and that there are no leading or trailing non-word chars in the name. I'd like to propose the same for other page/post/draft creation options.

I just found Tinkerer yesterday and it seems to be exactly what I've been looking for in a blogging tool!

I want to control the extension for internal links. For example, if you hide the extension using the MultiViews feature of Apache. This is a patch for that.

Fixes summary

• Change to app.config.html_link_suffix from hard code '.html'(Sphinx has html_link_suffix)
• Add html_link_suffix to MetaData class
• Append html_link_suffix to metadata.link

There are small typos in:

• blog/2012/07/05/tinkerer_0_4_beta_released.rst
• blog/2013/06/05/tinkerer_1_2_released.rst
• blog/doc/extensions.rst
• blog/pages/extensions.rst
• tinkerer/ext/filing.py
• tinkerer/ext/metadata.py
• tinkertest/test_categories.py
• tinkertest/test_cmdline.py
• tinkertest/test_master.py
• tinkertest/test_patch.py

Fixes:

• Should read repository rather than respository.
• Should read exists rather than exsits.
• Should read documentation rather than documententation.
• Should read development rather than developement.
• Should read containing rather than containig.
• Should read compatibility rather than compatiblity.
• Should read category rather than cateogry.
• Should read categories rather than categoires.
• Should read categories rather than caregories.
• Should read appended rather than appendend.
• Should read aggregate rather than aggreated.

Semi-automated pull request generated by https://github.com/timgates42/meticulous/blob/master/docs/NOTE.md

Sphinx 3.2.1 already defines a config value named author, so this patch just passes when an ExtensionError is raised when trying to add_config_value("author").

The tests pass with tox (upgraded to py38) and local output appears to be unchanged from building before this patch.

closes #112

closes #117

There's a new nonfatal error which may be resolved by a fix for #114 (~Allow passthrough configuration from conf.py)

tinker --build
Running Sphinx v3.2.1
WARNING: The config value `html_add_permalinks' has type `NoneType', defaults to `str'.

Is that a regression, or can this be merged?

This issue is bypassed, not solved. I've posted it anyway, so that others might use the bypass solution described here or can help to get a better solution.

Issue of using sphinx stand alone and in combination with tinkerer

Tinkerer 1.7.2 has Sphinx 1.7.1 as requirement. The latest Sphinx is 3.1.2 (July 2020) and I'm also Sphinx as standalone (without tinkerer) and would like to use the latest version. When I first install sphinx:

pip install sphinx

I get version 3.1.2, but when I also install tinkerer:

pip install tinkerer

The output mentions that sphinx is downgraded to version 1.7.1 due to tinkerer requirements.

Considered but not chosen solutions

A python virtual environment cannot handle this as it can only handle different python versions, e.g. 3.7 and 3.8, but not different module versions. I've also considered pipenv and poetry, but these require configuration files and seems to be more oriented towards developers and I wanted something more simple. A virtual machine like virtualbox seems to be too much overload.

Chosen solution

The chosen solution uses pipx and is described here.

First remove tinkerer and sphinx from your installation:

pip uninstall sphinx
pip uninstall tinkerer

I'm using (K)ubuntu with the aptitude package manager and installed pipx via:

sudo aptitude -y install pipx
pipx ensurepath

Without aptitude, you might want to use apt or apt-get. Pipx can be installed in many other ways, see: pipx .

When I tried tinkerer this way to generate my website, I got error messages, and I found out Tinkerer calls Sphinx via a subprocess (instead of 'within' python), Hence, it breaks out of the pipx virtual python environment... This can be prevented by editing the file (using the nano editor):

nano ~/.local/pipx/venvs/tinkerer/lib/python3.8/site-packages/tinkerer/cmdline.py

and change line 49 from flags = ["sphinx-build"] to:

flags = ["~/.local/pipx/venvs/tinkerer/bin/sphinx-build"]

When you want to use additional modules, you also have to add them in pipx, as for example to add matplotlib:

pipx inject tinkerer requests matplotlib

I also made some changes to the conf.py file, perhaps not needed... but anyway:

• tinkerer imports at the beginning are commented out... they are already loaded elsewhere, hence not needed
• all references to tinkerer _themes and _static are removed... as I use self defined, locally stored _themes and _static
• some variables are hard-coded instead of referring to tinkerer settings as in:

source_suffix = ".rst"
master_doc = "master"

Now tinkerer generates my website fluently, using sphinx 1.7.1

I can now also install sphinx latest version:

pip install --upgrade sphinx

which gives me version 3.1.2. And this runs well

AFAIK there's no way to force Tinkerer where to build it's output. This is essential configuration that should be modifiable in conf.py with variables e.g. thinker_blog_dir, thinker_html_dir etc.

Hi,

I fixes to support Sphinx 2.3.0, and Python 3.7, 3.8. If you like, could you please merge the pull request?

The main modifications are as follows.

1. tox on python 3.7, python3.8.

• Appends setenv = LANG = C for Non-English LANG.
• Drops Python 3.6 supports. (Python 3.6 does not work on Debian GNU/Linux bullseye/Sid.)

2. Fixes config.author bug for Sphinx >= 1.8.0.
3. Fixes sphinx.main() of unittest for Sphinx >= 2.0.0.
4. Fixes setup argument test_requries.
5. Supports Sphinx version >= 2.3.0.

I develops and tests using Pipenv.

$ pipenv --python 3.7
$ pipenv install tox
$ pipenv run tox

But not fixes RemovedInSphinx30Warning.

/path/to/tinkerer/tinkerer/themes/boilerplate/search.html:20: 
RemovedInSphinx30Warning: To modify script_files in the theme is deprecated.
Please insert a <script> tag directly in your theme instead.

Best regards,

This currently fails with

$ tinker --build
Running Sphinx v1.8.1

Extension error:
Config value 'author' already present

The problem goes away by remove the line

app.add_config_value("author", "Winston Smith", True)

in https://github.com/vladris/tinkerer/blob/master/tinkerer/ext/blog.py#L126

The problem of course is that pip install tinkerer is currently broken because it does not pin the version of Sphinx to its tested version.