sebix / python-textile Goto Github PK

View Code? Open in Web Editor NEW

62.0 7.0 42.0 669 KB

A Python port of Textile, A humane web text generator

Home Page: https://github.com/textile/python-textile/

License: Other

Python 100.00%

textile markup-language markup-converter python

python-textile's Introduction

python-textile

python-textile is a Python port of Textile, Dean Allen’s humane web text generator.

Installation

pip install textile

Dependencies:

html5lib
regex (The regex package causes problems with PyPy, and is not installed as a dependency in such environments. If you are upgrading a textile install on PyPy which had regex previously included, you may need to uninstall it.)

Optional dependencies include:

PIL/Pillow (for checking image sizes). If needed, install via pip install 'textile[imagesize]'

Usage

import textile
>>> s = """
... _This_ is a *test.*
...
... * One
... * Two
... * Three
...
... Link to "Slashdot":http://slashdot.org/
... """
>>> html = textile.textile(s)
>>> print html
	<p><em>This</em> is a <strong>test.</strong></p>

	<ul>
		<li>One</li>
		<li>Two</li>
		<li>Three</li>
	</ul>

	<p>Link to <a href="http://slashdot.org/">Slashdot</a></p>
>>>

Notes:

Active development supports Python 3.5 or later.

python-textile's People

Contributors

Stargazers

Watchers

python-textile's Issues

Repeating list modifiers output incorrect HTML

If the input is something like this:

text="First line\n\n** Point 1\n* Point2\nLast line"
print textile.textile_restricted(text)

The output looks like this:

<p>First line</p>

<p> <ul>
    <li>Point 1</li>
</ul></li>
<ul>
    <li>Point2</li>
</ul><br />Last line</p>

Its the same for #, - and *. The behavior is the same for two or any more repeating.

2.1.5 : no longer possible to opt out of paragraph wrapper

in previous versions a leading space would prevent the text from being wrapped in a paragraph tag. 2.1.5 seems to have broken this.

Unicode support for links

Take the following textile link example:

"Superhuman":https://en.wikipedia.org/wiki/Superhuman
"Übermensch":https://de.wikipedia.org/wiki/Übermensch

This will be rendered as:

<p><a href="https://en.wikipedia.org/wiki/Superhuman">Superhuman</a><br />
&#8220;Übermensch&#8221;:https://de.wikipedia.org/wiki/Übermensch</p>

The second link will not be parsed correctly, since it contains the Unicode character 'Ü'. This happens because the links() function uses A-Za-z in the regular expression:

        pattern = r'''
            (?P<pre>[\s\[{(]|[%s])?         #leading text
            "                               #opening quote
            (?P<atts>%s)                    #block attributes
            (?P<text>[^"]+?)                #link text
            \s?
            (?:\((?P<title>[^)]+?)\)(?="))? #optional title
            ":                              #closing quote, colon
            (?P<url>(?:ftp|https?)?         #URL
                        (?: :// )?
                        [-A-Za-z0-9+&@#/?=~_()|!:,.;%%]*
                        [-A-Za-z0-9+&@#/=~_()|]
            )
            (?P<post>[^\w\/;]*?)        #trailing text
            (?=<|\s|$)
        ''' % (re.escape(punct), self.c)

        text = re.compile(pattern, re.X).sub(self.fLink, text)

Instead, \w in combination with the re.UNICODE flag should be used:

        pattern = r'''
            (?P<pre>[\s\[{(]|[%s])?         #leading text
            "                               #opening quote
            (?P<atts>%s)                    #block attributes
            (?P<text>[^"]+?)                #link text
            \s?
            (?:\((?P<title>[^)]+?)\)(?="))? #optional title
            ":                              #closing quote, colon
            (?P<url>(?:ftp|https?)?         #URL
                        (?: :// )?
                        [-\w+&@#/?=~()|!:,.;%%]*
                        [-\w+&@#/=~()|]
            )
            (?P<post>[^\w\/;]*?)        #trailing text
            (?=<|\s|$)
        ''' % (re.escape(punct), self.c)

        text = re.compile(pattern, re.X | re.UNICODE).sub(self.fLink, text)

There are other occurences of the a-z or A-Z expression in the source code... they should maybe replaced too.

ImportError on Python 3

Hi, I'm trying to use textile for a project of mine that I want to support Python 3. Importing the textile package triggers an ImportError.

You can reproduce this with tox. Here is my tox.ini file:

[tox]
envlist = py34

[testenv]
deps = nose
       coverage
commands = nosetests

I think the problem is related to relative imports. Here are my results:

matt@eden:~/python-textile$ tox -e py34
GLOB sdist-make: /home/matt/python-textile/setup.py
py34 inst-nodeps: /home/matt/python-textile/.tox/dist/textile-2.1.5.zip
py34 runtests: PYTHONHASHSEED='4057464796'
py34 runtests: commands[0] | nosetests
E
======================================================================
ERROR: Failure: ImportError (No module named 'functions')
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/matt/python-textile/.tox/py34/lib/python3.4/site-packages/nose/failure.py", line 39, in runTest
    raise self.exc_val.with_traceback(self.tb)
  File "/home/matt/python-textile/.tox/py34/lib/python3.4/site-packages/nose/loader.py", line 414, in loadTestsFromName
    addr.filename, addr.module)
  File "/home/matt/python-textile/.tox/py34/lib/python3.4/site-packages/nose/importer.py", line 47, in importFromPath
    return self.importFromDir(dir_path, fqname)
  File "/home/matt/python-textile/.tox/py34/lib/python3.4/site-packages/nose/importer.py", line 94, in importFromDir
    mod = load_module(part_fqname, fh, filename, desc)
  File "/home/matt/python-textile/.tox/py34/lib/python3.4/imp.py", line 245, in load_module
    return load_package(name, filename)
  File "/home/matt/python-textile/.tox/py34/lib/python3.4/imp.py", line 217, in load_package
    return methods.load()
  File "<frozen importlib._bootstrap>", line 1220, in load
  File "<frozen importlib._bootstrap>", line 1200, in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1129, in _exec
  File "<frozen importlib._bootstrap>", line 1448, in exec_module
  File "<frozen importlib._bootstrap>", line 321, in _call_with_frames_removed
  File "/home/matt/python-textile/textile/__init__.py", line 1, in <module>
    from functions import textile, textile_restricted, Textile
ImportError: No module named 'functions'

Name                      Stmts   Miss  Cover   Missing
-------------------------------------------------------
textile/__init__              2      1    50%   3
textile/functions           469    469     0%   4-1001
textile/tests/__init__      139    139     0%   2-442
textile/textilefactory       27     27     0%   1-74
textile/tools/__init__        0      0   100%   
textile/tools/imagesize      24     24     0%   1-41
textile/tools/sanitizer      20     20     0%   1-35
-------------------------------------------------------
TOTAL                       681    680     1%   
----------------------------------------------------------------------
Ran 1 test in 0.039s

FAILED (errors=1)
ERROR: InvocationError: '/home/matt/python-textile/.tox/py34/bin/nosetests'
___________________________________ summary ____________________________________
ERROR:   py34: commands failed

Empty lines in pre block break pre

Sorry, I can't think of a good title.

But if there are empty lines inside the pre, output is broken. Think the same thing happens with code or notextile blocks too. I will try to look at the code and figure out a fix, but I'm still early in my python development. :)

[# ] import textile

[# ] s = """<pre>
   ...: xxxxxxxx
   ...: yyyyyyyy
   ...: zzzzzzzz
   ...: </pre>
   ...: 
   ...: <pre>
   ...: aaaaaaa
   ...: 
   ...: bbbbbbb
   ...: 
   ...: ccccccc
   ...: </pre>
   ...: """

[# ] textile.textile(s)
Out[5]: '<pre>\nxxxxxxxx\nyyyyyyyy\nzzzzzzzz\n</pre>\n\n\t<p><pre><br />aaaaaaa</p>\n\n\t<p>bbbbbbb</p>\n\n\t<p>ccccccc<br /></pre></p>'

[# ] !grep version setup.py
    version='2.1.5',

Missing tags

The 'upstream' repository for textile has changed over time, and it seems some tags have been lost on the way.

As far as I can tell, version 2.1.4 corresponds to commit 84bdad3 and version 2.1.5 corresponds to commit 4554407.

Could you re-add them?

bad em markup in anchor href attribute

Generally I like how textile let's HTML pass through, but I recently came across a problem along the lines of the following:

import textile
print textile.textile("""<a href="_link_">label</a>""")

from which python-textile 2.1.5 produces:

<p><a href="<em>link</em>&#8220;>label</a></p>

which not only is wrong, but introduces a run-away attribute value that swallows parts of the page!

auto_link=True breaks link syntax

Given you have the following textile text:

"Superhuman":https://en.wikipedia.org/wiki/Superhuman
https://en.wikipedia.org/wiki/Superhuman
www.wikipedia.org
foo en.wikipedia.org/wiki/Superhuman bar

Rendered with auto_link=True, the output is:

<p>
&#8220;Superhuman&#8221;:<a href="https://en.wikipedia.org/wiki/Superhuman" rel="nofollow">https://en.wikipedia.org/wiki/Superhuman</a><br />
<a href="https://en.wikipedia.org/wiki/Superhuman" rel="nofollow">https://en.wikipedia.org/wiki/Superhuman</a><br />
<a href="www.wikipedia.org" rel="nofollow">www.wikipedia.org</a><br />
foo <a href="en.wikipedia.org/wiki/Superhuman" rel="nofollow">en.wikipedia.org/wiki/Superhuman</a> bar
</p>

Obviously autoLink() breaks the textile link syntax ("a link":http://example.org/). I'd rather expect the following output, with both working:

<p>
<a href="https://en.wikipedia.org/wiki/Superhuman" rel="nofollow">Superhuman</a><br />
<a href="https://en.wikipedia.org/wiki/Superhuman" rel="nofollow">https://en.wikipedia.org/wiki/Superhuman</a><br />
<a href="www.wikipedia.org" rel="nofollow">www.wikipedia.org</a><br />
foo <a href="en.wikipedia.org/wiki/Superhuman" rel="nofollow">en.wikipedia.org/wiki/Superhuman</a> bar
</p>

It probably has something to do with the leading \b in the autoLink() pattern which not only matches end of words:

pattern = re.compile(r"""\b((?:https?://|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:'".,<>?«»“”‘’]))""", re.U | re.I)
        return pattern.sub(r'"\1":\1', text)

if there is not blank line between tables, textile will combine them together as one table

|6.12.27global| 10h00m39s239ms  074 | 38.46 | 1040 |some comments|
##title##
 || date ||version    ||data         ||data	||data	||comments||
|..|...

There are 2 tables, but textile always combine them together as one table.
I have to add blank lines like this:

|6.12.27global| 10h00m39s239ms  074 | 38.46 | 1040 |some comments|

##title##

 || date ||version    ||data         ||data	||data	||comments||
|..|...

EDIT 2017-02-06T21:04: Fix formatting

URLs containing ^ or * are not idenitified.

If a URL contains either of the above two characters, they are not correctly identified by the URL parser. The text will show up un-parsed. Test cases:

"Marketing Operations Managers":http://www.linkedin.com/groups?home=&gid=138990&goback=%2Enmp_*1_*1_*1_*1_*1_*1_*1_*1_*1&trk=grp-name 

"US Government bonds":http://finance.yahoo.com/echarts?s=^TYX+Interactive#chart2:symbol=^tyx;range=6m;indicator=volume;charttype=line;crosshair=on;ohlcvalues=0;logscale=on;source=undefined