libervia-backend: sat/plugins/plugin_misc_text

comparison sat/plugins/plugin_misc_text_syntaxes.py @ 2624:56f94936df1e

code style reformatting using black

author	Goffi <goffi@goffi.org>
date	Wed, 27 Jun 2018 20:14:46 +0200
parents	26edcf3a30eb
children	003b8b4b56a7

comparison

equal deleted inserted replaced

-:49533de4540b
+:56f94936df1e
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 from sat.core.i18n import _, D_
 from sat.core.constants import Const as C
 from sat.core.log import getLogger
 log = getLogger(__name__)
 from twisted.internet import defer
 from twisted.internet.threads import deferToThread
 from sat.core import exceptions
 try:
 from lxml import html
 from lxml.html import clean
 except ImportError:
-raise exceptions.MissingModule(u"Missing module lxml, please download/install it from http://lxml.de/")
+raise exceptions.MissingModule(
+u"Missing module lxml, please download/install it from http://lxml.de/"
+)
 from cgi import escape
 import re
 CATEGORY = D_("Composition")
 _SYNTAX_XHTML = "XHTML"
 _SYNTAX_CURRENT = "@CURRENT@"
 # TODO: check/adapt following list
 # list initialy based on feedparser list (http://pythonhosted.org/feedparser/html-sanitization.html)
-STYLES_WHITELIST = ("azimuth", "background-color", "border-bottom-color", "border-collapse", "border-color", "border-left-color", "border-right-color", "border-top-color", "clear", "color", "cursor", "direction", "display", "elevation", "float", "font", "font-family", "font-size", "font-style", "font-variant", "font-weight", "height", "letter-spacing", "line-height", "overflow", "pause", "pause-after", "pause-before", "pitch", "pitch-range", "richness", "speak", "speak-header", "speak-numeral", "speak-punctuation", "speech-rate", "stress", "text-align", "text-decoration", "text-indent", "unicode-bidi", "vertical-align", "voice-family", "volume", "white-space", "width")
+STYLES_WHITELIST = (
+"azimuth",
-SAFE_ATTRS = html.defs.safe_attrs.union(('style', 'poster', 'controls'))
+"background-color",
-STYLES_VALUES_REGEX = r'^(' + '|'.join(['([a-z-]+)', # alphabetical names
+"border-bottom-color",
-'(#[0-9a-f]+)', # hex value
+"border-collapse",
-'(\d+(.\d+)? *(|%|em|ex|px|in|cm|mm|pt|pc))', # values with units (or not)
+"border-color",
-'rgb\( *((\d+(.\d+)?), *){2}(\d+(.\d+)?) *\)', # rgb function
+"border-left-color",
-'rgba\( *((\d+(.\d+)?), *){3}(\d+(.\d+)?) *\)', # rgba function
+"border-right-color",
-]) + ') *(!important)?$' # we accept "!important" at the end
+"border-top-color",
+"clear",
+"color",
+"cursor",
+"direction",
+"display",
+"elevation",
+"float",
+"font",
+"font-family",
+"font-size",
+"font-style",
+"font-variant",
+"font-weight",
+"height",
+"letter-spacing",
+"line-height",
+"overflow",
+"pause",
+"pause-after",
+"pause-before",
+"pitch",
+"pitch-range",
+"richness",
+"speak",
+"speak-header",
+"speak-numeral",
+"speak-punctuation",
+"speech-rate",
+"stress",
+"text-align",
+"text-decoration",
+"text-indent",
+"unicode-bidi",
+"vertical-align",
+"voice-family",
+"volume",
+"white-space",
+"width",
+)
+SAFE_ATTRS = html.defs.safe_attrs.union(("style", "poster", "controls"))
+STYLES_VALUES_REGEX = (
+r"^("
++ "|".join(
+[
+"([a-z-]+)",  # alphabetical names
+"(#[0-9a-f]+)",  # hex value
+"(\d+(.\d+)? *(|%|em|ex|px|in|cm|mm|pt|pc))",  # values with units (or not)
+"rgb\( *((\d+(.\d+)?), *){2}(\d+(.\d+)?) *\)",  # rgb function
+"rgba\( *((\d+(.\d+)?), *){3}(\d+(.\d+)?) *\)",  # rgba function
+]
+)
++ ") *(!important)?$"
+)  # we accept "!important" at the end
 STYLES_ACCEPTED_VALUE = re.compile(STYLES_VALUES_REGEX)
 PLUGIN_INFO = {
 C.PI_NAME: "Text syntaxes",
 C.PI_IMPORT_NAME: "TEXT-SYNTAXES",
 C.PI_TYPE: "MISC",
 C.PI_PROTOCOLS: [],
 C.PI_DEPENDENCIES: [],
 C.PI_MAIN: "TextSyntaxes",
 C.PI_HANDLER: "no",
-C.PI_DESCRIPTION: _("""Management of various text syntaxes (XHTML-IM, Markdown, etc)""")
+C.PI_DESCRIPTION: _(
+"""Management of various text syntaxes (XHTML-IM, Markdown, etc)"""
+),
 }
 class TextSyntaxes(object):
 """ Text conversion class
 </individual>
 </params>
 """
 params_data = {
-'category_name': CATEGORY,
+"category_name": CATEGORY,
-'category_label': _(CATEGORY),
+"category_label": _(CATEGORY),
-'name': NAME,
+"name": NAME,
-'label': _(NAME),
+"label": _(NAME),
-'syntaxes': syntaxes,
+"syntaxes": syntaxes,
 }
 def __init__(self, host):
 log.info(_("Text syntaxes plugin initialization"))
 self.host = host
-self.addSyntax(self.SYNTAX_XHTML, lambda xhtml: defer.succeed(xhtml), lambda xhtml: defer.succeed(xhtml),
+self.addSyntax(
-TextSyntaxes.OPT_NO_THREAD)
+self.SYNTAX_XHTML,
+lambda xhtml: defer.succeed(xhtml),
+lambda xhtml: defer.succeed(xhtml),
+TextSyntaxes.OPT_NO_THREAD,
+)
 # TODO: text => XHTML should add <a/> to url like in frontends
 #       it's probably best to move sat_frontends.tools.strings to sat.tools.common or similar
-self.addSyntax(self.SYNTAX_TEXT, lambda text: escape(text), lambda xhtml: self._removeMarkups(xhtml), [TextSyntaxes.OPT_HIDDEN])
+self.addSyntax(
+self.SYNTAX_TEXT,
+lambda text: escape(text),
+lambda xhtml: self._removeMarkups(xhtml),
+[TextSyntaxes.OPT_HIDDEN],
+)
 try:
 import markdown, html2text
-def _html2text(html, baseurl=''):
+def _html2text(html, baseurl=""):
 h = html2text.HTML2Text(baseurl=baseurl)
 h.body_width = 0  # do not truncate the lines, it breaks the long URLs
 return h.handle(html)
-self.addSyntax(self.SYNTAX_MARKDOWN, markdown.markdown, _html2text, [TextSyntaxes.OPT_DEFAULT])
+self.addSyntax(
+self.SYNTAX_MARKDOWN,
+markdown.markdown,
+_html2text,
+[TextSyntaxes.OPT_DEFAULT],
+)
 except ImportError:
 log.warning(u"markdown or html2text not found, can't use Markdown syntax")
-log.info(u"You can download/install them from https://pythonhosted.org/Markdown/ and https://github.com/Alir3z4/html2text/")
+log.info(
-host.bridge.addMethod("syntaxConvert", ".plugin", in_sign='sssbs', out_sign='s',
+u"You can download/install them from https://pythonhosted.org/Markdown/ and https://github.com/Alir3z4/html2text/"
-async=True, method=self.convert)
+)
-host.bridge.addMethod("syntaxGet", ".plugin", in_sign='s', out_sign='s',
+host.bridge.addMethod(
-method=self.getSyntax)
+"syntaxConvert",
+".plugin",
+in_sign="sssbs",
+out_sign="s",
+async=True,
+method=self.convert,
+)
+host.bridge.addMethod(
+"syntaxGet", ".plugin", in_sign="s", out_sign="s", method=self.getSyntax
+)
 def _updateParamOptions(self):
 data_synt = TextSyntaxes.syntaxes
 default_synt = TextSyntaxes.default_syntax
 syntaxes = []
 syntaxes.sort(key=lambda synt: synt.lower())
 options = []
 for syntax in syntaxes:
-selected = 'selected="true"' if syntax == default_synt else ''
+selected = 'selected="true"' if syntax == default_synt else ""
 options.append(u'<option value="%s" %s/>' % (syntax, selected))
-TextSyntaxes.params_data["options"] = u'\n'.join(options)
+TextSyntaxes.params_data["options"] = u"\n".join(options)
 self.host.memory.updateParams(TextSyntaxes.params % TextSyntaxes.params_data)
 def getCurrentSyntax(self, profile):
 """ Return the selected syntax for the given profile
 @param profile: %(doc_profile)s
 @return: profile selected syntax
 """
-return self.host.memory.getParamA(NAME, CATEGORY , profile_key=profile)
+return self.host.memory.getParamA(NAME, CATEGORY, profile_key=profile)
 def _logError(self, failure, action=u"converting syntax"):
-log.error(u"Error while {action}: {failure}".format(action=action, failure=failure))
+log.error(
+u"Error while {action}: {failure}".format(action=action, failure=failure)
+)
 return failure
 def cleanXHTML(self, xhtml):
 """ Clean XHTML text by removing potentially dangerous/malicious parts
 @param xhtml: raw xhtml text to clean (or lxml's HtmlElement)
 """
 def blocking_cleaning(xhtml):
 """ Clean XHTML and style attributes """
 def clean_style(styles_raw):
 """" Remove styles not in the whitelist,
 or where the value doesn't match the regex """
 styles = styles_raw.split(";")
 cleaned_styles = []
 for style in styles:
 try:
-key, value = style.split(':')
+key, value = style.split(":")
 except ValueError:
 continue
 key = key.lower().strip()
 if key not in STYLES_WHITELIST:
 continue
 if not STYLES_ACCEPTED_VALUE.match(value):
 continue
 if value == "none":
 continue
 cleaned_styles.append((key, value))
-return "; ".join(["%s: %s" % (key_, value_) for key_, value_ in cleaned_styles])
+return "; ".join(
+["%s: %s" % (key_, value_) for key_, value_ in cleaned_styles]
+)
 if isinstance(xhtml, basestring):
 xhtml_elt = html.fromstring(xhtml)
 elif isinstance(xhtml, html.HtmlElement):
 xhtml_elt = xhtml
 else:
 log.error("Only strings and HtmlElements can be cleaned")
 raise exceptions.DataError
-cleaner = clean.Cleaner(style=False,
+cleaner = clean.Cleaner(
-add_nofollow=False,
+style=False, add_nofollow=False, safe_attrs=SAFE_ATTRS
-safe_attrs=SAFE_ATTRS)
+)
 xhtml_elt = cleaner.clean_html(xhtml_elt)
 for elt in xhtml_elt.xpath("//*[@style]"):
-elt.set("style", clean_style(elt.get('style')))
+elt.set("style", clean_style(elt.get("style")))
-return html.tostring(xhtml_elt, encoding=unicode, method='xml')
+return html.tostring(xhtml_elt, encoding=unicode, method="xml")
 d = deferToThread(blocking_cleaning, xhtml)
 d.addErrback(self._logError, action=u"cleaning syntax")
 return d
-def convert(self, text, syntax_from, syntax_to=_SYNTAX_XHTML, safe=True, profile=None):
+def convert(
+self, text, syntax_from, syntax_to=_SYNTAX_XHTML, safe=True, profile=None
+):
 """Convert a text between two syntaxes
 @param text: text to convert
 @param syntax_from: source syntax (e.g. "markdown")
 @param syntax_to: dest syntax (e.g.: "XHTML")
 if TextSyntaxes.OPT_NO_THREAD in syntaxes[syntax_from]["flags"]:
 d = defer.maybeDeferred(syntaxes[syntax_from]["to"], text)
 else:
 d = deferToThread(syntaxes[syntax_from]["to"], text)
-#TODO: keep only body element and change it to a div here ?
+# TODO: keep only body element and change it to a div here ?
 if safe:
 d.addCallback(self.cleanXHTML)
 if TextSyntaxes.OPT_NO_THREAD in syntaxes[syntax_to]["flags"]:
 # converters can add new lines that disturb the microblog change detection
 d.addCallback(lambda text: text.rstrip())
 return d
-def addSyntax(self, name, to_xhtml_cb, from_xhtml_cb, flags = None):
+def addSyntax(self, name, to_xhtml_cb, from_xhtml_cb, flags=None):
 """Add a new syntax to the manager
 @param name: unique name of the syntax
 @param to_xhtml_cb: callback to convert from syntax to XHTML
 @param from_xhtml_cb: callback to convert from XHTML to syntax
 TextSyntaxes.OPT_HIDDEN: do not show in parameters
 TextSyntaxes.OPT_NO_THREAD: do not defer to thread when converting (the callback may then return a deferred)
 """
 flags = flags if flags is not None else []
 if TextSyntaxes.OPT_HIDDEN in flags and TextSyntaxes.OPT_DEFAULT in flags:
-raise ValueError(u"{} and {} are mutually exclusive".format(TextSyntaxes.OPT_HIDDEN, TextSyntaxes.OPT_DEFAULT))
+raise ValueError(
+u"{} and {} are mutually exclusive".format(
+TextSyntaxes.OPT_HIDDEN, TextSyntaxes.OPT_DEFAULT
+)
+)
 syntaxes = TextSyntaxes.syntaxes
 key = name.lower().strip()
 if key in syntaxes:
-raise exceptions.ConflictError(u"This syntax key already exists: {}".format(key))
+raise exceptions.ConflictError(
-syntaxes[key] = {"name": name, "to": to_xhtml_cb, "from": from_xhtml_cb, "flags": flags}
+u"This syntax key already exists: {}".format(key)
+)
+syntaxes[key] = {
+"name": name,
+"to": to_xhtml_cb,
+"from": from_xhtml_cb,
+"flags": flags,
+}
 if TextSyntaxes.OPT_DEFAULT in flags:
 TextSyntaxes.default_syntaxe = key
 self._updateParamOptions()
 """Remove XHTML markups from the given string.
 @param xhtml: the XHTML string to be cleaned
 @return: the cleaned string
 """
-cleaner = clean.Cleaner(kill_tags=['style'])
+cleaner = clean.Cleaner(kill_tags=["style"])
 cleaned = cleaner.clean_html(html.fromstring(xhtml))
 return html.tostring(cleaned, encoding=unicode, method="text")

Mercurial > libervia-backend

comparison sat/plugins/plugin_misc_text_syntaxes.py @ 2624:56f94936df1e