Added TeX escaping utility functions and template filters. Removed

html escaping and added TeX escaping for relevant parts of the bibtext
template.  Fixes issue ietf-tools#2459.
 - Legacy-Id: 14711

Author: levkowetz

ietf/templates/doc/document_bibtex.bib

@@ -1,6 +1,7 @@
 {% spaceless %}
-
+{% autoescape off %}
 {% load ietf_filters %}
+{% load textfilters %}
 
 {% if doc.get_state_slug == "rfc" %}
 {% if doc.stream|slugify == "legacy" %}
@@ -24,13 +25,14 @@ @techreport{
 	publisher =	{% templatetag openbrace %}Internet Engineering Task Force{% templatetag closebrace %},
 	note =		{% templatetag openbrace %}Work in Progress{% templatetag closebrace %},
 	url =		{% templatetag openbrace %}https://datatracker.ietf.org/doc/html/{{doc.name}}-{{doc.rev}}{% templatetag closebrace %},{% endif %}
-        author =	{% templatetag openbrace %}{% for author in doc.documentauthor_set.all %}{{ author.person.name}}{% if not forloop.last %} and {% endif %}{% endfor %}{% templatetag closebrace %},
-	title =		{% templatetag openbrace %}{% templatetag openbrace %}{{doc.title}}{% templatetag closebrace %}{% templatetag closebrace %},
+        author =	{% templatetag openbrace %}{% for author in doc.documentauthor_set.all %}{{ author.person.name|texescape}}{% if not forloop.last %} and {% endif %}{% endfor %}{% templatetag closebrace %},
+	title =		{% templatetag openbrace %}{% templatetag openbrace %}{{doc.title|texescape}}{% templatetag closebrace %}{% templatetag closebrace %},
 	pagetotal =	{{ doc.pages }},
 	year =		{{ doc.pub_date.year }},
 	month =		{{ doc.pub_date|date:"b" }},{% if not doc.rfc_number or doc.pub_date.day == 1 and doc.pub_date.month == 4  %}
 	day =		{{ doc.pub_date.day }},{% endif %}
-	abstract =	{% templatetag openbrace %}{{ doc.abstract|clean_whitespace }}{% templatetag closebrace %},
+	abstract =	{% templatetag openbrace %}{{ doc.abstract|clean_whitespace|texescape }}{% templatetag closebrace %},
 {% templatetag closebrace %}
 
+{% endautoescape %}
 {% endspaceless %}

ietf/utils/templatetags/textfilters.py

@@ -1,11 +1,13 @@
 from __future__ import unicode_literals
 
-from django.template.library import Library
+from django import template
 from django.template.defaultfilters import stringfilter
 
-from ietf.utils.text import xslugify as _xslugify
+import debug                            # pyflakes:ignore
 
-register = Library()
+from ietf.utils.text import xslugify as _xslugify, texescape
+
+register = template.Library()
 
 @register.filter(is_safe=True)
 @stringfilter
@@ -26,3 +28,40 @@ def format(format, values):
             tmp[f.name] = getattr(values, f.name)
         values = tmp
     return format.format(**values)
+
+# ----------------------------------------------------------------------
+
+# from django.utils.safestring import mark_safe
+# class TeXEscapeNode(template.Node):
+#     """TeX escaping, rather than html escaping.
+# 
+#     Mostly, this tag is _not_ the right thing to use in a template that produces TeX
+#     markup, as it will escape all the markup characters, which is not what you want.
+#     Use the '|texescape' filter instead on text fragments where escaping is needed
+#     """
+#     def __init__(self, nodelist):
+#         self.nodelist = nodelist
+# 
+#     def render(self, context):
+#         saved_autoescape = context.autoescape
+#         context.autoescape = False
+#         text = self.nodelist.render(context)
+#         text = texescape(text)
+#         context.autoescape = saved_autoescape
+#         return mark_safe(text)
+# 
+# @register.tag('texescape')
+# def do_texescape(parser, token):
+#     args = token.contents.split()
+#     if len(args) != 1:
+#         raise TemplateSyntaxError("'texescape' tag takes no arguments.")
+#     nodelist = parser.parse(('endtexescape',))
+#     parser.delete_first_token()
+#     return TeXEscapeNode(nodelist)
+    
+@register.filter('texescape')
+@stringfilter
+def texescape_filter(value):
+    "A TeX escape filter"
+    return texescape(value)
+    

ietf/utils/texescape.py

@@ -0,0 +1,136 @@
+# -*- coding: utf-8 -*-
+# Copied from https://github.com/sphinx-doc/sphinx/blob/master/sphinx/util/texescape.py
+# Copyright and license as indicated in the original and below
+"""
+    sphinx.util.texescape
+    ~~~~~~~~~~~~~~~~~~~~~
+
+    TeX escaping helper.
+
+    :copyright: Copyright 2007-2018 by the Sphinx team, see AUTHORS.
+    :license: BSD, see LICENSE for details.
+"""
+
+from __future__ import unicode_literals
+
+tex_replacements = [
+    # map TeX special chars
+    ('$', r'\$'),
+    ('%', r'\%'),
+    ('&', r'\&'),
+    ('#', r'\#'),
+    ('_', r'\_'),
+    ('{', r'\{'),
+    ('}', r'\}'),
+    ('[', r'{[}'),
+    (']', r'{]}'),
+    ('`', r'{}`'),
+    ('\\', r'\textbackslash{}'),
+    ('~', r'\textasciitilde{}'),
+    ('<', r'\textless{}'),
+    ('>', r'\textgreater{}'),
+    ('^', r'\textasciicircum{}'),
+    # map special Unicode characters to TeX commands
+    ('¶', r'\P{}'),
+    ('§', r'\S{}'),
+    ('€', r'\texteuro{}'),
+    ('∞', r'\(\infty\)'),
+    ('±', r'\(\pm\)'),
+    ('→', r'\(\rightarrow\)'),
+    ('‣', r'\(\rightarrow\)'),
+    ('✓', r'\(\checkmark\)'),
+    # used to separate -- in options
+    ('', r'{}'),
+    # map some special Unicode characters to similar ASCII ones
+    ('⎽', r'\_'),
+    ('–', r'\textendash{}'),
+    ('|', r'\textbar{}'),
+    ('ℯ', r'e'),
+    ('ⅈ', r'i'),
+    ('⁰', r'\(\sp{\text{0}}\)'),
+    ('¹', r'\(\sp{\text{1}}\)'),
+    ('²', r'\(\sp{\text{2}}\)'),
+    ('³', r'\(\sp{\text{3}}\)'),
+    ('⁴', r'\(\sp{\text{4}}\)'),
+    ('⁵', r'\(\sp{\text{5}}\)'),
+    ('⁶', r'\(\sp{\text{6}}\)'),
+    ('⁷', r'\(\sp{\text{7}}\)'),
+    ('⁸', r'\(\sp{\text{8}}\)'),
+    ('⁹', r'\(\sp{\text{9}}\)'),
+    ('₀', r'\(\sb{\text{0}}\)'),
+    ('₁', r'\(\sb{\text{1}}\)'),
+    ('₂', r'\(\sb{\text{2}}\)'),
+    ('₃', r'\(\sb{\text{3}}\)'),
+    ('₄', r'\(\sb{\text{4}}\)'),
+    ('₅', r'\(\sb{\text{5}}\)'),
+    ('₆', r'\(\sb{\text{6}}\)'),
+    ('₇', r'\(\sb{\text{7}}\)'),
+    ('₈', r'\(\sb{\text{8}}\)'),
+    ('₉', r'\(\sb{\text{9}}\)'),
+    # map Greek alphabet
+    ('α', r'\(\alpha\)'),
+    ('β', r'\(\beta\)'),
+    ('γ', r'\(\gamma\)'),
+    ('δ', r'\(\delta\)'),
+    ('ε', r'\(\epsilon\)'),
+    ('ζ', r'\(\zeta\)'),
+    ('η', r'\(\eta\)'),
+    ('θ', r'\(\theta\)'),
+    ('ι', r'\(\iota\)'),
+    ('κ', r'\(\kappa\)'),
+    ('λ', r'\(\lambda\)'),
+    ('μ', r'\(\mu\)'),
+    ('ν', r'\(\nu\)'),
+    ('ξ', r'\(\xi\)'),
+    ('ο', r'o'),
+    ('π', r'\(\pi\)'),
+    ('ρ', r'\(\rho\)'),
+    ('σ', r'\(\sigma\)'),
+    ('τ', r'\(\tau\)'),
+    ('υ', '\\(\\upsilon\\)'),
+    ('φ', r'\(\phi\)'),
+    ('χ', r'\(\chi\)'),
+    ('ψ', r'\(\psi\)'),
+    ('ω', r'\(\omega\)'),
+    ('Α', r'A'),
+    ('Β', r'B'),
+    ('Γ', r'\(\Gamma\)'),
+    ('Δ', r'\(\Delta\)'),
+    ('Ε', r'E'),
+    ('Ζ', r'Z'),
+    ('Η', r'H'),
+    ('Θ', r'\(\Theta\)'),
+    ('Ι', r'I'),
+    ('Κ', r'K'),
+    ('Λ', r'\(\Lambda\)'),
+    ('Μ', r'M'),
+    ('Ν', r'N'),
+    ('Ξ', r'\(\Xi\)'),
+    ('Ο', r'O'),
+    ('Π', r'\(\Pi\)'),
+    ('Ρ', r'P'),
+    ('Σ', r'\(\Sigma\)'),
+    ('Τ', r'T'),
+    ('Υ', '\\(\\Upsilon\\)'),
+    ('Φ', r'\(\Phi\)'),
+    ('Χ', r'X'),
+    ('Ψ', r'\(\Psi\)'),
+    ('Ω', r'\(\Omega\)'),
+    ('Ω', r'\(\Omega\)'),
+]
+
+tex_escape_map = {}
+tex_replace_map = {}
+tex_hl_escape_map_new = {}
+
+
+def init():
+    # type: () -> None
+    for a, b in tex_replacements:
+        tex_escape_map[ord(a)] = b
+        tex_replace_map[ord(a)] = '_'
+
+    for a, b in tex_replacements:
+        if a in '[]{}\\':
+            continue
+        tex_hl_escape_map_new[ord(a)] = b

ietf/utils/text.py

@@ -11,6 +11,8 @@
 
 import debug                            # pyflakes:ignore
 
+from texescape import init as texescape_init, tex_escape_map
+
 @keep_lazy(six.text_type)
 def xslugify(value):
     """
@@ -174,3 +176,9 @@ def dict_to_text(d):
     for k, v in d.items():
         t += "%s: %s\n" % (k, v)
     return t
+
+def texescape(s):
+    if not tex_escape_map:
+        texescape_init()
+    t = s.translate(tex_escape_map)
+    return t