From 6a1952e44cef4e3ec40d3b4509f20d8844cfb1f2 Mon Sep 17 00:00:00 2001
From: "J.-S. Caux" <J.S.Caux@uva.nl>
Date: Thu, 13 Jun 2019 09:19:50 +0200
Subject: [PATCH] Improve markup language detection
---
markup/utils.py | 122 ++++++++++++++++++++++++++++--------------------
markup/views.py | 1 +
2 files changed, 72 insertions(+), 51 deletions(-)
diff --git a/markup/utils.py b/markup/utils.py
index 3e0487510..d19851c0e 100644
--- a/markup/utils.py
+++ b/markup/utils.py
@@ -106,12 +106,48 @@ def match_rst_header(text, symbol=None):
return None
if symbol not in ReST_HEADER_REGEX_DICT.keys():
raise ValueError('symbol is not a ReST header symbol')
- print('Looking for %s in rst: %s' % (
- symbol,
- re.search(ReST_HEADER_REGEX_DICT[symbol], text, re.MULTILINE)))
return re.search(ReST_HEADER_REGEX_DICT[symbol], text, re.MULTILINE)
+def check_markers(markers):
+ """
+ Checks the consistency of a markers dictionary. Returns a detector.
+ """
+ if len(markers['rst']) > 0:
+ if len(markers['md']) > 0:
+ return {
+ 'language': 'plain',
+ 'errors': ('Inconsistency: Markdown and reStructuredText syntaxes are mixed:\n\n'
+ 'Markdown: %s\n\nreStructuredText: %s' % (
+ markers['md'].popitem(),
+ markers['rst'].popitem()))
+ }
+ if len(markers['plain_or_md']) > 0:
+ return {
+ 'language': 'plain',
+ 'errors': ('Inconsistency: plain/Markdown and reStructuredText '
+ 'syntaxes are mixed:\n\n'
+ 'Markdown: %s\n\nreStructuredText: %s' % (
+ markers['plain_or_md'].popitem(),
+ markers['rst'].popitem()))
+ }
+ return {
+ 'language': 'reStructuredText',
+ 'errors': None,
+ }
+
+ elif len(markers['md']) > 0:
+ return {
+ 'language': 'Markdown',
+ 'errors': None,
+ }
+
+ return {
+ 'language': 'plain',
+ 'errors': None,
+ }
+
+
def detect_markup_language(text):
"""
Detect whether text is plain text, Markdown or reStructuredText.
@@ -163,68 +199,48 @@ def detect_markup_language(text):
* if the ``math`` role or directive is found together with inline/displayed maths
"""
- # Start from the default assumption
- detector = {
- 'language': 'plain',
- 'errors': None
+ markers = {
+ 'plain_or_md': {},
+ 'md': {},
+ 'rst': {},
}
# Step 1: check maths
# Inline maths is of the form $ ... $ or \( ... \)
- inline_math = match_inline_math(text)
+ match = match_inline_math(text)
+ if match:
+ markers['plain_or_md']['inline_math'] = match
# Displayed maths is of the form \[ ... \] or $$ ... $$
- displayed_math = match_displayed_math(text)
-
- rst_math_role = match_rst_role(text, 'math')
- rst_math_directive = match_rst_directive(text, 'math')
-
- if rst_math_role or rst_math_directive:
- # reStructuredText presumed; check for errors
- if inline_math:
- detector['errors'] = (
- 'You have mixed inline maths ($ ... $ or \( ... \) ) with '
- 'reStructuredText markup.\n\nPlease use one or the other, but not both!')
- return detector
- elif displayed_math:
- detector['errors'] = (
- 'You have mixed displayed maths ($$ ... $$ or \[ ... \]) with '
- 'reStructuredText markup.\n\nPlease use one or the other, but not both!')
- return detector
- else:
- detector['language'] = 'reStructuredText'
- return detector
+ match = match_displayed_math(text)
+ if match:
+ markers['plain_or_md']['displayed_math'] = match
- # no rst math from here onwards
+ match = match_rst_role(text, 'math')
+ if match:
+ markers['rst']['math_role'] = match
+ match = match_rst_directive(text, 'math')
+ if match:
+ markers['rst']['math_directive'] = match
# Step 2: check headers and blockquotes
- md_header = match_md_header(text)
- print('md_header: %s' % md_header)
- md_blockquote = match_md_blockquote(text)
-
- rst_header = match_rst_header(text)
- print('rst_header: %s' % rst_header)
-
- if md_header or md_blockquote:
- if rst_math_role or rst_math_directive:
- if md_header:
- detector['errors'] = (
- 'You have mixed Markdown headers with reStructuredText math '
- 'roles/directives.\n\nPlease use one language only.')
- elif md_blockquote:
- detector['errors'] = (
- 'You have mixed Markdown blockquotes with reStructuredText math '
- 'roles/directives.\n\nPlease use one language only.')
- detector['language'] = 'Markdown'
+ match = match_md_header(text)
+ if match:
+ markers['md']['header'] = match
+ match = match_md_blockquote(text)
+ if match:
+ markers['md']['blockquote'] = match
- elif md_header or md_blockquote:
- detector['language'] = 'Markdown'
+ match = match_rst_header(text)
+ if match:
+ markers['rst']['header'] = match
- elif rst_header:
- detector['language'] = 'reStructuredText'
+ print('markers: \n%s' % markers)
+ detector = check_markers(markers)
+ print('detector: \n%s' % detector)
return detector
@@ -397,6 +413,10 @@ def process_markup(text, language_forced=None):
language = language_forced if language_forced else markup_detector['language']
markup['language'] = language
+ markup['errors'] = markup_detector['errors']
+
+ if markup['errors']:
+ return markup
if language == 'reStructuredText':
warnStream = StringIO()
diff --git a/markup/views.py b/markup/views.py
index b5e254e44..b6f1f5d67 100644
--- a/markup/views.py
+++ b/markup/views.py
@@ -22,6 +22,7 @@ def process(request):
"""
form = MarkupTextForm(request.POST or None)
if form.is_valid():
+ print('response: \n%s' % form.get_processed_markup())
return JsonResponse(form.get_processed_markup())
return JsonResponse({})
--
GitLab