merged the autolink patterns branch by Arun

3 files changed, 140 insertions, 16 deletions

diff --git a/askbot/conf/markup.py b/askbot/conf/markup.py
index 026c5536..e4202f05 100644
--- a/askbot/conf/markup.py
+++ b/askbot/conf/markup.py
@@ -4,17 +4,34 @@ Settings that modify processing of user text input
 
 from askbot.conf.settings_wrapper import settings
 from askbot.deps.livesettings import ConfigurationGroup
-from askbot.deps.livesettings import BooleanValue, StringValue
+from askbot.deps.livesettings import BooleanValue, StringValue, LongStringValue
 from django.utils.translation import ugettext as _
-import askbot
 from askbot import const
-import os
+import re
 
 MARKUP = ConfigurationGroup(
                     'MARKUP',
                     _('Markup formatting')
                 )
 
+def regex_settings_validation(*args):
+    """
+    Validate the regular expressions
+    """
+    try:
+
+        new_value = args[1]
+        regex_list = new_value.split('\n')
+        
+        for i in range(0, len(regex_list)):
+            re.compile(regex_list[i].strip())
+        return args[1]
+    
+    except Exception:
+        # The regex is invalid, so we overwrite it with empty string
+        return ""
+
+
 settings.register(
     BooleanValue(
         MARKUP,
@@ -63,3 +80,60 @@ settings.register(
         default = ''
     )
 )
+
+
+settings.register(
+    BooleanValue(
+        MARKUP,
+        'ENABLE_AUTO_LINKING',
+        description=_('Enable autolinking with specific patterns'),
+        help_text=_(
+            'If you enable this feature, '
+            'the application  will be able to '
+            'detect patterns and auto link to URLs'        
+        ),
+        default = False
+    )
+)
+
+
+settings.register(
+    LongStringValue(
+        MARKUP,
+        'AUTO_LINK_PATTERNS',
+        description=_('Regexes to detect the link patterns'),
+        help_text=_(
+            'Enter valid regular expressions for the patters,'
+            ' one per line.'
+            ' For example to'
+            ' detect a bug pattern like #bug123,'
+            ' use the following regex: #bug(\d+). The numbers'
+            ' captured by the pattern in the parentheses will'
+            ' be transferred to the link url template.'
+            ' Please look up more information about regular'
+            ' expressions elsewhere.'
+        ),
+        update_callback=regex_settings_validation,
+        default = ''
+        )
+    )
+
+settings.register(
+    LongStringValue(
+        MARKUP,
+        'AUTO_LINK_URLS',
+        description=_('URLs for autolinking'),
+        help_text=_(
+            'Here, please enter url templates for the patterns'
+            ' entered in the previous setting, also one entry per line.'
+            ' <strong>Make sure that number of lines in this setting'
+            ' and the previous one are the same</strong>'
+            ' For example template'
+            ' https://bugzilla.redhat.com/show_bug.cgi?id=\\1'
+            ' together with the pattern shown above'
+            ' and the entry in the post #123'
+            ' will produce link to the bug 123 in the redhat bug tracker.'
+        ),
+        default = ''
+    )
+)
diff --git a/askbot/doc/source/changelog.rst b/askbot/doc/source/changelog.rst
index 87933311..33a0199c 100644
--- a/askbot/doc/source/changelog.rst
+++ b/askbot/doc/source/changelog.rst
@@ -1,12 +1,13 @@
 Changes in Askbot
 =================
 
-0.7.24 (development version - not released)
--------------------------------------------
+Development version (not released yet)
+--------------------------------------
 * Added annotations for the meanings of user levels on the "moderation" page. (Jishnu)
+* Auto-link patterns - e.g. to bug databases - are configurable from settings. (Arun SAG)
 
-0.7.23 (Current)
--------------------
+0.7.23 (Current Version)
+------------------------
 * Greeting for anonymuos users can be changed from live settings (Hrishi)
 * Greeting for anonymous users is shown only once (Rag Sagar)
 * Added support for Akismet spam detection service (Adolfo Fitoria)
diff --git a/askbot/utils/markup.py b/askbot/utils/markup.py
index acb8d03d..60bde9a0 100644
--- a/askbot/utils/markup.py
+++ b/askbot/utils/markup.py
@@ -1,8 +1,13 @@
+"""methods that make parsing of post inputs possible,
+handling of markdown and additional syntax rules - 
+such as optional link patterns, video embedding and 
+Twitter-style @mentions"""
+
 import re
+import logging
 from askbot import const
 from askbot.conf import settings as askbot_settings
 from markdown2 import Markdown
-
 #url taken from http://regexlib.com/REDetails.aspx?regexp_id=501 by Brian Bothwell
 URL_RE = re.compile("((?<!(href|.src|data)=['\"])((http|https|ftp)\://([a-zA-Z0-9\.\-]+(\:[a-zA-Z0-9\.&amp;%\$\-]+)*@)*((25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9])\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[0-9])|localhost|([a-zA-Z0-9\-]+\.)*[a-zA-Z0-9\-]+\.(com|edu|gov|int|mil|net|org|biz|arpa|info|name|pro|aero|coop|museum|[a-zA-Z]{2}))(\:[0-9]+)*(/($|[a-zA-Z0-9\.\,\?\'\\\+&amp;%\$#\=~_\-]+))*))")
 
@@ -10,7 +15,10 @@ LINK_PATTERNS = [
     (URL_RE, r'\1'),
 ]
 
+
 def get_parser():
+    """returns an instance of configured ``markdown2`` parser
+    """
     extras = ['link-patterns', 'video']  
     if askbot_settings.ENABLE_MATHJAX or \
         askbot_settings.MARKUP_CODE_FRIENDLY:
@@ -22,6 +30,30 @@ def get_parser():
         #pip install -e git+git://github.com/andryuha/python-markdown2.git
         extras.append('video')
 
+    if askbot_settings.ENABLE_AUTO_LINKING:
+        pattern_list = askbot_settings.AUTO_LINK_PATTERNS.split('\n')
+        url_list = askbot_settings.AUTO_LINK_URLS.split('\n')
+        pairs = zip(pattern_list, url_list)#always takes equal number of items 
+        for item in pairs:
+            LINK_PATTERNS.append(
+                (
+                    re.compile(item[0]),
+                    item[1].strip()
+                )
+            )
+        
+        #Check whether  we have matching links for all key terms,
+        #Other wise we ignore the key terms
+        #May be we should do this test in update_callback?
+        #looks like this might be a defect of livesettings
+        #as there seems to be no way
+        #to validate entries that depend on each other
+        if len(pattern_list) != len(url_list):
+            settings_url = askbot_settings.APP_URL+'/settings/AUTOLINK/'
+            logging.critical(
+                "Number of autolink patterns didn't match the number "
+                "of url templates, fix this by visiting" + settings_url) 
+            
     return Markdown(
                 html4tags=True,
                 extras=extras,
@@ -30,18 +62,23 @@ def get_parser():
 
 
 def format_mention_in_html(mentioned_user):
+    """formats mention as url to the user profile"""
     url = mentioned_user.get_profile_url()
     username = mentioned_user.username
     return '<a href="%s">@%s</a>' % (url, username)
 
 def extract_first_matching_mentioned_author(text, anticipated_authors):
+    """matches beginning of ``text`` string with the names
+    of ``anticipated_authors`` - list of user objects.
+    Returns upon first match the first matched user object
+    and the remainder of the ``text`` that is left unmatched"""
 
     if len(text) == 0:
         return None, ''
 
-    for a in anticipated_authors:
-        if text.lower().startswith(a.username.lower()):
-            ulen = len(a.username)
+    for author in anticipated_authors:
+        if text.lower().startswith(author.username.lower()):
+            ulen = len(author.username)
             if len(text) == ulen:
                 text = ''
             elif text[ulen] in const.TWITTER_STYLE_MENTION_TERMINATION_CHARS:
@@ -50,17 +87,24 @@ def extract_first_matching_mentioned_author(text, anticipated_authors):
                 #near miss, here we could insert a warning that perhaps
                 #a termination character is needed
                 continue
-            return a, text
+            return author, text
     return None, text
 
 def extract_mentioned_name_seeds(text):
+    """Returns list of strings that
+    follow the '@' symbols in the text.
+    The strings will be 10 characters long,
+    or shorter, if the subsequent character
+    is one of the list accepted to be termination
+    characters.
+    """
     extra_name_seeds = set()
     while '@' in text:
         pos = text.index('@')
         text = text[pos+1:]#chop off prefix
         name_seed = ''
-        for c in text:
-            if c in const.TWITTER_STYLE_MENTION_TERMINATION_CHARS:
+        for char in text:
+            if char in const.TWITTER_STYLE_MENTION_TERMINATION_CHARS:
                 extra_name_seeds.add(name_seed)
                 name_seed = ''
                 break
@@ -68,12 +112,12 @@ def extract_mentioned_name_seeds(text):
                 extra_name_seeds.add(name_seed)
                 name_seed = ''
                 break
-            if c == '@':
+            if char == '@':
                 if len(name_seed) > 0:
                     extra_name_seeds.add(name_seed)
                     name_seed = ''
                 break
-            name_seed += c
+            name_seed += char
         if len(name_seed) > 0:
             #in case we run off the end of text
             extra_name_seeds.add(name_seed)
@@ -81,6 +125,11 @@ def extract_mentioned_name_seeds(text):
     return extra_name_seeds
 
 def mentionize_text(text, anticipated_authors):
+    """Returns a tuple of two items:
+    * modified text where @mentions are
+      replaced with urls to the corresponding user profiles
+    * list of users whose names matched the @mentions
+    """
     output = ''
     mentioned_authors = list()
     while '@' in text: