From 05cdfe05998fecc9048f36286ab10215d75bd553 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Gon=C3=A7alo=20Silva?= <goncalossilva@gmail.com>
Date: Wed, 24 Jul 2019 23:33:37 +0100
Subject: [PATCH] Use CommonMark's regexp for autolink / autoemail

Ref: https://github.com/commonmark/commonmark.js/blob/master/lib/inlines.js#L62-L64

Also avoids special handling in stats counter.
---
 uberwriter/markup_regex.py             | 2 +-
 uberwriter/stats_counter.py            | 9 ++-------
 uberwriter/text_view_markup_handler.py | 9 +++++----
 3 files changed, 8 insertions(+), 12 deletions(-)

diff --git a/uberwriter/markup_regex.py b/uberwriter/markup_regex.py
index d0fa219..57f5fcb 100644
--- a/uberwriter/markup_regex.py
+++ b/uberwriter/markup_regex.py
@@ -13,7 +13,7 @@ CODE = re.compile(
 LINK = re.compile(
     r"\[(?P<text>.*)\]\((?P<url>.+?)(?: \"(?P<title>.+)\")?\)")
 LINK_ALT = re.compile(
-    r"(?:<)(?P<url>https?://[^\s]+)(?:>)")
+    r"<(?P<text>[A-Za-z][A-Za-z0-9.+-]{1,31}:[^<>\x00-\x20]*|(?:[a-zA-Z0-9.!#$%&'*+\/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*))>")
 IMAGE = re.compile(
     r"!\[(?P<text>.*)\]\((?P<url>.+?)(?: \"(?P<title>.+)\")?\)")
 HORIZONTAL_RULE = re.compile(
diff --git a/uberwriter/stats_counter.py b/uberwriter/stats_counter.py
index 8c1feab..8c1e342 100644
--- a/uberwriter/stats_counter.py
+++ b/uberwriter/stats_counter.py
@@ -27,11 +27,8 @@ class StatsCounter:
 
     # List of regexp whose matches should be replaced by their "text" group. Order is important.
     MARKUP_REGEXP_REPLACE = (
-        BOLD_ITALIC, ITALIC, BOLD, STRIKETHROUGH, IMAGE, LINK, LIST, ORDERED_LIST, BLOCK_QUOTE,
-        HEADER, HEADER_UNDER, CODE_BLOCK, TABLE, MATH, FOOTNOTE_ID, FOOTNOTE
-    )
-    MARKUP_REGEXP_URL_REPLACE = (
-        LINK_ALT,
+        BOLD_ITALIC, ITALIC, BOLD, STRIKETHROUGH, IMAGE, LINK, LINK_ALT, LIST, ORDERED_LIST,
+        BLOCK_QUOTE, HEADER, HEADER_UNDER, CODE_BLOCK, TABLE, MATH, FOOTNOTE_ID, FOOTNOTE
     )
 
     # List of regexp whose matches should be removed. Order is important.
@@ -78,8 +75,6 @@ class StatsCounter:
                     child_conn.close()
                     return
 
-            for regexp in self.MARKUP_REGEXP_URL_REPLACE:
-                text = re.sub(regexp, r"\g<url>", text)
             for regexp in self.MARKUP_REGEXP_REPLACE:
                 text = re.sub(regexp, r"\g<text>", text)
             for regexp in self.MARKUP_REGEXP_REMOVE:
diff --git a/uberwriter/text_view_markup_handler.py b/uberwriter/text_view_markup_handler.py
index 57bab60..e0089bc 100644
--- a/uberwriter/text_view_markup_handler.py
+++ b/uberwriter/text_view_markup_handler.py
@@ -20,8 +20,9 @@ from multiprocessing import Pipe, Process
 import gi
 
 from uberwriter import helpers, markup_regex
-from uberwriter.markup_regex import STRIKETHROUGH, BOLD_ITALIC, BOLD, ITALIC, IMAGE, LINK, LINK_ALT, \
-    HORIZONTAL_RULE, LIST, ORDERED_LIST, BLOCK_QUOTE, HEADER, HEADER_UNDER, TABLE, MATH, CODE
+from uberwriter.markup_regex import STRIKETHROUGH, BOLD_ITALIC, BOLD, ITALIC, IMAGE, LINK,\
+    LINK_ALT, HORIZONTAL_RULE, LIST, ORDERED_LIST, BLOCK_QUOTE, HEADER, HEADER_UNDER, TABLE, MATH, \
+    CODE
 
 gi.require_version('Gtk', '3.0')
 from gi.repository import Gtk, GLib
@@ -203,11 +204,11 @@ class MarkupHandler:
                     result.append((tag_name, (), match.start(), match.start("text")))
                     result.append((tag_name, (), match.end("text"), match.end()))
 
-            # - "<url>" (gray out)
+            # Find "<url>" links (gray out).
             matches = re.finditer(LINK_ALT, text)
             for match in matches:
                 result.append((
-                    self.TAG_NAME_GRAY_TEXT, (), match.start("url"), match.end("url")))
+                    self.TAG_NAME_GRAY_TEXT, (), match.start("text"), match.end("text")))
 
             # Find "---" horizontal rule (center).
             matches = re.finditer(HORIZONTAL_RULE, text)