Add backslash escaping to new link format

For users to be able to add square brackets inside of links using
the new format, the regular expression used for parsing those links
contained a balancing group, which can be used for matching pairs
of tokens (in this case, opening and closing brackets, in that order).
However, this means that users could not post links with unmatched
brackets inside of them (ie. ones that contain single brackets, or
a closing bracket and then an opening one). Allow for escaping opening
and closing brackets using the backslash character.

The change substitutes this old fragment of the regex in the display
text group:

    [^\[\]]*        // any character other than closing/opening bracket

for this one:

    (((?<=\\)[\[\]])|[^\[\]])*

The second pattern in the alternative remains the same; the first one
performs the escaping, as follows:

    (
        (?<=\\)     // positive lookbehind expression:
                    // this match will succeed, if the next expression
                    // is preceded by a single backslash
        [\[\]]      // either an opening or closing brace
    )

Since the entire display group is matched, unfortunately the lookbehind
expression does not actually strip the backslashes, so they are
manually stripped in handleMatches.

As demonstrated in the unit tests attached, this also allows balanced
brackets to be mixed with escaped ones.
This commit is contained in:
Bartłomiej Dach 2019-09-03 23:18:39 +02:00
parent 8d52d282e9
commit a8f16503e2
3 changed files with 45 additions and 3 deletions

View File

@ -131,6 +131,42 @@ public void TestNewFormatLink()
Assert.AreEqual(11, result.Links[0].Length);
}
[Test]
public void TestNewFormatLinkWithEscapedBrackets()
{
Message result = MessageFormatter.FormatMessage(new Message { Content = "This is a [https://osu.ppy.sh nasty link with escaped brackets: \\] and \\[]" });
Assert.AreEqual("This is a nasty link with escaped brackets: ] and [", result.DisplayContent);
Assert.AreEqual(1, result.Links.Count);
Assert.AreEqual("https://osu.ppy.sh", result.Links[0].Url);
Assert.AreEqual(10, result.Links[0].Index);
Assert.AreEqual(41, result.Links[0].Length);
}
[Test]
public void TestNewFormatLinkWithBackslashesInside()
{
Message result = MessageFormatter.FormatMessage(new Message { Content = "This is a [https://osu.ppy.sh link \\ with \\ backslashes \\]" });
Assert.AreEqual("This is a link \\ with \\ backslashes \\", result.DisplayContent);
Assert.AreEqual(1, result.Links.Count);
Assert.AreEqual("https://osu.ppy.sh", result.Links[0].Url);
Assert.AreEqual(10, result.Links[0].Index);
Assert.AreEqual(27, result.Links[0].Length);
}
[Test]
public void TestNewFormatLinkWithEscapedAndBalancedBrackets()
{
Message result = MessageFormatter.FormatMessage(new Message { Content = "This is a [https://osu.ppy.sh [link [with \\] too many brackets \\[ ]]]" });
Assert.AreEqual("This is a [link [with ] too many brackets [ ]]", result.DisplayContent);
Assert.AreEqual(1, result.Links.Count);
Assert.AreEqual("https://osu.ppy.sh", result.Links[0].Url);
Assert.AreEqual(10, result.Links[0].Index);
Assert.AreEqual(36, result.Links[0].Length);
}
[Test]
public void TestMarkdownFormatLink()
{

View File

@ -127,6 +127,7 @@ private void testLinksGeneral()
addMessageWithChecks("is now playing [https://osu.ppy.sh/b/252238 IMAGE -MATERIAL- <Version 0>]", 1, true, expectedActions: LinkAction.OpenBeatmap);
addMessageWithChecks("Let's (try)[https://osu.ppy.sh/home] [https://osu.ppy.sh/b/252238 multiple links] https://osu.ppy.sh/home", 3,
expectedActions: new[] { LinkAction.External, LinkAction.OpenBeatmap, LinkAction.External });
addMessageWithChecks("[https://osu.ppy.sh/home New link format with escaped [and \\[ paired] braces]", expectedActions: LinkAction.External);
// note that there's 0 links here (they get removed if a channel is not found)
addMessageWithChecks("#lobby or #osu would be blue (and work) in the ChatDisplay test (when a proper ChatOverlay is present).");
addMessageWithChecks("I am important!", 0, false, true);

View File

@ -3,6 +3,7 @@
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text.RegularExpressions;
namespace osu.Game.Online.Chat
@ -16,7 +17,7 @@ public static class MessageFormatter
private static readonly Regex old_link_regex = new Regex(@"\(([^\)]*)\)\[([a-z]+://[^ ]+)\]");
// [https://osu.ppy.sh/b/1234 Beatmap [Hard] (poop)] -> Beatmap [hard] (poop) (https://osu.ppy.sh/b/1234)
private static readonly Regex new_link_regex = new Regex(@"\[([a-z]+://[^ ]+) ([^\[\]]*(((?<open>\[)[^\[\]]*)+((?<close-open>\])[^\[\]]*)+)*(?(open)(?!)))\]");
private static readonly Regex new_link_regex = new Regex(@"\[([a-z]+://[^ ]+) ((((?<=\\)[\[\]])|[^\[\]])*(((?<open>\[)(((?<=\\)[\[\]])|[^\[\]])*)+((?<close-open>\])(((?<=\\)[\[\]])|[^\[\]])*)+)*(?(open)(?!)))\]");
// [test](https://osu.ppy.sh/b/1234) -> test (https://osu.ppy.sh/b/1234) aka correct markdown format
private static readonly Regex markdown_link_regex = new Regex(@"\[([^\]]*)\]\(([a-z]+://[^ ]+)\)");
@ -48,7 +49,7 @@ public static class MessageFormatter
// Unicode emojis
private static readonly Regex emoji_regex = new Regex(@"(\uD83D[\uDC00-\uDE4F])");
private static void handleMatches(Regex regex, string display, string link, MessageFormatterResult result, int startIndex = 0, LinkAction? linkActionOverride = null)
private static void handleMatches(Regex regex, string display, string link, MessageFormatterResult result, int startIndex = 0, LinkAction? linkActionOverride = null, char[] escapeChars = null)
{
int captureOffset = 0;
@ -68,6 +69,10 @@ private static void handleMatches(Regex regex, string display, string link, Mess
if (displayText.Length == 0 || linkText.Length == 0) continue;
// Remove backslash escapes in front of the characters provided in escapeChars
if (escapeChars != null)
displayText = escapeChars.Aggregate(displayText, (current, c) => current.Replace($"\\{c}", c.ToString()));
// Check for encapsulated links
if (result.Links.Find(l => (l.Index <= index && l.Index + l.Length >= index + m.Length) || (index <= l.Index && index + m.Length >= l.Index + l.Length)) == null)
{
@ -183,7 +188,7 @@ private static MessageFormatterResult format(string toFormat, int startIndex = 0
var result = new MessageFormatterResult(toFormat);
// handle the [link display] format
handleMatches(new_link_regex, "{2}", "{1}", result, startIndex);
handleMatches(new_link_regex, "{2}", "{1}", result, startIndex, escapeChars: new[] { '[', ']' });
// handle the standard markdown []() format
handleMatches(markdown_link_regex, "{1}", "{2}", result, startIndex);