calvin.erfmann/DiscordChatExporter
1
0
Fork 0
mirror of https://github.com/Tyrrrz/DiscordChatExporter.git synced 2026-04-28 16:57:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Cleanup

Browse Source
This commit is contained in:
Tyrrrz
2021-02-22 03:15:09 +02:00
parent bed0ade732
commit ebe4d58a42
101 changed files with 330 additions and 310 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
DiscordChatExporter.Core/Markdown/Ast/EmojiNode.cs Normal file
View File

@@ -0,0 +1,27 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal class EmojiNode : MarkdownNode
{
public string? Id { get; }
public string Name { get; }
public bool IsAnimated { get; }
public bool IsCustomEmoji => !string.IsNullOrWhiteSpace(Id);
public EmojiNode(string? id, string name, bool isAnimated)
{
Id = id;
Name = name;
IsAnimated = isAnimated;
}
public EmojiNode(string name)
: this(null, name, false)
{
}
public override string ToString() => $"<Emoji> {Name}";
}
}

29
DiscordChatExporter.Core/Markdown/Ast/FormattedNode.cs Normal file
View File

@@ -0,0 +1,29 @@
using System.Collections.Generic;
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal enum TextFormatting
{
Bold,
Italic,
Underline,
Strikethrough,
Spoiler,
Quote
}
internal class FormattedNode : MarkdownNode
{
public TextFormatting Formatting { get; }
public IReadOnlyList<MarkdownNode> Children { get; }
public FormattedNode(TextFormatting formatting, IReadOnlyList<MarkdownNode> children)
{
Formatting = formatting;
Children = children;
}
public override string ToString() => $"<{Formatting}> (+{Children.Count})";
}
}

14
DiscordChatExporter.Core/Markdown/Ast/InlineCodeBlockNode.cs Normal file
View File

@@ -0,0 +1,14 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal class InlineCodeBlockNode : MarkdownNode
{
public string Code { get; }
public InlineCodeBlockNode(string code)
{
Code = code;
}
public override string ToString() => $"<Code> {Code}";
}
}

22
DiscordChatExporter.Core/Markdown/Ast/LinkNode.cs Normal file
View File

@@ -0,0 +1,22 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal class LinkNode : MarkdownNode
{
public string Url { get; }
public string Title { get; }
public LinkNode(string url, string title)
{
Url = url;
Title = title;
}
public LinkNode(string url)
: this(url, url)
{
}
public override string ToString() => $"<Link> {Title}";
}
}

6
DiscordChatExporter.Core/Markdown/Ast/MarkdownNode.cs Normal file
View File

@@ -0,0 +1,6 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal abstract class MarkdownNode
{
}
}

25
DiscordChatExporter.Core/Markdown/Ast/MentionNode.cs Normal file
View File

@@ -0,0 +1,25 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal enum MentionType
{
Meta,
User,
Channel,
Role
}
internal class MentionNode : MarkdownNode
{
public string Id { get; }
public MentionType Type { get; }
public MentionNode(string id, MentionType type)
{
Id = id;
Type = type;
}
public override string ToString() => $"<{Type} mention> {Id}";
}
}

17
DiscordChatExporter.Core/Markdown/Ast/MultiLineCodeBlockNode.cs Normal file
View File

@@ -0,0 +1,17 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal class MultiLineCodeBlockNode : MarkdownNode
{
public string Language { get; }
public string Code { get; }
public MultiLineCodeBlockNode(string language, string code)
{
Language = language;
Code = code;
}
public override string ToString() => $"<{Language}> {Code}";
}
}

14
DiscordChatExporter.Core/Markdown/Ast/TextNode.cs Normal file
View File

@@ -0,0 +1,14 @@
namespace DiscordChatExporter.Core.Markdown.Ast
{
internal class TextNode : MarkdownNode
{
public string Text { get; }
public TextNode(string text)
{
Text = text;
}
public override string ToString() => Text;
}
}

291
DiscordChatExporter.Core/Markdown/MarkdownParser.cs Normal file
View File

@@ -0,0 +1,291 @@
using System.Collections.Generic;
using System.Linq;
using System.Text.RegularExpressions;
using DiscordChatExporter.Core.Markdown.Ast;
using DiscordChatExporter.Core.Markdown.Matching;
namespace DiscordChatExporter.Core.Markdown
{
// The following parsing logic is meant to replicate Discord's markdown grammar as close as possible
internal static partial class MarkdownParser
{
private const RegexOptions DefaultRegexOptions =
RegexOptions.Compiled |
RegexOptions.CultureInvariant |
RegexOptions.Multiline;
/* Formatting */
// Capture any character until the earliest double asterisk not followed by an asterisk
private static readonly IMatcher<MarkdownNode> BoldFormattedNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\*\\*(.+?)\\*\\*(?!\\*)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Bold, Parse(p.Slice(m.Groups[1])))
);
// Capture any character until the earliest single asterisk not preceded or followed by an asterisk
// Opening asterisk must not be followed by whitespace
// Closing asterisk must not be preceded by whitespace
private static readonly IMatcher<MarkdownNode> ItalicFormattedNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\*(?!\\s)(.+?)(?<!\\s|\\*)\\*(?!\\*)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Italic, Parse(p.Slice(m.Groups[1])))
);
// Capture any character until the earliest triple asterisk not followed by an asterisk
private static readonly IMatcher<MarkdownNode> ItalicBoldFormattedNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\*(\\*\\*.+?\\*\\*)\\*(?!\\*)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Italic, Parse(p.Slice(m.Groups[1]), BoldFormattedNodeMatcher))
);
// Capture any character except underscore until an underscore
// Closing underscore must not be followed by a word character
private static readonly IMatcher<MarkdownNode> ItalicAltFormattedNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("_([^_]+)_(?!\\w)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Italic, Parse(p.Slice(m.Groups[1])))
);
// Capture any character until the earliest double underscore not followed by an underscore
private static readonly IMatcher<MarkdownNode> UnderlineFormattedNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("__(.+?)__(?!_)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Underline, Parse(p.Slice(m.Groups[1])))
);
// Capture any character until the earliest triple underscore not followed by an underscore
private static readonly IMatcher<MarkdownNode> ItalicUnderlineFormattedNodeMatcher =
new RegexMatcher<MarkdownNode>(
new Regex("_(__.+?__)_(?!_)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Italic,
Parse(p.Slice(m.Groups[1]), UnderlineFormattedNodeMatcher))
);
// Capture any character until the earliest double tilde
private static readonly IMatcher<MarkdownNode> StrikethroughFormattedNodeMatcher =
new RegexMatcher<MarkdownNode>(
new Regex("~~(.+?)~~", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Strikethrough, Parse(p.Slice(m.Groups[1])))
);
// Capture any character until the earliest double pipe
private static readonly IMatcher<MarkdownNode> SpoilerFormattedNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\|\\|(.+?)\\|\\|", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Spoiler, Parse(p.Slice(m.Groups[1])))
);
// Capture any character until the end of the line
// Opening 'greater than' character must be followed by whitespace
private static readonly IMatcher<MarkdownNode> SingleLineQuoteNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("^>\\s(.+\n?)", DefaultRegexOptions),
(p, m) => new FormattedNode(TextFormatting.Quote, Parse(p.Slice(m.Groups[1])))
);
// Repeatedly capture any character until the end of the line
// This one is tricky as it ends up producing multiple separate captures which need to be joined
private static readonly IMatcher<MarkdownNode> RepeatedSingleLineQuoteNodeMatcher =
new RegexMatcher<MarkdownNode>(
new Regex("(?:^>\\s(.+\n?)){2,}", DefaultRegexOptions),
(_, m) =>
{
var content = string.Concat(m.Groups[1].Captures.Select(c => c.Value));
return new FormattedNode(TextFormatting.Quote, Parse(content));
}
);
// Capture any character until the end of the input
// Opening 'greater than' characters must be followed by whitespace
private static readonly IMatcher<MarkdownNode> MultiLineQuoteNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("^>>>\\s(.+)", DefaultRegexOptions | RegexOptions.Singleline),
(p, m) => new FormattedNode(TextFormatting.Quote, Parse(p.Slice(m.Groups[1])))
);
/* Code blocks */
// Capture any character except backtick until a backtick
// Blank lines at the beginning and end of content are trimmed
// There can be either one or two backticks, but equal number on both sides
private static readonly IMatcher<MarkdownNode> InlineCodeBlockNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("(`{1,2})([^`]+)\\1", DefaultRegexOptions | RegexOptions.Singleline),
m => new InlineCodeBlockNode(m.Groups[2].Value.Trim('\r', '\n'))
);
// Capture language identifier and then any character until the earliest triple backtick
// Language identifier is one word immediately after opening backticks, followed immediately by newline
// Blank lines at the beginning and end of content are trimmed
private static readonly IMatcher<MarkdownNode> MultiLineCodeBlockNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("```(?:(\\w*)\\n)?(.+?)```", DefaultRegexOptions | RegexOptions.Singleline),
m => new MultiLineCodeBlockNode(m.Groups[1].Value, m.Groups[2].Value.Trim('\r', '\n'))
);
/* Mentions */
// Capture @everyone
private static readonly IMatcher<MarkdownNode> EveryoneMentionNodeMatcher = new StringMatcher<MarkdownNode>(
"@everyone",
_ => new MentionNode("everyone", MentionType.Meta)
);
// Capture @here
private static readonly IMatcher<MarkdownNode> HereMentionNodeMatcher = new StringMatcher<MarkdownNode>(
"@here",
_ => new MentionNode("here", MentionType.Meta)
);
// Capture <@123456> or <@!123456>
private static readonly IMatcher<MarkdownNode> UserMentionNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("<@!?(\\d+)>", DefaultRegexOptions),
m => new MentionNode(m.Groups[1].Value, MentionType.User)
);
// Capture <#123456>
private static readonly IMatcher<MarkdownNode> ChannelMentionNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("<#(\\d+)>", DefaultRegexOptions),
m => new MentionNode(m.Groups[1].Value, MentionType.Channel)
);
// Capture <@&123456>
private static readonly IMatcher<MarkdownNode> RoleMentionNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("<@&(\\d+)>", DefaultRegexOptions),
m => new MentionNode(m.Groups[1].Value, MentionType.Role)
);
/* Emojis */
// Capture any country flag emoji (two regional indicator surrogate pairs)
// ... or "miscellaneous symbol" character
// ... or surrogate pair
// ... or digit followed by enclosing mark
// (this does not match all emojis in Discord but it's reasonably accurate enough)
private static readonly IMatcher<MarkdownNode> StandardEmojiNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("((?:[\\uD83C][\\uDDE6-\\uDDFF]){2}|[\\u2600-\\u26FF]|\\p{Cs}{2}|\\d\\p{Me})",
DefaultRegexOptions),
m => new EmojiNode(m.Groups[1].Value)
);
// Capture <:lul:123456> or <a:lul:123456>
private static readonly IMatcher<MarkdownNode> CustomEmojiNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("<(a)?:(.+?):(\\d+?)>", DefaultRegexOptions),
m => new EmojiNode(m.Groups[3].Value, m.Groups[2].Value, !string.IsNullOrWhiteSpace(m.Groups[1].Value))
);
/* Links */
// Capture [title](link)
private static readonly IMatcher<MarkdownNode> TitledLinkNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\[(.+?)\\]\\((.+?)\\)", DefaultRegexOptions),
m => new LinkNode(m.Groups[2].Value, m.Groups[1].Value)
);
// Capture any non-whitespace character after http:// or https:// until the last punctuation character or whitespace
private static readonly IMatcher<MarkdownNode> AutoLinkNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("(https?://\\S*[^\\.,:;\"\'\\s])", DefaultRegexOptions),
m => new LinkNode(m.Groups[1].Value)
);
// Same as auto link but also surrounded by angular brackets
private static readonly IMatcher<MarkdownNode> HiddenLinkNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("<(https?://\\S*[^\\.,:;\"\'\\s])>", DefaultRegexOptions),
m => new LinkNode(m.Groups[1].Value)
);
/* Text */
// Capture the shrug emoticon
// This escapes it from matching for formatting
private static readonly IMatcher<MarkdownNode> ShrugTextNodeMatcher = new StringMatcher<MarkdownNode>(
@"¯\_(ツ)_/¯",
p => new TextNode(p.ToString())
);
// Capture some specific emojis that don't get rendered
// This escapes it from matching for emoji
private static readonly IMatcher<MarkdownNode> IgnoredEmojiTextNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("(\\u26A7|\\u2640|\\u2642|\\u2695|\\u267E|\\u00A9|\\u00AE|\\u2122)", DefaultRegexOptions),
m => new TextNode(m.Groups[1].Value)
);
// Capture any "symbol/other" character or surrogate pair preceded by a backslash
// This escapes it from matching for emoji
private static readonly IMatcher<MarkdownNode> EscapedSymbolTextNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\\\(\\p{So}|\\p{Cs}{2})", DefaultRegexOptions),
m => new TextNode(m.Groups[1].Value)
);
// Capture any non-whitespace, non latin alphanumeric character preceded by a backslash
// This escapes it from matching for formatting or other tokens
private static readonly IMatcher<MarkdownNode> EscapedCharacterTextNodeMatcher = new RegexMatcher<MarkdownNode>(
new Regex("\\\\([^a-zA-Z0-9\\s])", DefaultRegexOptions),
m => new TextNode(m.Groups[1].Value)
);
// Combine all matchers into one
// Matchers that have similar patterns are ordered from most specific to least specific
private static readonly IMatcher<MarkdownNode> AggregateNodeMatcher = new AggregateMatcher<MarkdownNode>(
// Escaped text
ShrugTextNodeMatcher,
IgnoredEmojiTextNodeMatcher,
EscapedSymbolTextNodeMatcher,
EscapedCharacterTextNodeMatcher,
// Formatting
ItalicBoldFormattedNodeMatcher,
ItalicUnderlineFormattedNodeMatcher,
BoldFormattedNodeMatcher,
ItalicFormattedNodeMatcher,
UnderlineFormattedNodeMatcher,
ItalicAltFormattedNodeMatcher,
StrikethroughFormattedNodeMatcher,
SpoilerFormattedNodeMatcher,
MultiLineQuoteNodeMatcher,
RepeatedSingleLineQuoteNodeMatcher,
SingleLineQuoteNodeMatcher,
// Code blocks
MultiLineCodeBlockNodeMatcher,
InlineCodeBlockNodeMatcher,
// Mentions
EveryoneMentionNodeMatcher,
HereMentionNodeMatcher,
UserMentionNodeMatcher,
ChannelMentionNodeMatcher,
RoleMentionNodeMatcher,
// Links
TitledLinkNodeMatcher,
AutoLinkNodeMatcher,
HiddenLinkNodeMatcher,
// Emoji
StandardEmojiNodeMatcher,
CustomEmojiNodeMatcher
);
// Minimal set of matchers for non-multimedia formats (e.g. plain text)
private static readonly IMatcher<MarkdownNode> MinimalAggregateNodeMatcher = new AggregateMatcher<MarkdownNode>(
// Mentions
EveryoneMentionNodeMatcher,
HereMentionNodeMatcher,
UserMentionNodeMatcher,
ChannelMentionNodeMatcher,
RoleMentionNodeMatcher,
// Emoji
CustomEmojiNodeMatcher
);
private static IReadOnlyList<MarkdownNode> Parse(StringPart stringPart, IMatcher<MarkdownNode> matcher) =>
matcher
.MatchAll(stringPart, p => new TextNode(p.ToString()))
.Select(r => r.Value)
.ToArray();
}
internal static partial class MarkdownParser
{
private static IReadOnlyList<MarkdownNode> Parse(StringPart stringPart) => Parse(stringPart, AggregateNodeMatcher);
private static IReadOnlyList<MarkdownNode> ParseMinimal(StringPart stringPart) => Parse(stringPart, MinimalAggregateNodeMatcher);
public static IReadOnlyList<MarkdownNode> Parse(string input) => Parse(new StringPart(input));
public static IReadOnlyList<MarkdownNode> ParseMinimal(string input) => ParseMinimal(new StringPart(input));
}
}

45
DiscordChatExporter.Core/Markdown/MarkdownVisitor.cs Normal file
View File

@@ -0,0 +1,45 @@
using System;
using System.Collections.Generic;
using DiscordChatExporter.Core.Markdown.Ast;
namespace DiscordChatExporter.Core.Markdown
{
internal abstract class MarkdownVisitor
{
protected virtual MarkdownNode VisitText(TextNode text) => text;
protected virtual MarkdownNode VisitFormatted(FormattedNode formatted)
{
Visit(formatted.Children);
return formatted;
}
protected virtual MarkdownNode VisitInlineCodeBlock(InlineCodeBlockNode inlineCodeBlock) => inlineCodeBlock;
protected virtual MarkdownNode VisitMultiLineCodeBlock(MultiLineCodeBlockNode multiLineCodeBlock) => multiLineCodeBlock;
protected virtual MarkdownNode VisitLink(LinkNode link) => link;
protected virtual MarkdownNode VisitEmoji(EmojiNode emoji) => emoji;
protected virtual MarkdownNode VisitMention(MentionNode mention) => mention;
public MarkdownNode Visit(MarkdownNode node) => node switch
{
TextNode text => VisitText(text),
FormattedNode formatted => VisitFormatted(formatted),
InlineCodeBlockNode inlineCodeBlock => VisitInlineCodeBlock(inlineCodeBlock),
MultiLineCodeBlockNode multiLineCodeBlock => VisitMultiLineCodeBlock(multiLineCodeBlock),
LinkNode link => VisitLink(link),
EmojiNode emoji => VisitEmoji(emoji),
MentionNode mention => VisitMention(mention),
_ => throw new ArgumentOutOfRangeException(nameof(node))
};
public void Visit(IEnumerable<MarkdownNode> nodes)
{
foreach (var node in nodes)
Visit(node);
}
}
}

46
DiscordChatExporter.Core/Markdown/Matching/AggregateMatcher.cs Normal file
View File

@@ -0,0 +1,46 @@
using System.Collections.Generic;
namespace DiscordChatExporter.Core.Markdown.Matching
{
internal class AggregateMatcher<T> : IMatcher<T>
{
private readonly IReadOnlyList<IMatcher<T>> _matchers;
public AggregateMatcher(IReadOnlyList<IMatcher<T>> matchers)
{
_matchers = matchers;
}
public AggregateMatcher(params IMatcher<T>[] matchers)
: this((IReadOnlyList<IMatcher<T>>) matchers)
{
}
public ParsedMatch<T>? TryMatch(StringPart stringPart)
{
ParsedMatch<T>? earliestMatch = null;
// Try to match the input with each matcher and get the match with the lowest start index
foreach (var matcher in _matchers)
{
// Try to match
var match = matcher.TryMatch(stringPart);
// If there's no match - continue
if (match == null)
continue;
// If this match is earlier than previous earliest - replace
if (earliestMatch == null || match.StringPart.StartIndex < earliestMatch.StringPart.StartIndex)
earliestMatch = match;
// If the earliest match starts at the very beginning - break,
// because it's impossible to find a match earlier than that
if (earliestMatch.StringPart.StartIndex == stringPart.StartIndex)
break;
}
return earliestMatch;
}
}
}

49
DiscordChatExporter.Core/Markdown/Matching/IMatcher.cs Normal file
View File

@@ -0,0 +1,49 @@
using System;
using System.Collections.Generic;
namespace DiscordChatExporter.Core.Markdown.Matching
{
internal interface IMatcher<T>
{
ParsedMatch<T>? TryMatch(StringPart stringPart);
}
internal static class MatcherExtensions
{
public static IEnumerable<ParsedMatch<T>> MatchAll<T>(this IMatcher<T> matcher,
StringPart stringPart, Func<StringPart, T> transformFallback)
{
// Loop through segments divided by individual matches
var currentIndex = stringPart.StartIndex;
while (currentIndex < stringPart.EndIndex)
{
// Find a match within this segment
var match = matcher.TryMatch(stringPart.Slice(currentIndex, stringPart.EndIndex - currentIndex));
// If there's no match - break
if (match == null)
break;
// If this match doesn't start immediately at current index - transform and yield fallback first
if (match.StringPart.StartIndex > currentIndex)
{
var fallbackPart = stringPart.Slice(currentIndex, match.StringPart.StartIndex - currentIndex);
yield return new ParsedMatch<T>(fallbackPart, transformFallback(fallbackPart));
}
// Yield match
yield return match;
// Shift current index to the end of the match
currentIndex = match.StringPart.StartIndex + match.StringPart.Length;
}
// If EOL wasn't reached - transform and yield remaining part as fallback
if (currentIndex < stringPart.EndIndex)
{
var fallbackPart = stringPart.Slice(currentIndex);
yield return new ParsedMatch<T>(fallbackPart, transformFallback(fallbackPart));
}
}
}
}

15
DiscordChatExporter.Core/Markdown/Matching/ParsedMatch.cs Normal file
View File

@@ -0,0 +1,15 @@
namespace DiscordChatExporter.Core.Markdown.Matching
{
internal class ParsedMatch<T>
{
public StringPart StringPart { get; }
public T Value { get; }
public ParsedMatch(StringPart stringPart, T value)
{
StringPart = stringPart;
Value = value;
}
}
}

40
DiscordChatExporter.Core/Markdown/Matching/RegexMatcher.cs Normal file
View File

@@ -0,0 +1,40 @@
using System;
using System.Text.RegularExpressions;
namespace DiscordChatExporter.Core.Markdown.Matching
{
internal class RegexMatcher<T> : IMatcher<T>
{
private readonly Regex _regex;
private readonly Func<StringPart, Match, T> _transform;
public RegexMatcher(Regex regex, Func<StringPart, Match, T> transform)
{
_regex = regex;
_transform = transform;
}
public RegexMatcher(Regex regex, Func<Match, T> transform)
: this(regex, (p, m) => transform(m))
{
}
public ParsedMatch<T>? TryMatch(StringPart stringPart)
{
var match = _regex.Match(stringPart.Target, stringPart.StartIndex, stringPart.Length);
if (!match.Success)
return null;
// Overload regex.Match(string, int, int) doesn't take the whole string into account,
// it effectively functions as a match check on a substring.
// Which is super weird because regex.Match(string, int) takes the whole input in context.
// So in order to properly account for ^/$ regex tokens, we need to make sure that
// the expression also matches on the bigger part of the input.
if (!_regex.IsMatch(stringPart.Target.Substring(0, stringPart.EndIndex), stringPart.StartIndex))
return null;
var stringPartMatch = stringPart.Slice(match.Index, match.Length);
return new ParsedMatch<T>(stringPartMatch, _transform(stringPartMatch, match));
}
}
}

36
DiscordChatExporter.Core/Markdown/Matching/StringMatcher.cs Normal file
View File

@@ -0,0 +1,36 @@
using System;
namespace DiscordChatExporter.Core.Markdown.Matching
{
internal class StringMatcher<T> : IMatcher<T>
{
private readonly string _needle;
private readonly StringComparison _comparison;
private readonly Func<StringPart, T> _transform;
public StringMatcher(string needle, StringComparison comparison, Func<StringPart, T> transform)
{
_needle = needle;
_comparison = comparison;
_transform = transform;
}
public StringMatcher(string needle, Func<StringPart, T> transform)
: this(needle, StringComparison.Ordinal, transform)
{
}
public ParsedMatch<T>? TryMatch(StringPart stringPart)
{
var index = stringPart.Target.IndexOf(_needle, stringPart.StartIndex, stringPart.Length, _comparison);
if (index >= 0)
{
var stringPartMatch = stringPart.Slice(index, _needle.Length);
return new ParsedMatch<T>(stringPartMatch, _transform(stringPartMatch));
}
return null;
}
}
}

36
DiscordChatExporter.Core/Markdown/Matching/StringPart.cs Normal file
View File

@@ -0,0 +1,36 @@
using System.Text.RegularExpressions;
namespace DiscordChatExporter.Core.Markdown.Matching
{
internal readonly struct StringPart
{
public string Target { get; }
public int StartIndex { get; }
public int Length { get; }
public int EndIndex { get; }
public StringPart(string target, int startIndex, int length)
{
Target = target;
StartIndex = startIndex;
Length = length;
EndIndex = startIndex + length;
}
public StringPart(string target)
: this(target, 0, target.Length)
{
}
public StringPart Slice(int newStartIndex, int newLength) => new(Target, newStartIndex, newLength);
public StringPart Slice(int newStartIndex) => Slice(newStartIndex, EndIndex - newStartIndex);
public StringPart Slice(Capture capture) => Slice(capture.Index, capture.Length);
public override string ToString() => Target.Substring(StartIndex, Length);
}
}