From b1498484e8c7575368e4981e14243d080e79adfb Mon Sep 17 00:00:00 2001 From: Martin Dvorak Date: Fri, 5 Jan 2024 22:15:37 +0100 Subject: [PATCH] Blacklisting http and https in autolinking. --- lib/src/mind/ai/autolinking/autolinking_mind.cpp | 9 +++++++++ ...cmark_aho_corasick_block_autolinking_preprocessor.cpp | 3 ++- 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/lib/src/mind/ai/autolinking/autolinking_mind.cpp b/lib/src/mind/ai/autolinking/autolinking_mind.cpp index 701b3584..c2b595b1 100644 --- a/lib/src/mind/ai/autolinking/autolinking_mind.cpp +++ b/lib/src/mind/ai/autolinking/autolinking_mind.cpp @@ -76,6 +76,15 @@ void AutolinkingMind::updateTrieIndex() addThingToTrie(n); } + // remove functional blacklist + vector blacklist = { + "http", + "https", + }; + for_each(blacklist.begin(), blacklist.end(), [this](const string& s) { + trie->removeWord(s); + }); + // IMPROVE: add also tags #ifdef DO_MF_DEBUG diff --git a/lib/src/mind/ai/autolinking/cmark_aho_corasick_block_autolinking_preprocessor.cpp b/lib/src/mind/ai/autolinking/cmark_aho_corasick_block_autolinking_preprocessor.cpp index 492061da..16012c37 100644 --- a/lib/src/mind/ai/autolinking/cmark_aho_corasick_block_autolinking_preprocessor.cpp +++ b/lib/src/mind/ai/autolinking/cmark_aho_corasick_block_autolinking_preprocessor.cpp @@ -269,7 +269,8 @@ void injectThingsLinks(cmark_node* srcNode, Mind& mind) * Preprocessor. */ -const string CmarkAhoCorasickBlockAutolinkingPreprocessor::TRAILING_CHARS = string{" \t,:;.!?<>{}&()-+/*\\_=%~#$^[]'\""}; +const string CmarkAhoCorasickBlockAutolinkingPreprocessor::TRAILING_CHARS + = string{" \t,:;.!?<>{}&()-+/*\\_=%~#$^[]'\""}; CmarkAhoCorasickBlockAutolinkingPreprocessor::CmarkAhoCorasickBlockAutolinkingPreprocessor(Mind& mind) : AutolinkingPreprocessor{mind}