Ignore markdown codeblocks for tags (#66)

* Ignore markdown codeblocks for tags

* Changed `content_ex_tags` to use `content` instead of `content_ex_codeblock`

* Bumped `INDEX_SCHEMA_VERSION`

diff --git a/flatnotes/flatnotes.py b/flatnotes/flatnotes.py
index 782387b7323196b199f552326a96cc315925a769..c904a7906616096a36f0db9d5911cf11a7910403 100644 (file)
--- a/flatnotes/flatnotes.py
+++ b/flatnotes/flatnotes.py
@@ -20,7 +20,7 @@ from helpers import empty_dir, re_extract, strip_ext
 from logger import logger
 
 MARKDOWN_EXT = ".md"
-INDEX_SCHEMA_VERSION = "3"
+INDEX_SCHEMA_VERSION = "4"
 
 StemmingFoldingAnalyzer = StemmingAnalyzer() | CharsetFilter(accent_map)
 
@@ -162,6 +162,7 @@ class SearchResult(Note):
 
 class Flatnotes(object):
     TAGS_RE = re.compile(r"(?:(?<=^#)|(?<=\s#))\w+(?=\s|$)")
+    CODEBLOCK_RE = re.compile(r"`{1,3}.*?`{1,3}", re.DOTALL)
     TAGS_WITH_HASH_RE = re.compile(r"(?:(?<=^)|(?<=\s))#\w+(?=\s|$)")
 
     def __init__(self, dir: str) -> None:
@@ -203,7 +204,9 @@ class Flatnotes(object):
 
         - The content without the tags.
         - A set of tags converted to lowercase."""
-        content_ex_tags, tags = re_extract(cls.TAGS_RE, content)
+        content_ex_codeblock = re.sub(cls.CODEBLOCK_RE, '', content)
+        _, tags = re_extract(cls.TAGS_RE, content_ex_codeblock)
+        content_ex_tags, _ = re_extract(cls.TAGS_RE, content)
         try:
             tags = [tag.lower() for tag in tags]
             return (content_ex_tags, set(tags))