Correctly deal with multiple links on one line (#30)

* 🐞 Fix greedy matching of [[]] links. * 🐞 Fix greedy matching of []() links.
BasilPH · Jul 12, 2020 · dc6888b · dc6888b
1 parent dc0283c
commit dc6888b
Show file tree

Hide file tree

Showing 7 changed files with 15 additions and 12 deletions.
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "vizel"
-version = "0.2.0"
+version = "0.2.1"
 description = "Vizualise a Zettelkasten"
 authors = ["Basil Philipp <basil@interdimensional-television.com>"]
 license = "GPL-3.0-only"

diff --git a/tests/data/zettelkasten_md/03272020061037-electrodermal-activity.md b/tests/data/zettelkasten_md/03272020061037-electrodermal-activity.md
@@ -31,4 +31,7 @@ nocite: |
 ---
 ---
 
+Here we have to links on one line to test the regex. [A first link](202002251025_This_is_the_first_test_zettel.md) and a [second line](202003211727_This_is_the_second_test_zettel.md). 
+
+
 ### **References:**
diff --git a/tests/data/zettelkasten_md/202002251025_This_is_the_first_test_zettel.md b/tests/data/zettelkasten_md/202002251025_This_is_the_first_test_zettel.md
@@ -1,9 +1,7 @@
 # 202002251025 This is the first test Zettel
 tags: #artificialintelligence #zettelkasten
 
-This Zettel references this other Zettel here[[202002241029]].
-
-It points to the second Zettel [[202003211727]].
+This Zettel references this other Zettel here[[202002241029]]. It points to the second Zettel [[202003211727]].
 
 
 ## It also has subtitles, fancy

diff --git a/tests/data/zettelkasten_txt/03272020061037-electrodermal-activity.txt b/tests/data/zettelkasten_txt/03272020061037-electrodermal-activity.txt
@@ -31,4 +31,6 @@ nocite: |
 ---
 ---
 
+Here we have to links on one line to test the regex. [A first link](202002251025_This_is_the_first_test_zettel.txt) and a [second line](202003211727_This_is_the_second_test_zettel.txt).
+
 ### **References:**
diff --git a/tests/data/zettelkasten_txt/202002251025_This_is_the_first_test_zettel.txt b/tests/data/zettelkasten_txt/202002251025_This_is_the_first_test_zettel.txt
@@ -1,9 +1,7 @@
 # 202002251025 This is the first test Zettel
 tags: #artificialintelligence #zettelkasten
 
-This Zettel references this other Zettel here[[202002241029]].
-
-It points to the second Zettel [[202003211727]].
+This Zettel references this other Zettel here[[202002241029]]. It points to the second Zettel [[202003211727]].
 
 
 ## It also has subtitles, fancy

diff --git a/tests/test_vizel.py b/tests/test_vizel.py
@@ -47,9 +47,9 @@ def test_stats(zettelkasten_directory, stderr_expected):
     assert result.exit_code == 0
     stdout_output = (
         '7 Zettel\n'
-        '4 references between Zettel\n'
+        '6 references between Zettel\n'
         '2 Zettel with no references\n'
-        '4 connected components\n'
+        '3 connected components\n'
     )
     assert result.stdout == stdout_output
 

diff --git a/vizel/cli.py b/vizel/cli.py
@@ -68,11 +68,13 @@ def _load_references(zettel_path, zettel_directory_path):
     zettel_filenames = sorted(
         [os.path.basename(f) for f in glob.glob(os.path.join(zettel_directory_path, '*[.md|.txt]'))])
 
-    # Extract references for the [[]] link format
-    references += _extract_valid_references('\[\[(.+)\]\]', zettel_path, zettel_filenames)
+    # Extract references for the [[ID]] link format
+    # Look for [[, and then match anything that isn't ]]. End with ]].
+    references += _extract_valid_references('\[\[([^\]\]]+)\]\]', zettel_path, zettel_filenames)
 
     # Extract references for the markdown link format
-    references += _extract_valid_references('\[.*\]\((.+)\)', zettel_path, zettel_filenames)
+    # Look for [, and then match anything that isn't ]. Then look for ( and match anything that isn't ). End with ).
+    references += _extract_valid_references('\[[^\]]+\]\(([^\)]+)\)', zettel_path, zettel_filenames)
 
     return references