diff --git a/tests/cluecode/data/copyrights/AliasDotCom_Website.html b/tests/cluecode/data/copyrights/AliasDotCom_Website.html
new file mode 100644
index 00000000000..bb26b5763a2
--- /dev/null
+++ b/tests/cluecode/data/copyrights/AliasDotCom_Website.html
@@ -0,0 +1,21 @@
+<h1>Disclaimer for AliasDotCom Ltd.</h1>
+
+<p>Copyright (C) AliasDotCom Ltd. All rights reserved.</p>
+
+<p>If you require any more information or have any questions about our site's disclaimer, please feel free to contact us by email at noone@example.com.</p>
+
+<h2>Disclaimers for aliasdotcom</h2>
+
+<p>All the information on this website - https://www.aliasdotcom.com - is published in good faith and for general information purpose only. aliasdotcom does not make any warranties about the completeness, reliability and accuracy of this information. Any action you take upon the information you find on this website (aliasdotcom), is strictly at your own risk. aliasdotcom will not be liable for any losses and/or damages in connection with the use of our website.</p>
+
+<p>From our website, you can visit other websites by following hyperlinks to such external sites. While we strive to provide only quality links to useful and ethical websites, we have no control over the content and nature of these sites. These links to other websites do not imply a recommendation for all the content found on these sites. Site owners and content may change without notice and may occur before we have the opportunity to remove a link which may have gone 'bad'.</p>
+
+<p>Please be also aware that when you leave our website, other sites may have different privacy policies and terms which are beyond our control. Please be sure to check the Privacy Policies of these sites as well as their "Terms of Service" before engaging in any business or uploading any information.</p>
+
+<h2>Consent</h2>
+
+<p>By using our website, you hereby consent to our disclaimer and agree to its terms.</p>
+
+<h2>Update</h2>
+
+<p>Should we update, amend or make any changes to this document, those changes will be prominently posted here.</p>
diff --git a/tests/cluecode/data/copyrights/AliasDotCom_Website.html.yml b/tests/cluecode/data/copyrights/AliasDotCom_Website.html.yml
new file mode 100644
index 00000000000..aec5a9b9319
--- /dev/null
+++ b/tests/cluecode/data/copyrights/AliasDotCom_Website.html.yml
@@ -0,0 +1,9 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copyright (c) AliasDotCom Ltd.
+
+holders:
+  - AliasDotCom Ltd.
diff --git a/tests/cluecode/data/copyrights/alphanumeric_holder_name.txt b/tests/cluecode/data/copyrights/alphanumeric_holder_name.txt
new file mode 100644
index 00000000000..c997d2c7968
--- /dev/null
+++ b/tests/cluecode/data/copyrights/alphanumeric_holder_name.txt
@@ -0,0 +1 @@
+Copyright (c) corps69
diff --git a/tests/cluecode/data/copyrights/alphanumeric_holder_name.txt.yml b/tests/cluecode/data/copyrights/alphanumeric_holder_name.txt.yml
new file mode 100644
index 00000000000..ed048a797c9
--- /dev/null
+++ b/tests/cluecode/data/copyrights/alphanumeric_holder_name.txt.yml
@@ -0,0 +1,3 @@
+what:
+  - copyrights
+  - holders
diff --git a/tests/cluecode/data/copyrights/copr.txt b/tests/cluecode/data/copyrights/copr.txt
new file mode 100644
index 00000000000..8ebfc55c002
--- /dev/null
+++ b/tests/cluecode/data/copyrights/copr.txt
@@ -0,0 +1 @@
+copr. (c) Foobar Pvt. Ltd.
diff --git a/tests/cluecode/data/copyrights/copr.txt.yml b/tests/cluecode/data/copyrights/copr.txt.yml
new file mode 100644
index 00000000000..104c5442320
--- /dev/null
+++ b/tests/cluecode/data/copyrights/copr.txt.yml
@@ -0,0 +1,19 @@
+what:
+  - copyrights
+  - holders
+  - copyrights_summary
+  - holders_summary
+
+copyrights:
+  - copr. (c) Foobar Pvt. Ltd.
+
+holders:
+  - Foobar Pvt. Ltd.
+
+copyrights_summary:
+  - value: copr. (c) Foobar Pvt. Ltd.
+    count: 1
+
+holders_summary:
+  - value: Foobar Pvt. Ltd.
+    count: 1
diff --git a/tests/cluecode/data/copyrights/copyright_in_docstring.py b/tests/cluecode/data/copyrights/copyright_in_docstring.py
new file mode 100644
index 00000000000..1968fca34d2
--- /dev/null
+++ b/tests/cluecode/data/copyrights/copyright_in_docstring.py
@@ -0,0 +1,3 @@
+"""
+Copr. (c) 1999 Random Corp. Ltd.
+"""
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/copyright_in_docstring.py.yml b/tests/cluecode/data/copyrights/copyright_in_docstring.py.yml
new file mode 100644
index 00000000000..8ae7011edb8
--- /dev/null
+++ b/tests/cluecode/data/copyrights/copyright_in_docstring.py.yml
@@ -0,0 +1,9 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copr. (c) 1999 Random Corp. Ltd.
+
+holders:
+  - Random Corp. Ltd.
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/copyright_without_icon.txt b/tests/cluecode/data/copyrights/copyright_without_icon.txt
new file mode 100644
index 00000000000..69111159c5b
--- /dev/null
+++ b/tests/cluecode/data/copyrights/copyright_without_icon.txt
@@ -0,0 +1 @@
+Copyright Foobar Pvt. Ltd.. All rights reserved.
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/copyright_without_icon.txt.yml b/tests/cluecode/data/copyrights/copyright_without_icon.txt.yml
new file mode 100644
index 00000000000..83bf8978f5a
--- /dev/null
+++ b/tests/cluecode/data/copyrights/copyright_without_icon.txt.yml
@@ -0,0 +1,19 @@
+what:
+  - copyrights
+  - holders
+  - copyrights_summary
+  - holders_summary
+
+copyrights:
+  - Copyright Foobar Pvt. Ltd
+
+holders:
+  - Foobar Pvt. Ltd
+
+copyrights_summary:
+  - value: Copyright Foobar Pvt. Ltd
+    count: 1
+
+holders_summary:
+  - value: Foobar Pvt. Ltd
+    count: 1
diff --git a/tests/cluecode/data/copyrights/gibberish_holder_name.c b/tests/cluecode/data/copyrights/gibberish_holder_name.c
new file mode 100644
index 00000000000..e1e57eff270
--- /dev/null
+++ b/tests/cluecode/data/copyrights/gibberish_holder_name.c
@@ -0,0 +1,3 @@
+/*
+ * Copyright (C) %34hkj12iaksd2@//0
+ */
diff --git a/tests/cluecode/data/copyrights/gibberish_holder_name.c.yml b/tests/cluecode/data/copyrights/gibberish_holder_name.c.yml
new file mode 100644
index 00000000000..587e47039f9
--- /dev/null
+++ b/tests/cluecode/data/copyrights/gibberish_holder_name.c.yml
@@ -0,0 +1,5 @@
+what:
+  - copyrights
+  - holders
+  - copyrights_summary
+  - holders_summary
diff --git a/tests/cluecode/data/copyrights/gibberish_holdername.c b/tests/cluecode/data/copyrights/gibberish_holdername.c
new file mode 100644
index 00000000000..566bede73ca
--- /dev/null
+++ b/tests/cluecode/data/copyrights/gibberish_holdername.c
@@ -0,0 +1,3 @@
+/*
+ * Copyright (c) Micr0s?3t4ii2\ew
+ */
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/gibberish_holdername.c.yml b/tests/cluecode/data/copyrights/gibberish_holdername.c.yml
new file mode 100644
index 00000000000..7190db99743
--- /dev/null
+++ b/tests/cluecode/data/copyrights/gibberish_holdername.c.yml
@@ -0,0 +1,3 @@
+what:
+  - copyrights
+  - holders
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/gibberish_with_real_copyright.txt b/tests/cluecode/data/copyrights/gibberish_with_real_copyright.txt
new file mode 100644
index 00000000000..7f8e9f7f3ef
--- /dev/null
+++ b/tests/cluecode/data/copyrights/gibberish_with_real_copyright.txt
@@ -0,0 +1 @@
+(c) adfe??+_93\ee\adfds\/ad<ds>Copyright(c) Microsoft
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/gibberish_with_real_copyright.txt.yml b/tests/cluecode/data/copyrights/gibberish_with_real_copyright.txt.yml
new file mode 100644
index 00000000000..db1a04dbb54
--- /dev/null
+++ b/tests/cluecode/data/copyrights/gibberish_with_real_copyright.txt.yml
@@ -0,0 +1,9 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copyright (c) Microsoft
+
+holders:
+  - Microsoft
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/in_docstring.py b/tests/cluecode/data/copyrights/in_docstring.py
new file mode 100644
index 00000000000..82a6223fdd6
--- /dev/null
+++ b/tests/cluecode/data/copyrights/in_docstring.py
@@ -0,0 +1,3 @@
+"""
+© 1989-2000 Baz Corporation. All rights reserved.
+"""
diff --git a/tests/cluecode/data/copyrights/in_docstring.py.yml b/tests/cluecode/data/copyrights/in_docstring.py.yml
new file mode 100644
index 00000000000..b2f9504a786
--- /dev/null
+++ b/tests/cluecode/data/copyrights/in_docstring.py.yml
@@ -0,0 +1,10 @@
+what:
+  - copyrights
+  - holders
+  - authors
+
+copyrights:
+  - (c) 1989-2000 Baz Corporation
+
+holders:
+  - Baz Corporation
diff --git a/tests/cluecode/data/copyrights/invalid_copyright_year.py b/tests/cluecode/data/copyrights/invalid_copyright_year.py
new file mode 100644
index 00000000000..127f149d5f2
--- /dev/null
+++ b/tests/cluecode/data/copyrights/invalid_copyright_year.py
@@ -0,0 +1,4 @@
+# Copyright (C) RandomCompany 2990-2999
+
+def fn():
+    pass
diff --git a/tests/cluecode/data/copyrights/invalid_copyright_year.py.yml b/tests/cluecode/data/copyrights/invalid_copyright_year.py.yml
new file mode 100644
index 00000000000..dcc72bbcc8b
--- /dev/null
+++ b/tests/cluecode/data/copyrights/invalid_copyright_year.py.yml
@@ -0,0 +1,15 @@
+what:
+  - copyrights
+  - holders
+  - copyrights_summary
+
+copyrights:
+  - Copyright (c) RandomCompany
+
+holders:
+  - RandomCompany
+
+copyrights_summary:
+  - value: Copyright (c) RandomCompany
+    count: 1
+
diff --git a/tests/cluecode/data/copyrights/long_holder_name.txt b/tests/cluecode/data/copyrights/long_holder_name.txt
new file mode 100644
index 00000000000..5c36c3db229
--- /dev/null
+++ b/tests/cluecode/data/copyrights/long_holder_name.txt
@@ -0,0 +1 @@
+Copyright (c) 2000-2021 Adolph Blaine Charles David Earl Frederick Gerald Hubert Irvin John Kenneth Lloyd Martin Nero Oliver Paul Quincy Randolph Sherman Thomas Uncas Victor William Xerxes Yancy Zeus Wolfeschlegelsteinhausenbergerdorffwelchevoralternwarengewissenhaftschaferswessenschafewarenwohlgepflegeundsorgfaltigkeitbeschutzenvonangreifendurchihrraubgierigfeindewelchevoralternzwolftausendjahresvorandieerscheinenvanderersteerdemenschderraumschiffgebrauchlichtalsseinursprungvonkraftgestartseinlangefahrthinzwischensternartigraumaufdersuchenachdiesternwelchegehabtbewohnbarplanetenkreisedrehensichundwohinderneurassevonverstandigmenschlichkeitkonntefortpflanzenundsicherfreuenanlebenslanglichfreudeundruhemitnichteinfurchtvorangreifenvonandererintelligentgeschopfsvonhinzwischensternartigraum
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/long_holder_name.txt.yml b/tests/cluecode/data/copyrights/long_holder_name.txt.yml
new file mode 100644
index 00000000000..8784bc6c3eb
--- /dev/null
+++ b/tests/cluecode/data/copyrights/long_holder_name.txt.yml
@@ -0,0 +1,14 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copyright (c) 2000-2021 Adolph Blaine Charles David Earl Frederick Gerald Hubert Irvin John Kenneth Lloyd Martin Nero Oliver Paul Quincy Randolph Sherman Thomas Uncas Victor William Xerxes Yancy Zeus
+holders:
+  - Adolph Blaine Charles David Earl Frederick Gerald Hubert Irvin John Kenneth Lloyd Martin Nero Oliver Paul Quincy Randolph Sherman Thomas Uncas Victor William Xerxes Yancy Zeus
+
+expected_failures:
+  - copyrights
+  - holders
+notes: we should not detect very long names
+  
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/long_name.txt b/tests/cluecode/data/copyrights/long_name.txt
new file mode 100644
index 00000000000..1452c4af7fe
--- /dev/null
+++ b/tests/cluecode/data/copyrights/long_name.txt
@@ -0,0 +1 @@
+Copyright (c) 2021 Adolph Blaine Charles David Earl Frederick Gerald Hubert Irvin John Kenneth Lloyd Martin Nero Oliver Paul Quincy Randolph Sherman Thomas Uncas Victor William Xerxes Yancy Zeus Wolfeschlegelsteinhausenbergerdorffwelchevoralternwarengewissenhaftschaferswessenschafewarenwohlgepflegeundsorgfaltigkeitbeschutzenvonangreifendurchihrraubgierigfeindewelchevoralternzwolftausendjahresvorandieerscheinenvanderersteerdemenschderraumschiffgebrauchlichtalsseinursprungvonkraftgestartseinlangefahrthinzwischensternartigraumaufdersuchenachdiesternwelchegehabtbewohnbarplanetenkreisedrehensichundwohinderneurassevonverstandigmenschlichkeitkonntefortpflanzenundsicherfreuenanlebenslanglichfreudeundruhemitnichteinfurchtvorangreifenvonandererintelligentgeschopfsvonhinzwischensternartigraum
diff --git a/tests/cluecode/data/copyrights/long_name.txt.yml b/tests/cluecode/data/copyrights/long_name.txt.yml
new file mode 100644
index 00000000000..130a8510207
--- /dev/null
+++ b/tests/cluecode/data/copyrights/long_name.txt.yml
@@ -0,0 +1,14 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copyright (c) 2021 Adolph Blaine Charles David Earl Frederick Gerald Hubert Irvin John Kenneth Lloyd Martin Nero Oliver Paul Quincy Randolph Sherman Thomas Uncas Victor William Xerxes Yancy Zeus
+
+holders:
+  - Adolph Blaine Charles David Earl Frederick Gerald Hubert Irvin John Kenneth Lloyd Martin Nero Oliver Paul Quincy Randolph Sherman Thomas Uncas Victor William Xerxes Yancy Zeus
+
+expected_failures:
+  - copyrights
+  - holders
+notes: we should not detect very long names
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/multiple_copyrights.cpp b/tests/cluecode/data/copyrights/multiple_copyrights.cpp
new file mode 100644
index 00000000000..2c2b703bd95
--- /dev/null
+++ b/tests/cluecode/data/copyrights/multiple_copyrights.cpp
@@ -0,0 +1,6 @@
+/*
+ * Copyright(C) Microsoft. All rights reserved.
+ * Copyright(c) 2000-2001 Alias Ltd. All rights reserved. This is just an example corporation
+ * named Alias
+ * Copyright (c) Microsoft.
+ */
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/multiple_copyrights.cpp.yml b/tests/cluecode/data/copyrights/multiple_copyrights.cpp.yml
new file mode 100644
index 00000000000..7b4ed5437d9
--- /dev/null
+++ b/tests/cluecode/data/copyrights/multiple_copyrights.cpp.yml
@@ -0,0 +1,20 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copyright (c) Microsoft
+  - Copyright (c) 2000-2001 Alias Ltd.
+  - Copyright (c) Microsoft
+
+holders:
+  - Microsoft
+  - Alias Ltd.
+  - Microsoft
+
+copyrights_summary:
+  - value: Copyright (c) Microsoft
+    count: 2
+
+  - value: Copyright (c) 2000-2001 Alias Ltd.
+    count: 1
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/multiple_copyrights.txt b/tests/cluecode/data/copyrights/multiple_copyrights.txt
new file mode 100644
index 00000000000..399b09fb988
--- /dev/null
+++ b/tests/cluecode/data/copyrights/multiple_copyrights.txt
@@ -0,0 +1,4 @@
+\\ewasdfasd\Copyright(c) Microsoft
+\\'''?0Copyright(c) Adobe 1999'''
+Copyright(c)Microsoft
+weasdfkjadfadflkjl
diff --git a/tests/cluecode/data/copyrights/multiple_copyrights.txt.yml b/tests/cluecode/data/copyrights/multiple_copyrights.txt.yml
new file mode 100644
index 00000000000..034e745fb41
--- /dev/null
+++ b/tests/cluecode/data/copyrights/multiple_copyrights.txt.yml
@@ -0,0 +1,20 @@
+what:
+  - copyrights
+  - holders
+  - copyrights_summary
+
+copyrights:
+  - Copyright (c) Microsoft
+  - (c) Adobe 1999
+  - Copyright (c) Microsoft
+
+holders:
+  - Microsoft
+  - Adobe
+  - Microsoft
+
+copyrights_summary:
+  - value: Copyright (c) Microsoft
+    count: 2
+  - value: (c) Adobe
+    count: 1
diff --git a/tests/cluecode/data/copyrights/no_punctuation_after_holder_name.txt b/tests/cluecode/data/copyrights/no_punctuation_after_holder_name.txt
new file mode 100644
index 00000000000..23bc582b098
--- /dev/null
+++ b/tests/cluecode/data/copyrights/no_punctuation_after_holder_name.txt
@@ -0,0 +1 @@
+(c) 1990-2005 Microsoft All rights reserved
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/no_punctuation_after_holder_name.txt.yml b/tests/cluecode/data/copyrights/no_punctuation_after_holder_name.txt.yml
new file mode 100644
index 00000000000..7eb276d922d
--- /dev/null
+++ b/tests/cluecode/data/copyrights/no_punctuation_after_holder_name.txt.yml
@@ -0,0 +1,9 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - (c) 1990-2005 Microsoft
+
+holders:
+  - Microsoft
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/unlimited_copr.html b/tests/cluecode/data/copyrights/unlimited_copr.html
new file mode 100644
index 00000000000..bc1a61b1fdf
--- /dev/null
+++ b/tests/cluecode/data/copyrights/unlimited_copr.html
@@ -0,0 +1,21 @@
+<h1>Disclaimer for Unlimited Ltd.</h1>
+
+<p>Copyright (c) Unlimited Ltd.</p>
+
+<p>If you require any more information or have any questions about our site's disclaimer, please feel free to contact us by email at unlimited.ceo@unlimited.com.</p>
+
+<h2>Disclaimers for unlimited</h2>
+
+<p>All the information on this website - https://www.unlimited.com - is published in good faith and for general information purpose only. unlimited does not make any warranties about the completeness, reliability and accuracy of this information. Any action you take upon the information you find on this website (unlimited), is strictly at your own risk. unlimited will not be liable for any losses and/or damages in connection with the use of our website.</p>
+
+<p>From our website, you can visit other websites by following hyperlinks to such external sites. While we strive to provide only quality links to useful and ethical websites, we have no control over the content and nature of these sites. These links to other websites do not imply a recommendation for all the content found on these sites. Site owners and content may change without notice and may occur before we have the opportunity to remove a link which may have gone 'bad'.</p>
+
+<p>Please be also aware that when you leave our website, other sites may have different privacy policies and terms which are beyond our control. Please be sure to check the Privacy Policies of these sites as well as their "Terms of Service" before engaging in any business or uploading any information.</p>
+
+<h2>Consent</h2>
+
+<p>By using our website, you hereby consent to our disclaimer and agree to its terms.</p>
+
+<h2>Update</h2>
+
+<p>Should we update, amend or make any changes to this document, those changes will be prominently posted here.</p>
\ No newline at end of file
diff --git a/tests/cluecode/data/copyrights/unlimited_copr.html.yml b/tests/cluecode/data/copyrights/unlimited_copr.html.yml
new file mode 100644
index 00000000000..36f9716e9f2
--- /dev/null
+++ b/tests/cluecode/data/copyrights/unlimited_copr.html.yml
@@ -0,0 +1,9 @@
+what:
+  - copyrights
+  - holders
+
+copyrights:
+  - Copyright (c) Unlimited Ltd.
+
+holders:
+  - Unlimited Ltd.
\ No newline at end of file