Upgrade isolate_tests.py to support extracting code from docs

1 files changed, 48 insertions, 3 deletions

diff --git a/scripts/isolate_tests.py b/scripts/isolate_tests.py
index a1d1c75c..d075b07f 100755
--- a/scripts/isolate_tests.py
+++ b/scripts/isolate_tests.py
@@ -1,6 +1,6 @@
 #!/usr/bin/python
 #
-# This script reads C++ source files and writes all
+# This script reads C++ or RST source files and writes all
 # multi-line strings into individual files.
 # This can be used to extract the Solidity test cases
 # into files for e.g. fuzz testing as
@@ -12,7 +12,7 @@ import os
 import hashlib
 from os.path import join
 
-def extract_cases(path):
+def extract_test_cases(path):
     lines = open(path, 'rb').read().splitlines()
 
     inside = False
@@ -34,6 +34,44 @@ def extract_cases(path):
 
     return tests
 
+# Contract sources are indented by 4 spaces.
+# Look for `pragma solidity` and abort a line not indented properly.
+# If the comment `// This will not compile` is above the pragma,
+# the code is skipped.
+def extract_docs_cases(path):
+    # Note: this code works, because splitlines() removes empty new lines
+    #       and thus even if the empty new lines are missing indentation
+    lines = open(path, 'rb').read().splitlines()
+
+    ignore = False
+    inside = False
+    tests = []
+
+    for l in lines:
+      if inside:
+        # Abort if indentation is missing
+        m = re.search(r'^[^ ]+', l)
+        if m:
+          inside = False
+        else:
+          tests[-1] += l + '\n'
+      else:
+        m = re.search(r'^    // This will not compile', l)
+        if m:
+          ignore = True
+
+        if ignore:
+          # Abort if indentation is missing
+          m = re.search(r'^[^ ]+', l)
+          if m:
+            ignore = False
+        else:
+          m = re.search(r'^    pragma solidity .*[0-9]+\.[0-9]+\.[0-9]+;$', l)
+          if m:
+            inside = True
+            tests += [l]
+
+    return tests
 
 def write_cases(tests):
     for test in tests:
@@ -41,8 +79,15 @@ def write_cases(tests):
 
 if __name__ == '__main__':
     path = sys.argv[1]
+    docs = False
+    if len(sys.argv) > 2 and sys.argv[2] == 'docs':
+      docs = True
 
     for root, dir, files in os.walk(path):
         for f in files:
-            cases = extract_cases(join(root, f))
+            path = join(root, f)
+            if docs:
+              cases = extract_docs_cases(path)
+            else:
+              cases = extract_test_cases(path)
             write_cases(cases)