Add some details about verification to the manual.

1 files changed, 59 insertions, 0 deletions

diff --git a/doc/manual/verifier.py b/doc/manual/verifier.py
new file mode 100644
index 000000000..877a118a1
--- /dev/null
+++ b/doc/manual/verifier.py
@@ -0,0 +1,59 @@
+#!/usr/bin/python3
+
+from pathlib import Path
+import re
+import sys
+import subprocess
+
+if len(sys.argv) < 3:
+    print(f"Syntax: {sys.argv[0]} <path-to-libdcp-source-tree> <ERROR|BV21_ERROR|WARNING>")
+    sys.exit(1)
+
+libdcp = Path(sys.argv[1])
+type = sys.argv[2]
+header = libdcp / "src" / "verify.h"
+
+types = ("BV21_ERROR", "ERROR", "WARNING")
+
+def find_type(name):
+    """
+    Search source code to find where a given code is used and hence find out whether
+    it represents an error, Bv2.1 "error" or warning.
+    """
+    previous = ''
+    for source in ["verify_j2k.cc", "dcp.cc", "verify.cc"]:
+        path = libdcp / "src" / source
+        with open(path) as s:
+            for line in s:
+                if line.find(name) != -1:
+                    line_with_previous = previous + line
+                    for t in types:
+                        if line_with_previous.find(t) != -1:
+                            return t
+                    assert False
+                previous = line
+
+
+print('<itemizedlist>')
+
+active = False
+with open(header) as h:
+    for line in h:
+        strip = line.strip()
+        if strip == "enum class Code {":
+            active = True
+        elif strip == "};":
+            active = False
+        elif active:
+            if strip.startswith('/**'):
+                text = strip.replace('/**', '').replace('*/', '').strip()
+            elif not strip.startswith('/*') and not strip.startswith('*') and strip.endswith(','):
+                this_type = find_type(strip[:-1])
+                if this_type == type:
+                    text = re.sub(r"\[.*?\]", lambda m: f'(Bv2.1 {m[0][7:-1]})', text)
+                    text = text.replace('<', '&lt;')
+                    text = text.replace('>', '&gt;')
+                    text = re.sub(r"_(.*?)_", r"<code>\1</code>", text)
+                    print(f'<listitem>{text}.</listitem>')
+
+print('</itemizedlist>')