Updated *.pak file format to support both UTF8 and UTF16

tools/data_pack/data_pack.py

Index: tools/data_pack/data_pack.py
diff --git a/tools/data_pack/data_pack.py b/tools/data_pack/data_pack.py
index ffd2cb2b0aa8639f52a4ea6ac5297f52a11a8a80..7059dea47b427ef3c0f917af74339940f564f4f4 100755
--- a/tools/data_pack/data_pack.py
+++ b/tools/data_pack/data_pack.py
@@ -9,8 +9,12 @@ See base/pack_file* for details.
 
 import struct
 
-FILE_FORMAT_VERSION = 3
-HEADER_LENGTH = 2 * 4  # Two uint32s. (file version and number of entries)
+FILE_FORMAT_VERSION = 4
+HEADER_LENGTH = 3 * 4  # Three uint32s. (file version, number of entries and
+                       # encoding of text resources)

[tony, 2011/08/29 21:24:39]

Can we use something smaller than an uint32 for the encoding?  Maybe an int16 or
int8 (ie., unsigned char)?

[marcvs, 2011/08/30 07:58:55]

Changed to uint8 (== unsigned char).

+BINARY = 0
+UTF8 = 1
+UTF16 = 2

[tony, 2011/08/29 21:24:39]

BINARY, UTF8, UTF16 = range(3)

[marcvs, 2011/08/30 07:58:55]

Done. (also in the other files)

 
 class WrongFileVersion(Exception):
   pass
@@ -21,13 +25,14 @@ def ReadDataPack(input_file):
   original_data = data
 
   # Read the header.
-  version, num_entries = struct.unpack("<II", data[:HEADER_LENGTH])
+  version, num_entries, encoding = struct.unpack("<III", data[:HEADER_LENGTH])
   if version != FILE_FORMAT_VERSION:
+    print "Wrong file version in ", input_file
     raise WrongFileVersion
 
   resources = {}
   if num_entries == 0:
-    return resources
+    return resources, encoding
 
   # Read the index and data.
   data = data[HEADER_LENGTH:]
@@ -38,15 +43,15 @@ def ReadDataPack(input_file):
     next_id, next_offset = struct.unpack("<HI", data[:kIndexEntrySize])
     resources[id] = original_data[offset:next_offset]
 
-  return resources
+  return resources, encoding
 
-def WriteDataPack(resources, output_file):
+def WriteDataPack(resources, output_file, encoding):
   """Write a map of id=>data into output_file as a data pack."""
   ids = sorted(resources.keys())
   file = open(output_file, "wb")
 
   # Write file header.
-  file.write(struct.pack("<II", FILE_FORMAT_VERSION, len(ids)))
+  file.write(struct.pack("<III", FILE_FORMAT_VERSION, len(ids), encoding))
 
   # Each entry is a uint16 and a uint32. We have one extra entry for the last
   # item.
@@ -67,9 +72,9 @@ def WriteDataPack(resources, output_file):
 def main():
   # Just write a simple file.
   data = { 1: "", 4: "this is id 4", 6: "this is id 6", 10: "" }
-  WriteDataPack(data, "datapack1.pak")
+  WriteDataPack(data, "datapack1.pak", UTF8)
   data2 = { 1000: "test", 5: "five" }
-  WriteDataPack(data2, "datapack2.pak")
+  WriteDataPack(data2, "datapack2.pak", UTF8)
   print "wrote datapack1 and datapack2 to current directory."
 
 if __name__ == '__main__':