Abstract
This post of my blog responds to a request by a customer to find and extract a VBA macro in a PowerPoint file, specifically one stored in the older binary file format, also known to many as BFF.
Introduction
This post will take you through steps outlined in [MS-PPT] “PowerPoint (.ppt) Binary File Format” section 2.1.2 “PowerPoint Document Stream”. In this section is an algorithm for finding various important structures and objects in the PowerPoint format. Because I only needed the VbaProjectStgCompressedAtom object, specified in 2.10.42, which contains the storage for the VBA macro, I didn’t need to go through the entire algorithm. Instead, the approach I used was to look for the part of the algorithm which most directly leads to the object (or its immediate parent), and then work backward in the steps. This is not actually that hard to do, but care must when
moving backward in the algorithm not to miss important prerequisite structures.
Because of this, it is also important for me to note that taking this approach is not a substitute for reading the specification and understanding the overall structure of a PowerPoint document format.
Also, what you see here is an example process based on one PowerPoint document that was created specifically for this exercise.
The document used in this example was created in PowerPoint 2013 by adding a VBA macro, placing password protection on it, and then saving it as PowerPoint 97-2003 format (i.e. BFF). Below are screen shots for this first part.
![]()
Figure 1.0
The Developer tab you see in Figure 1.0, can be accessed after going to File | Options…, selecting Customize Ribbon and checking the Developer box on the right like in Figure 1.1.
![]()
Figure 1.1
Once you have the Developer tab, click the Visual Basic button and you will be able to add code like in Figure 1.2.
![]()
Figure 1.2
Then, to password protect this sample macro, right click on “Module 1” in the left hand tree view and select VBA Project Properties… In the resulting dialog you will be able to enable protection and set the password like in Figure 1.3.
![]()
Figure 1.3
Now exit out, saving the VBA macro. Then save the file as PowerPoint 97-2003 and you have a similar sample file to what I’m using for the rest of this blog. Now the fun begins while we follow the steps in [MS-PPT]!
As in my other blogs on binary formats like this one, I use SweetScape’s 010 editor to display the data in structures along the way. I mention this because in my previous posts, I’ve been asked what I used. [MS-PPT] “PowerPoint (.ppt) Binary File Format” is hyperlinked here for reference but I will not include links for all the sections as I walk through the steps below.
From here down to [5. Read the VbaProjectStgrecord…] I will merely list the steps in section 2.1.2 followed by the resulting data found at required locations. The steps and the highlighting speak for themselves. The excerpts from [MS-PPT] will be in italics. My story line is in normal font and the hex output will be in courier font. I’ve also tried to highlight using color to make it easier to match fields with my comments and calculations.
[MS-PPT] 2.1.2 PowerPoint Document Stream
Part 1: Construct the persist object directory.
1. Read the CurrentUserAtomrecord (section 2.3.2) from the Current User Stream(section 2.1.1). All seek operations in the steps that follow this step are in the PowerPoint Document Stream.
![]()
1000h: 00 00 F6 0F 20 00 00 00 14 00 00 00 5F C0 91 E3 ..ö. ......._À‘ã
1010h: 1A 99 00 00 08 00 F4 03 03 00 62 1C 54 6F 6D 20 .™....ô...b.Tom
1020h: 4A 65 62 6F 08 00 00 00 54 00 6F 00 6D 00 20 00 Jebo....T.o.m. .
1030h: 4A 00 65 00 62 00 6F 00 J.e.b.o.
2. Seek, in the PowerPoint Document Stream, to the offset specified by the offsetToCurrentEditfield of the CurrentUserAtomrecord identified in step 1.
CurrentUserAtomrecord.offsetToCurrentEdit = 1A 99 00 00
3. Read the UserEditAtomrecord at the current offset. Let this record be a live record.
0x1200 + 0x991A = 0xAB1A
AB10h: 00 00 F3 8D 00 00 D3 8A 00 00 00 00 F5 0F 1C 00 ..ó
AB20h: 00 00 00 00 00 00 81 11 00 03 00 00 00 00 FE 98 ......
AB30h: 00 00 01 00 00 00 04 00 00 00 01 00 62 1C 00 00 ............b...
Part 2: Identify the document persist object.
1. Read the docPersistIdReffield of the UserEditAtomrecord first identified in step 3 of Part 1, that is, the UserEditAtomrecord closest to the end of the stream.
docPersistIdReffield = 0x00000001
2. Lookup the value of the docPersistIdReffield in the persist object directory constructed in step 8 of Part 1 to find the stream offset of a persist object.
PersistDirectoryAtomrecord @ 0xAAFE
offsetLastEdit is 0x00000000
2.3.5 PersistDirectoryEntry bytes:
AAF0h: 83 87 97 5F 7A 7B 23 3A FC 07 00 00 FF FF 00 00 ƒ‡—_z{#:ü...ÿÿ..
AB00h: 72 17 14 00 00 00 01 00 40 00 00 00 00 00 F6 0B r.......@.....ö.
AB10h: 00 00 F3 8D 00 00 D3 8A 00 00 00 00 F5 0F 1C 00 ..ó
PersistOffsetEntry = 00 00 00 00
3. Seek to the stream offset specified in step 2.
4. Read theDocumentContainerrecordat the current offset. Let this record be a live record.
1200h: 0F 00 E8 03 EE 0B 00 00 01 00 E9 03 28 00 00 00 ..è.î.....é.(...
1210h: 00 1E 00 00 E0 10 00 00 E0 10 00 00 80 16 00 00 ....à...à...€...
1220h: 05 00 00 00 0A 00 00 00 00 00 00 00 00 00 00 00 ................
Part 11: Identify the VBA project persist object.
1. Read the DocInfoListContainerrecord (section 2.4.4), if present, specified by the docInfoList field of the DocumentContainerrecord identified in step 4 of Part 2. If not present, skip to step 6.
docInfoList @0x148C
1480h: 00 00 00 00 00 00 00 80 00 00 00 00 0F 00 D0 07 .......€......Ð.
1490h: 44 01 00 00 1F 00 14 04 1C 00 00 00 00 00 15 04 D...............
14A0h: 14 00 00 00 85 D1 F0 08 00 CA 9A 3B AD 07 94 C7 ....…Ñð..Êš;.”Ç
14B0h: 00 CA 9A 3B 01 02 00 00 1F 00 13 04 3C 00 00 00 .Êš;........<...
14C0h: 00 00 FD 03 34 00 00 00 01 00 00 00 01 00 00 00 ..ý.4...........
14D0h: 01 00 00 00 01 00 00 00 54 B5 C6 00 DC AF C6 00 ........TµÆ.ܯÆ.
14E0h: 08 00 00 00 18 B0 C6 00 68 B2 C6 00 1C B2 C6 00 .....°Æ.h²Æ..²Æ.
14F0h: 00 00 00 00 00 00 00 00 00 01 C6 00 0F 00 FA 03 ..........Æ...ú.
1500h: 47 00 00 00 00 00 FE 03 03 00 00 00 00 00 00 00 G.....þ.........
1510h: 00 FD 03 34 00 00 00 72 00 00 00 64 00 00 00 72 .ý.4...r...d...r
1520h: 00 00 00 64 00 00 00 3C AF C6 00 08 00 00 00 30 ...d...<¯Æ.....0
1530h: B0 C6 00 03 00 00 00 7C B2 C6 00 4C AF C6 00 86 °Æ.....|²Æ.L¯Æ.†
1540h: 01 00 00 66 00 00 00 01 00 C6 00 1F 00 FF 03 14 ...f.....Æ...ÿ..
1550h: 00 00 00 02 00 00 04 0C 00 00 00 03 00 00 00 01 ................
1560h: 00 00 00 02 00 00 00 0F 00 88 13 69 00 00 00 0F .........ˆ.i....
2. Read theVBAInfoContainer(section 2.4.10) child record, if present, of the DocInfoListContainerrecord identified in step 1. If no such child record exists, skip to step 6.
1540h: 01 00 00 66 00 00 00 01 00 C6 00 1F 00 FF 03 14 ...f.....Æ...ÿ..
1550h: 00 00 00 02 00 00 04 0C 00 00 00 03 00 00 00 01 ................
1560h: 00 00 00 02 00 00 00 0F 00 88 13 69 00 00 00 0F .........ˆ.i....
3. Lookup the value of the vbaInfoAtom.persistIdReffield of the VBAInfoContainerrecord identified in step 2 in the persist object directory constructed in step 8 of Part 1 to find the stream offset of a persist object.
vbaInfoAtom.persistIdReffield = 03 00 00 00 = 3
4. Seek to the stream offset specified in step 3.
From the persist object directory above: F3 8D 00 00 = offset 0x8DF3
PowerPoint Document stream @ 0x1200
So VbaProjectStorage record is at 0x1200 + 0x8DF3 = 0x9FF3
5. Read the VbaProjectStgrecord at the current offset. Let this record be a live record.
I copied the VbaProjectStgCompressedAtom bytes from 010 editor into another file. This new file, then became a self-contained compound file by the nature of its internal organization.
9FF0h: 00 00 00 10 00 11 10 03 0B 00 00 00 22 00 00 78 ............"..x
A000h: 01 EC 59 7D 6C 1C C5 15 7F BB 77 B6 CF 97 D8 B9 .ìY}l.Å.»w¶Ï—ع
A010h: 18 13 12 93 C2 C6 06 E3 24 B6 D9 DD 3B DF 07 B1 ...“ÂÆ.ã$¶ÙÝ;ß.±
A020h: 69 6E EF C3 71 64 63 2B 46 71 A5 1E 84 B3 6F 8D inïÃqdc+Fq¥.„³o
…
…
…
Size of the compressed storage is: 03 0B 00 00 = 0xB03
I copied this entire block of bytes into a separate file and decompressed them to obtain a fully independent compound file.
The decompression algorithm is referenced in 2.10.42 “VbaProjectStgCompressedAtom” and says:
The original bytes of the storage are compressed by the algorithm specified in [RFC1950] and are decompressed by the algorithm specified in [RFC1951].
So at this point, you will need to write some code to decompress this using the zlib compression library. There is example code shown in Figure 1.4 at the end of this blog.
Back to the block of bytes in a separate file; this is indeed a [MS-CFB] conforming document with the VBA storages as specified in [MS-OVBA]. Opening that in 010 or Offvis, you can see the storages:
![]()
Now we can look at some of the stream data to convince ourselves that we have a valid VBA macro storage here. Since it is password protected, some of the fields are encrypted and look like just random data. But we can look at, for instance, a stream named “PROJECT” which according to [MS-OVBA]:
2.3.1 PROJECT Stream: Project Information
The PROJECT stream (1) specifies properties of the VBA project.
So, looking at the bytes at the beginning of it:
1AC0h: 49 44 3D 22 7B 30 30 30 30 30 30 30 30 2D 30 30 ID="{00000000-00
1AD0h: 30 30 2D 30 30 30 30 2D 30 30 30 30 2D 30 30 30 00-0000-0000-000
1AE0h: 30 30 30 30 30 30 30 30 30 7D 22 0D 0A 4D 6F 64 000000000}"..Mod
1AF0h: 75 6C 65 3D 4D 6F 64 75 6C 65 31 0D 0A 48 65 6C ule=Module1..Hel
1B00h: 70 46 69 6C 65 3D 22 22 0D 0A 4E 61 6D 65 3D 22 pFile=""..Name="
1B10h: 56 42 41 50 72 6F 6A 65 63 74 22 0D 0A 48 65 6C VBAProject"..Hel
1B20h: 70 43 6F 6E 74 65 78 74 49 44 3D 22 30 22 0D 0A pContextID="0"..
1B30h: 56 65 72 73 69 6F 6E 43 6F 6D 70 61 74 69 62 6C VersionCompatibl
1B40h: 65 33 32 3D 22 33 39 33 32 32 32 30 30 30 22 0D e32="393222000".
1B50h: 0A 43 4D 47 3D 22 45 42 45 39 34 37 37 45 33 36 .CMG="EBE9477E36
1B60h: 38 32 33 36 38 32 33 32 38 36 33 32 38 36 22 0D 82368232863286".
1B70h: 0A 44 50 42 3D 22 44 36 44 34 37 41 36 42 38 45 .DPB="D6D47A6B8E
1B80h: 42 44 36 34 44 41 36 34 44 41 39 42 32 36 36 35 BD64DA64DA9B2665
1B90h: 44 41 31 37 43 37 39 36 32 31 44 46 44 31 30 37 DA17C79621DFD107
1BA0h: 31 31 46 39 39 33 36 31 44 30 39 32 44 43 39 31 11F99361D092DC91
1BB0h: 45 46 41 42 37 36 34 34 43 37 35 42 33 41 38 30 EFAB7644C75B3A80
1BC0h: 35 39 36 42 22 0D 0A 47 43 3D 22 43 31 43 33 36 596B"..GC="C1C36
1BD0h: 44 36 45 36 45 36 45 36 45 36 45 22 0D 0A 0D 0A D6E6E6E6E6E"....
1BE0h: 5B 48 6F 73 74 20 45 78 74 65 6E 64 65 72 20 49 [Host Extender I
1BF0h: 6E 66 6F 5D 0D 0A 26 48 30 30 30 30 30 30 30 31 nfo]..&H00000001
1C00h: 3D 7B 33 38 33 32 44 36 34 30 2D 43 46 39 30 2D ={3832D640-CF90-
1C10h: 31 31 43 46 2D 38 45 34 33 2D 30 30 41 30 43 39 11CF-8E43-00A0C9
1C20h: 31 31 30 30 35 41 7D 3B 56 42 45 3B 26 48 30 30 11005A};VBE;&H00
1C30h: 30 30 30 30 30 30 0D 0A 0D 0A 5B 57 6F 72 6B 73 000000....[Works
1C40h: 70 61 63 65 5D 0D 0A 4D 6F 64 75 6C 65 31 3D 32 pace]..Module1=2
1C50h: 36 2C 20 32 36 2C 20 31 34 32 38 2C 20 36 37 30 6, 26, 1428, 670
1C60h: 2C 20 0D 0A , ..
From this, we see properties like Name="VBAProject” and Module=Module1 which gives us some confidence that we have indeed the properties for a VBA macro. For questions or comments about the contents of this blog, please contact Interoperability Documentation Help.
Figure 1.4 Example code to decompress the VbaProjectStgCompressedAtom
'===================================================================
' DISCLAIMER:
'-------------------------------------------------------------------
'
' This sample is provided as is and is not meant for use on a
' production environment. It is provided only for illustrative
' purposes. The end user must test and modify the sample to suit
' their target environment. '
' Microsoft can make no representation concerning the content of
' this sample. Microsoft is providing this information only as a
' convenience to you. This is to inform you that Microsoft has not
' tested the sample and therefore cannot make any representations
' regarding the quality, safety, or suitability of any code or
' information found here.
'
'===================================================================
// assumptions for the input file:
// 1. it has nothing in it but the a) 4 byte uncompressed length and b) compressed data
// 2. it is small enough to read in and process in a reasonable buffer and processing time
//
int main(int argc, char *argv[])
{
int BlockSizeUncompress=0x1000;
int cprLevel=Z_BEST_COMPRESSION;
FILE* streamIn = NULL;
FILE* streamOut = NULL;
enum
{
paramExe,
paramFileIn,
paramFileOut,
paramBlockSizeUncompress,
paramCompressLevel
};
if (argc<paramFileOut)
{
printf("run decompress <FileIn> <FileOut> [BlockSizeUncompress] [compres. level]\n"
"For example: decompress inputfile.bin uncompressed.data.bin");
return 0;
}
if (argc>paramBlockSizeUncompress)
BlockSizeUncompress=atol(argv[paramBlockSizeUncompress]);
if (argc>paramCompressLevel)
cprLevel=(int)atol(argv[paramCompressLevel]);
byte* pbOriginal;
byte* pbCompressed;
// open the input file for reading binary data
streamIn=fopen(argv[paramFileIn], "rb");
if (streamIn==NULL)
return 0;
// open the output file for writing binary data
streamOut=fopen(argv[paramFileOut], "wb");
if (streamOut==NULL)
return 0;
// Beginning of main decompression functionality
{
z_stream zcpr;
int ret=Z_OK;
long lOrigDone = 0;
int step=0;
memset(&zcpr,0,sizeof(z_stream));
inflateInit(&zcpr);
struct _stat buf;
// find out how big the input file is
if (_stat(argv[paramFileIn], &buf))
{
printf("failed to read file size");
return -4;
}
// allocate enough bytes for the whole file.
pbCompressed = (byte*)
malloc(buf.st_size);
size_t corig = 0;
long lorig = 0;
// read in the first 4 bytes which is the size of the uncompressed
// data
if (fread(&lorig, 1, 4, streamIn) == 0)
{
printf("failed to read uncompressed size");
return -4;
}
corig = lorig;
pbOriginal = (byte*) malloc(corig);
ZeroMemory(pbOriginal, sizeof(pbOriginal));
// read in the compressed data from the input file
if (fread(pbCompressed, 1, 4096, streamIn) == 0)
{
printf("failed to read compressed data");
return -4;
}
zcpr.next_in = pbCompressed;
zcpr.next_out = pbOriginal;
zcpr.avail_in = buf.st_size;
zcpr.avail_out = lorig;
// call zlib to uncompress or inflate the data
do
{
ret=inflate(&zcpr,Z_SYNC_FLUSH);
step++;
} while (ret==Z_OK);
// write out the uncompressed data to the output file.
fwrite(pbOriginal, 1, lorig, streamOut);
inflateEnd(&zcpr);
}
// clean up buffers and file pointers
free(pbCompressed);
free(pbOriginal);
fclose(streamIn);
fclose(streamOut);
return 0;
}
![]()
