Mercurial > libpst
annotate TODO @ 359:a3e674fade6c
From Jeffrey Morlan:
pst_parse_block misreads Table Contexts (aka "type 2") with a
multi-block Row Matrix ("ind2"). Rows are never split between blocks -
every block except the last has padding at the end which should be
ignored. I've only seen this affect the recipients table, but
presumably it could affect attachments too. This was causing
out-of-bounds memory ranges to be returned from pst_getBlockOffset and
later access; patch fixes both the table reading issue and adds a
missing bounds check to pst_getBlockOffset (so as not to risk a
segfault if the PST is corrupted).
author | Carl Byington <carl@five-ten-sg.com> |
---|---|
date | Wed, 06 Jul 2016 10:20:12 -0700 |
parents | 2f38c4ce606f |
children |
rev | line source |
---|---|
47 | 1 |
51 | 2 Need testing on big-endian machines. We may be missing some LEnn_CPU() |
3 calls. | |
47 | 4 |
139
1b3922080ca8
add forensic headers to capture some other data of interest; switch back to quoted From separator line
Carl Byington <carl@five-ten-sg.com>
parents:
108
diff
changeset
|
5 pst2diii needs header and mime type updates from readpst. |
1b3922080ca8
add forensic headers to capture some other data of interest; switch back to quoted From separator line
Carl Byington <carl@five-ten-sg.com>
parents:
108
diff
changeset
|
6 |
198
7c60d6d1c681
decode more recurrence mapi elements
Carl Byington <carl@five-ten-sg.com>
parents:
195
diff
changeset
|
7 At the next soname bump (to libpst.so.5) we should |
201
3850a3b11745
fixes for parallel readpst
Carl Byington <carl@five-ten-sg.com>
parents:
200
diff
changeset
|
8 move some of readpst into the shared library, in particular write_normal_email() |