-
pst file with 20,000 relatively small attachments is not processed correctly by libpff.
do you want to create your own test case, or shall I generate one for you?
thank you.
2009-06-30 21:44:29 UTC in libpff
-
pst file containing message with 20+k recipients causes libpff.
i'll send zipped pst file separately, since the built in upload only allows 256K.
2009-06-24 14:48:28 UTC in libpff
-
pst file containing message with 300 attachments causes libpff to fail.
libpff_descriptor_local_read: invalid descriptor index values - missing data identifier.
libpff_item_values_local_read: unable to read local descriptor for identifier: 1649.
libpff_item_determine_attachments: error while read local descriptor identifier: 1649.
libpff_item_get_amount_of_attachments: unable to...
2009-06-24 14:08:03 UTC in libpff
-
The libpff_item_values_get_entry_multi_value code only works for string and binary mv types. It fails for guid (and will fail for ints) because those types do not have a 4 byte length field.
2009-06-22 13:16:18 UTC in libpff
-
Thank you. The functionality is the main thing. The other detail changes are fine with me.
2009-06-22 10:45:11 UTC in libpff
-
Please consider making the enclosed changes for the notification stuff. This makes life a lot easier for those of us wrapping libpff with managed code.
2009-06-18 18:05:32 UTC in libpff
-
A standard Outlook Sync Issues / Server Failures folder contains some messages which our OutLook and MFCMAPI display. But libpff does not find any subitems for this folder. On the surface (via MFCMAPI) it appears that this is the same as any other content table.
2009-06-15 19:24:14 UTC in libpff
-
libpff_table_read: unsupported table type: 0xe011.
libpff_item_values_local_read: unable to read table.
libpff_attachment_get_item: error while read local descriptor identifier: 32809.
2009-06-15 18:24:47 UTC in libpff
-
There is no foolproof way to deal with this. I currently deal with it by checking to see whether or not all the chars are between decimal 32 and 126.
Out of curiousity, I looked at the MFCMAPI source code, and it tests the scenario by converting to an A string and a W string, and then compares the lengths. If the A string length is less than the W string length, it is the expected (UTF16...
2009-06-15 13:21:56 UTC in libpff
-
Thank you. I have not yet tested this (mainly because I do not want strings in UTF8 format), but I will get to the MV binary tests next week.
Also, besides having a live 1048 type scenario, I also came across some 1003 scenarios, in case you want some testing for those types once you implement them.
2009-06-12 17:35:16 UTC in libpff
