November 9, 2019
View b16d5b7

Add NULL check in cli_isnumber

Andrew authored on 2019/11/09 04:25:42
March 6, 2019
View e010ed3

Clean up of PDF object finding logic. Changes include recording object sizes as objects are found, identifying object streams in the object parsing section instead of the PDF parsing section, and limiting of stream and other object parsing to the size of the object instead of the size of the PDF. It is also easier to read and includes more inline documentation.

Micah Snyder authored on 2019/03/06 11:15:41
March 3, 2019
View 9180468

bb12284 - Fix to prevent path traversal when using cli_genfname() to generate filenames that may retain path and filename information. Changed scanrar so that it will no longer retain path information for extracted files.

Micah Snyder authored on 2019/03/03 03:05:17
January 26, 2019
View c442ca9

Updating and cleaning up copyright notices in 0.101 branch.

Micah Snyder authored on 2019/01/26 00:15:50
January 24, 2019
View bf16d3b

fuzz - 12528 - fixing left shift issue with OLE2 and utf16 to ascii decoding

Mickey Sola authored on 2019/01/24 05:58:49
January 16, 2019
View ca9c606

fuzz - 12260 - fixing undefined shift issue when handling javascript escape sequences during hex to int conversion

Mickey Sola authored on 2019/01/16 02:11:23
December 6, 2018
View 95b2d68

Replacing libclamav/cltypes.h with clamav-types.h.in, which generates a header clamav-types.h that we install alongside clamav.h.

Micah Snyder authored on 2018/12/06 10:46:20
November 16, 2018
View fef9404

bb12220: Converting strnlen() calls to cli_strnlen() for systems such as Solaris 10 where strnlen() is not available. Adding #else clause to cli_get_filepath_from_filedesc() for platforms where we have not implemented a mechanism to determine the filename from the file descriptor.

Micah Snyder (micasnyd) authored on 2018/11/16 23:24:10
October 16, 2018
View 6ad41ab

bcomp - fixing case where automatic detection would fail against little endian hex values; removing code for little endian decimal support; fixing some clang warnings; fixes for hexidecimal detection in sli_strnto functions; updating documentation

Mickey Sola authored on 2018/10/16 03:29:35
August 16, 2018
View 9e408e7

bb4007 - adding pcre byte sequence comparison functions

Mickey Sola authored on 2018/08/16 06:04:25
August 15, 2018
View 89d5207

Added new pdf object stream parsing capability.

Micah Snyder (micasnyd) authored on 2018/08/15 06:00:31
July 31, 2018
View d39cb65

Updating libclamunrar from legacy C implementation to modern unrar 5.6.5. API changes and supporting changes included to pass the filepath of the scanned file into libclamav through the cli_ctx structure, required by the unrar library to open archives. The filename argument may be optional for the scandesc scanning variant, but libclamav will make a best effort to identify the filename from the file descriptor if it was not provided. In addition, included the ability to prefix temp file and directory names with file basenames.

Micah Snyder authored on 2018/07/31 09:19:28
June 13, 2018
View f842e96

Replacing strntol with strntoul to ensure proper (un)signedness when parsing numbers from PDFs.

Micah Snyder authored on 2018/06/13 09:47:21
June 3, 2018
View bf6e777

bb12133: Wrapping cli_strntol to provide easy error detection. Applying cli_strntol_wrap with error checking. Adding logic to identify when a parsing error is in fact a new revision of the PDF.

Micah Snyder authored on 2018/06/03 09:58:35
June 2, 2018
View 53cbdee

bb12133: Implementing cli_strntol based on gnu gcc's strtol implementation with modifications to limit string buffer length for non-null terminated strings. Using cli_strntol in pdf.c for added safety.

Micah Snyder authored on 2018/06/02 03:23:25