aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* * ldap.m4: Revert the mozldap change: mozldap has some significantDavid Shaw2007-08-022-48/+32
| | | | | | | | differences around TLS compared to OpenLDAP, and will segfault (!) if the NSS library is not initialized first. Nothing impossible to do, but I think I'll wait until someone actually asks us for mozldap support.
* * gpgkeys_ldap.c (main): Fix bug in setting up whether to verify peerDavid Shaw2007-07-292-1/+7
| | | | | | SSL cert. This used to work with older OpenLDAP, but is now more strictly handled.
* * gpgkeys_ldap.c: Fix build warning with mozldap.David Shaw2007-07-272-0/+7
|
* * gpgkeys_ldap.c (search_key, main): Fix bug where searching for fooDavid Shaw2007-07-272-5/+19
| | | | | | bar (no quotes) on the command line resulted in searching for "foo\2Abar" due to LDAP quoting. The proper search is "foo*bar".
* * ldap.m4: If we don't find ldap or winldap, try for mozldap.David Shaw2007-07-272-24/+52
|
* * gpg.c (main): Fix typo. Noted by John Clizbe.David Shaw2007-07-192-1/+3
|
* * armor.c (parse_header_line): Improve test so that the header testDavid Shaw2007-07-192-16/+25
| | | | | only allows "Hash" in the signed data section.
* * configure.ac: It seems that AM_GNU_GETTEXT doesn't like being runDavid Shaw2007-07-182-4/+10
| | | | | | conditionally and it confuses the includes when doing later type checks. This is bug#810.
* * defs.inc (all_hash_algos): See "all_cipher_algos", below.David Shaw2007-07-183-34/+30
| | | | | | * sigs.test: Use it here, and also test with >=160 bit hashes for DSA2.
* * armor.c (is_armor_tag): New. Detect if an armor header matchesDavid Shaw2007-07-182-2/+34
| | | | | | | 2440bis-21. (parse_header_line): Call it here, as bis-21 requires warning the user (but continuing to process the message) when seeing an unknown header.
* Add an extra warning for Camellia.Werner Koch2007-07-0936-3925/+3968
| | | | | | Minor W32 fix Address change.
* * configure.ac: AM_PROG_CC_C_O is desired by automake 1.10 to doDavid Shaw2007-07-082-0/+7
| | | | | | per-target builds (like we do in the keyserver dir), but exists in the currently-used 1.9.3 as well.
* * encode.c (encode_crypt): Missed one call to setup_plaintext_name().David Shaw2007-07-062-16/+7
| | | | | This is bug#809.
* * mk-w32-dist: Remove underline markup from man pages. ReportedWerner Koch2007-06-222-3/+13
| | | | | by Dirk Traulsen.
* Add Camellia. Do not enable this if you are not doing interopDavid Shaw2007-06-1311-9/+1763
| | | | | | | testing. It is not (yet) legal OpenPGP, is not interop tested yet (obviously), and it's a great way to lose your data. Just don't do it.
* * sign.c (mk_notation_policy_etc): expect all sigs that this is calledDavid Shaw2007-06-122-43/+24
| | | | | | | | for are >=v4. (write_signature_packets, make_keysig_packet): Only call it for >=v4 sigs. This allows --force-v3-sigs and --force-v4-certs to enable or disable notations, policies, and keyserver URLs. This is bug#800.
* * defs.inc (all_cipher_algos): New function to return all ciphers weDavid Shaw2007-05-146-102/+27
| | | | | | | | | | support. This is safer than the previous setup which could hide that some ciphers weren't being tested. Plus, this automatically tests any new ciphers we may add. * conventional.test, encrypt-test, encrypt-dsa.test, conventional-mdc.test: Use it here.
* * conventional.test, encrypt.test, encrypt-dsa.test,David Shaw2007-05-025-21/+31
| | | | | | conventional-mdc.test: Fix some broken tests that were only testing 3DES instead of all available ciphers.
* * packet.h, mainproc.c (reset_literals_seen): New function to resetDavid Shaw2007-04-175-7/+24
| | | | | | | | | the literals count. * verify.c (verify_one_file), decrypt.c (decrypt_messages): Call it here so we allow multiple literals in --multifile mode (in different files - not concatenated together).
* * gpgkeys_hkp.c (main): Show curl or fake-curl version string.David Shaw2007-04-163-3/+7
| | | | | * Makefile.am: Link with libcompat.a for ascii_str(n)casecmp.
* * gpgkeys_curl.c, gpgkeys_ldap.c, gpgkeys_hkp.c, ksutil.c: Rename allDavid Shaw2007-04-165-38/+48
| | | | | | str(n)casecmp to ascii_str(n)casecmp. We want the locale-independent ones here.
* Move some ascii_foo functions to libcompatDavid Shaw2007-04-166-74/+84
|
* * gpg.c, pkclist.c: #include <strings.h> for strcasecmp if it isDavid Shaw2007-04-163-4/+16
| | | | | | present. Note that autoconf protects us against a strings.h that cannot be used together with string.h.
* util/Werner Koch2007-04-165-51/+67
| | | | | | | | | | | * secmem.c (init_pool): Avoid assigning a negative value to a size_t. ./ * acinclude.m4: Fix last change. Make test self-conatined by checking for sysconf and getpagesize. Remove indentation for the sake of broken C-89 cpps.
* Don't build shmtest unless we have shm.David Shaw2007-04-164-1/+13
|
* Use sysconf() when possible as not all platforms have getpagesize().David Shaw2007-04-165-7/+32
|
* * argparse.c (default_strusage): Copyright 2007.David Shaw2007-04-155-6/+15
| | | | | * cert.c, srv.c, pka.c: Need arpa/inet.h for ntohs().
* * configure.ac: QNX puts resolver functions in libsocket. From SeanDavid Shaw2007-04-152-6/+11
| | | | | Boudreau.
* * parse-packet.c (parse_marker): New. Enforce that the markerDavid Shaw2007-04-042-22/+65
| | | | | | contains 'P', 'G', 'P', and nothing but. (parse): Call it here. (skip_packet): No longer need to handle marker packets here.
* * keyserver.c: Windows Vista doesn't grok X_OK and so fails access()David Shaw2007-03-142-2/+16
| | | | | | tests. Previous versions interpreted X_OK as F_OK anyway, so we'll just use F_OK directly.
* * gpgkeys_curl.c (main): Use curl_version_info to verify that theDavid Shaw2007-03-134-6/+43
| | | | | | | | | protocol we're about to use is actually available. * curl-shim.h, curl-shim.c (curl_free): Make into a macro. (curl_version_info): New. Only advertises "http" for our shim, of course.
* Add extra fflush (stdout)Werner Koch2007-03-082-0/+15
|
* * keyedit.c (keyedit_menu): If we modify the keyblock (viaDavid Shaw2007-03-063-2/+20
| | | | | | | | | fix_keyblock() or collapse_uids()) make sure we reprocess the keyblock so the flags are correct. Noted by Robin H. Johnson. * getkey.c (fixup_uidnode): Properly clear flags that don't apply to us (revoked, expired) so that we can reprocess a uid.
* Post release updateWerner Koch2007-03-052-2/+6
|
* Preparing 1.4.7gnupg-1.4.7Werner Koch2007-03-0544-17761/+17943
|
* UpdatedWerner Koch2007-02-263-3463/+3323
|
* [W32] Make use of the LANGUAGE envvar.Werner Koch2007-02-263-71/+140
|
* * parse-packet.c (parse_signature): It's hex.David Shaw2007-02-143-1/+11
| | | | | | * getkey.c (merge_selfsigs_subkey): Avoid listing the contents of a backsig when list mode is on. Noted by Timo Schulz.
* * secmem.c (ptr_into_pool_p): New.Werner Koch2007-02-122-1/+24
| | | | | | | (m_is_secure): Implement in terms of above. Also check that the pool has been initialized.
* * gpgkeys_ldap.c (send_key): Missing a free().David Shaw2007-02-113-1/+16
| | | | | | * curl-shim.c (curl_easy_perform): Some debugging items that may be handy.
* * http.c (do_parse_uri): Remove the hkp port 11371 detection. WeDavid Shaw2007-02-112-2/+6
| | | | | | implement hkp in the keyserver handler, and the support here makes it appear like a bad hkp request actually succeeded.
* Fix for Debian bug 402592Werner Koch2007-02-042-3/+24
|
* Missed one caller.David Shaw2007-02-011-1/+1
|
* * main.h, keygen.c (ask_expire_interval, parse_expire_string): Pass inDavid Shaw2007-02-016-27/+39
| | | | | | the time to use to calculate the expiration offset, rather than querying it internally. Change all callers.
* * ttyio.c (do_get): Assume that anything read from the user withoutDavid Shaw2007-02-012-1/+8
| | | | | | echoing to the screen is sensitive and put it in secure memory. Suggested by Benjamin Bennett.
* * keygen.c (do_generate_keypair, proc_parameter_file,David Shaw2007-01-312-35/+62
| | | | | | generate_keypair, generate_subkeypair): Pass a timestamp through to all the gen_xxx functions.
* * keyedit.c (sign_uids): Another multiple to single timestampDavid Shaw2007-01-312-12/+9
| | | | | operation.
* * sign.c (write_plaintext_packet): Take timestamp from outside.David Shaw2007-01-312-10/+20
| | | | | | | | Change all callers. (sign_file, clearsign_file, sign_symencrypt_file): Calculate one timestamp for the whole transaction. Different timestamps for each step can leak performance information.
* * main.h, keygen.c (ask_expiredate), keyedit.c (menu_expire): RemoveDavid Shaw2007-01-314-10/+9
| | | | | the need for ask_expiredate().
* * keygen.c (make_backsig, write_direct_sig, write_selfsigs)David Shaw2007-01-302-7/+17
| | | | | | | | (write_keybinding): Have the various selfsigs created during key generation use the timestamp of their parent key. This helps avoid leaking information about the performance of the machine where the keys were generated. Noted by Christian Danner.
generated by cgit v1.2.3 (git 2.25.1) at 2025-06-20 20:58:46 +0000