platform/upstream/evolution-data-server.git
21 years agoGet rid of the #ifdef ENABLE_THREADS since we no longer plan to
Jeffrey Stedfast [Wed, 9 Jul 2003 19:05:13 +0000 (19:05 +0000)]
Get rid of the #ifdef ENABLE_THREADS since we no longer plan to

2003-07-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-block-file.c: Get rid of the #ifdef ENABLE_THREADS since
we no longer plan to support/maintain this.

* camel.c: Same.

* camel-certdb.c: Here too.

* camel-charset-map.c: And here.

* camel-cipher-context.c: "

* camel-data-wrapper.c: "

* camel-digest-folder.c: "

* camel-exception.c: "

* camel-folder.c: "

* camel-folder-summary.c: "

* camel-lock-client.c: "

* camel-mime-utils.c: "

* camel-object.c: "

* camel-operation.c: "

* camel-partition-table.c: "

* camel-sasl-popb4smtp.c: "

* camel-service.c: "

* camel-session.c: "

* camel-store.c: "

* camel-store-summary.c: "

* camel-text-index.c: "

* camel-transport.c: "

* camel-vee-folder.c: "

* camel-tcp-stream-openssl.c: Removed pthread.h, it isn't needed.

21 years agoremove reference to camel-pgp-mime.h
Larry Ewing [Wed, 9 Jul 2003 12:49:59 +0000 (12:49 +0000)]
remove reference to camel-pgp-mime.h

2003-07-09  Larry Ewing  <lewing@ximian.com>

* camel.h: remove reference to camel-pgp-mime.h

21 years agoRemoved.
Jeffrey Stedfast [Tue, 8 Jul 2003 19:29:58 +0000 (19:29 +0000)]
Removed.

2003-07-08  Jeffrey Stedfast  <fejj@ximian.com>

* camel-pgp-mime.[c,h]: Removed.

21 years agoUpdated (new copy/paste from e-iconv).
Jeffrey Stedfast [Tue, 8 Jul 2003 18:36:40 +0000 (18:36 +0000)]
Updated (new copy/paste from e-iconv).

2003-07-08  Jeffrey Stedfast  <fejj@ximian.com>

* camel-iconv.c: Updated (new copy/paste from e-iconv).

* camel-block-file.c (camel_block_file_get_block): Use
camel_read() rather than libc read.

* camel-tcp-stream-raw.c (stream_read): Use camel_read().
(stream_write): Use camel_write().

* camel-stream-fs.c (stream_read): Use camel_read().
(stream_write): Use camel_write().

21 years agoUse camel_read(). (stream_write): Use camel_write().
Jeffrey Stedfast [Tue, 8 Jul 2003 17:48:32 +0000 (17:48 +0000)]
Use camel_read(). (stream_write): Use camel_write().

2003-07-08  Jeffrey Stedfast  <fejj@ximian.com>

* camel-tcp-stream-raw.c (stream_read): Use camel_read().
(stream_write): Use camel_write().

* camel-stream-fs.c (stream_read): Use camel_read().
(stream_write): Use camel_write().

21 years agoUse camel_mkdir().
Jeffrey Stedfast [Mon, 7 Jul 2003 20:58:24 +0000 (20:58 +0000)]
Use camel_mkdir().

2003-07-07  Jeffrey Stedfast  <fejj@ximian.com>

* providers/nntp/camel-nntp-folder.c (camel_nntp_folder_new): Use
camel_mkdir().

* providers/imap/camel-imap-folder.c (camel_imap_folder_new): Use
camel_mkdir().

* camel-session.c (get_storage_path): Use camel_mkdir().

* camel-store.c (camel_mkdir_hier): Removed.

* camel-data-cache.c (camel_data_cache_new): Updated to use
camel_mkdir().
(data_cache_path): Same.

* camel-file-utils.c (camel_mkdir): Renamed and documented.
(camel_file_util_safe_filename): Documented.
(camel_read): Moved here from camel-io.c
(camel_write): Same.

* camel-io.[c,h]: Removed.

* camel-uid-cache.c (camel_uid_cache_new): Use the
camel-file-utils.c version of mkdir.

21 years agoremoved debug printfs
Jeffrey Stedfast [Mon, 7 Jul 2003 19:20:25 +0000 (19:20 +0000)]
removed debug printfs

21 years agoUpdated for string-utils namespace changes.
Jeffrey Stedfast [Mon, 7 Jul 2003 19:18:57 +0000 (19:18 +0000)]
Updated for string-utils namespace changes.

2003-07-07  Jeffrey Stedfast  <fejj@ximian.com>

* camel-session.c (camel_session_init): Updated for string-utils
namespace changes.

* camel-provider.c: Updated for string-utils namespace changes.

* camel-mime-part.c (init_header_name_table): Updated for
string-utils namespace changes.

* camel-mime-message.c (camel_mime_message_class_init): Updated
for string-utils namespace changes.
(camel_mime_message_init): Same.

* camel-mime-filter-enriched.c
(camel_mime_filter_enriched_class_init): Updated for string-utils
namespace changes.

* camel-folder-summary.c (camel_folder_summary_init): Updated for
string-utils namespace changes.

* camel-string-utils.[c,h]: Renamed from string-utils.[c,h] and
also namespaced all functions.

21 years agoDon't quote the charset value, the qop value, nor the response value.
Jeffrey Stedfast [Wed, 2 Jul 2003 23:43:12 +0000 (23:43 +0000)]
Don't quote the charset value, the qop value, nor the response value.

2003-07-01  Jeffrey Stedfast  <fejj@ximian.com>

* camel-sasl-digest-md5.c (digest_response): Don't quote the
charset value, the qop value, nor the response value. Fixes bug
#45712.

21 years agoUse gmtime_r() instead of using gmtime() and memcpy() to try and be
Jeffrey Stedfast [Wed, 2 Jul 2003 16:47:29 +0000 (16:47 +0000)]
Use gmtime_r() instead of using gmtime() and memcpy() to try and be

2003-07-01  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-utils.c (header_format_date): Use gmtime_r() instead
of using gmtime() and memcpy() to try and be "atomic".

21 years agofree the summary hash
Dan Winship [Tue, 1 Jul 2003 18:47:04 +0000 (18:47 +0000)]
free the summary hash

* camel-folder-search.c (camel_folder_search_finalize): free the
summary hash

21 years agoPut day of week into outgoing email.
David Woodhouse [Tue, 1 Jul 2003 18:10:17 +0000 (18:10 +0000)]
Put day of week into outgoing email.

2003-06-24  David Woodhouse  <dwmw2@infradead.org>

* camel-mime-utils.c (header_format_date): Put day of week into
outgoing email.

21 years agoAdded prototype for camel_folder_summary_remove_range().
Jeffrey Stedfast [Wed, 25 Jun 2003 15:43:46 +0000 (15:43 +0000)]
Added prototype for camel_folder_summary_remove_range().

2003-06-25  Jeffrey Stedfast  <fejj@ximian.com>

* camel-folder-summary.h: Added prototype for
camel_folder_summary_remove_range().

2003-06-25  Not Zed  <NotZed@Ximian.com>

** See bug #45386

* camel-service.c (camel_gethostbyname, camel_gethostbyaddr): Make
sure we have an exception that we test against.

21 years ago** See bug #43887
Not Zed [Wed, 25 Jun 2003 10:20:09 +0000 (10:20 +0000)]
** See bug #43887

2003-06-20  Not Zed  <NotZed@Ximian.com>

** See bug #43887

* camel-mime-filter-enriched.c (camel_enriched_to_html): simple
wrapper to convert enriched to html in one go.

21 years ago** See bug #31745
Not Zed [Fri, 20 Jun 2003 05:04:58 +0000 (05:04 +0000)]
** See bug #31745

2003-06-16  Not Zed  <NotZed@Ximian.com>

** See bug #31745

* providers/imap/camel-imap-store-summary.c
(camel_imap_store_summary_namespace_new): Workaround a shell bug -
if the namespace has '#' in it, then strip it.

21 years agoif we got cancelled, the message is floating, so free it. (struct
Not Zed [Fri, 20 Jun 2003 03:53:35 +0000 (03:53 +0000)]
if we got cancelled, the message is floating, so free it. (struct

2003-06-18  Not Zed  <NotZed@Ximian.com>

        * camel-service.c (get_hostbyaddr, get_hostbyname): if we got
        cancelled, the message is floating, so free it.
        (struct _lookup_msg): Add a cancelled tag.
        (camel_gethostbyname, camel_gethostbyaddr): if we get a
        failure/cancel, cancel the lookup thread and detach, so we dont
        have to wait for it to return.  cleanup changed to handle the case
        where we didn't get a reply message.

21 years agoMade static (to match the prototype).
Jeffrey Stedfast [Wed, 18 Jun 2003 17:55:40 +0000 (17:55 +0000)]
Made static (to match the prototype).

2003-06-13  Jeffrey Stedfast  <fejj@ximian.com>

* providers/pop3/camel-pop3-folder.c (pop3_finalize): Made static
(to match the prototype).

21 years agodon't leak the summary when reloading it. Fixes a very large leak.
Larry Ewing [Wed, 18 Jun 2003 15:13:12 +0000 (15:13 +0000)]
don't leak the summary when reloading it. Fixes a very large leak.

2003-06-13  Larry Ewing  <lewing@ximian.com>

* camel-folder-thread.c (camel_folder_thread_messages_apply):
don't leak the summary when reloading it.  Fixes a very large
leak.

21 years agoCalculate ranges to remove folder info's more efficiently. affects
Not Zed [Wed, 18 Jun 2003 06:14:09 +0000 (06:14 +0000)]
Calculate ranges to remove folder info's more efficiently. affects

2003-06-17  Not Zed  <NotZed@Ximian.com>

* camel-vee-folder.c (vee_folder_remove_folder): Calculate ranges
to remove folder info's more efficiently.  affects shutdown
performance on big vfolders signifinantly.
(vee_folder_build_folder): do the same here, when rebuilding a
folder's definition.

* camel-folder-summary.c (camel_folder_summary_remove_index): new
function to drop a range of index entries in one hit.

21 years ago** See bug #44322
Not Zed [Tue, 17 Jun 2003 03:20:59 +0000 (03:20 +0000)]
** See bug #44322

2003-06-16  Not Zed  <NotZed@Ximian.com>

** See bug #44322

* providers/imap/camel-imap-command.c (imap_command_strdup_vprintf):
If we are outputting a folder name, make sure we calculate buffer
size based on the raw/utf7 version

** See bug #44121

* camel-multipart-signed.c (signed_get_part): If we can't parse
the content, but we have a stream, just use that as the content.

21 years agoFix for bug #40788.
Jeffrey Stedfast [Fri, 13 Jun 2003 19:06:52 +0000 (19:06 +0000)]
Fix for bug #40788.

2003-06-05  Jeffrey Stedfast  <fejj@ximian.com>

Fix for bug #40788.

* providers/pop3/camel-pop3-engine.c (camel_pop3_engine_new): Now
takes a flags argument. Currently there is only 1 flag which can
be used to disable Pop3 server extensions.
(get_capabilities): Don't check for Pop3 server extensions if the
DISABLE_EXTENSIONS flag is set on the engine.
(camel_pop3_engine_iterate): If we get a response that is neither
+OK nor -ERR, default to treating it like a -ERR.

* providers/pop3/camel-pop3-store.c (connect_to_server): Check for
the disable_extensions param.

* providers/pop3/camel-pop3-provider.c: Define a checkbox to
disable all POP3 extension support.

21 years agoPut the ChangeLog.pre-1-4 in the Makefiles.
Ettore Perazzoli [Thu, 12 Jun 2003 21:13:56 +0000 (21:13 +0000)]
Put the ChangeLog.pre-1-4 in the Makefiles.

21 years agoPartial fix for bug #44457.
Jeffrey Stedfast [Thu, 12 Jun 2003 04:48:05 +0000 (04:48 +0000)]
Partial fix for bug #44457.

2003-06-11  Jeffrey Stedfast  <fejj@ximian.com>

Partial fix for bug #44457.

* camel-mime-part-utils.c
(simple_data_wrapper_construct_from_parser): Make sure to set
rawtext to FALSE if we successfully convert the text to UTF-8.

* camel-data-wrapper.c (camel_data_wrapper_init): Default the
value of rawtext to TRUE instead of FALSE. This way if the mailer
decides to try displaying a non-textual part as text, it knows
that it needs to convert the content to UTF-8.

21 years agosome days you just wish you had that subatomic 100 megaton kill-o-zap bomb.
Jeffrey Stedfast [Wed, 11 Jun 2003 20:36:06 +0000 (20:36 +0000)]
some days you just wish you had that subatomic 100 megaton kill-o-zap bomb.

2003-06-04  Jeffrey Stedfast  <fejj@ximian.com>

* camel-uid-cache.c (camel_uid_cache_new): Create the directory
with mode 0777 and the cache file itself with mode 0666. Let the
user's umask filter the permissions. Instead of saving the fd on
the Cache object, instead save the filename. Use camel_read()
instead of expecting read() to just always work without getting an
EINTR/etc.
(maybe_write_uid): Don't do anything if cache->fd == -1, this
means an error has occured in a previous callback. Replace the 2
calls to write() with camel_write() and check their return
values. If either of them fails, set cache->fd to -1 (GHashTable
doesn't give us a way to abort foreach'ing thru the table).
(camel_uid_cache_save): Save to a temp file instead of overwriting
the original. Do proper error checking, etc. Also added some
smarts about whether to try and overwrite the old cache even if we
haven't successfully saved all the uids in the cache.
(camel_uid_cache_destroy): Free the cache->filename, no longer
need to close (cache->fd).

21 years agouse g_utf8_strdown properly.
Larry Ewing [Wed, 11 Jun 2003 19:38:33 +0000 (19:38 +0000)]
use g_utf8_strdown properly.

2003-06-11  Larry Ewing  <lewing@ximian.com>

* camel-text-index.c (text_index_normalise): use g_utf8_strdown properly.

21 years agoAdd the CAMEL_BESTENC_TEXT bit to enctype if the part is a text part.
Jeffrey Stedfast [Wed, 11 Jun 2003 15:59:30 +0000 (15:59 +0000)]
Add the CAMEL_BESTENC_TEXT bit to enctype if the part is a text part.

2003-06-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-message.c (find_best_encoding): Add the
CAMEL_BESTENC_TEXT bit to enctype if the part is a text part.

* camel-mime-filter-bestenc.c
(camel_mime_filter_bestenc_get_best_encoding): If we have any
nul-bytes or if the content is non-text and contains any 8bit
octets, we need to use base64. Fixes bug #44344.

21 years agoRoll over ChangeLogs.
Ettore Perazzoli [Mon, 9 Jun 2003 17:31:12 +0000 (17:31 +0000)]
Roll over ChangeLogs.

21 years agoTurn off debugging.
Jeffrey Stedfast [Sun, 1 Jun 2003 23:49:29 +0000 (23:49 +0000)]
Turn off debugging.

2003-06-01  Jeffrey Stedfast  <fejj@ximian.com>

* broken-date-parser.c (d): Turn off debugging.

* providers/pop3/camel-pop3-engine.c (get_capabilities): Move the
code that prepends the 'password' authtype to the auth list so
that we don't add it again when regetting the capabilities.

21 years agoTurned off a debugging printf.
Jeffrey Stedfast [Fri, 30 May 2003 15:50:07 +0000 (15:50 +0000)]
Turned off a debugging printf.

2003-05-30  Jeffrey Stedfast  <fejj@ximian.com>

* camel-filter-driver.c (run_only_once): Turned off a debugging
printf.

21 years agoThis should finish fixing all variations of bug #42854, which, funnily
Jeffrey Stedfast [Wed, 28 May 2003 00:42:19 +0000 (00:42 +0000)]
This should finish fixing all variations of bug #42854, which, funnily

2003-05-27  Jeffrey Stedfast  <fejj@ximian.com>

This should finish fixing all variations of bug #42854, which,
funnily enough specified "copying" as the scenario for reproducing
this bug, but somehow I found/fixed it only for expunging for some
reason? I can't even guess as to how this happened. Oh well,
expunging needed fixing either way.

* providers/imap/camel-imap-folder.c (imap_transfer_online): Sort
the uids here too (just like I had fixed for expunge a week or so
ago).
(imap_transfer_resyncing): Sort them here too.
(imap_search_by_uids): Sort here too, even though I don't think it
is necessary. Can't hurt.
(imap_expunge_uids_offline): Might as well sort here too.

21 years agoLets try this again. We don't actually want to free the key here, because
Jeffrey Stedfast [Thu, 22 May 2003 20:17:02 +0000 (20:17 +0000)]
Lets try this again. We don't actually want to free the key here, because

2003-05-22  Jeffrey Stedfast  <fejj@ximian.com>

* camel-object.c (camel_object_bag_destroy): Lets try this
again. We don't actually want to free the key here, because
camel_object_bag_remove() does that for us. This should fix bug
#43406 properly.

21 years agoCreate a second GPtrArray for the keys. We don't want to use
Jeffrey Stedfast [Wed, 21 May 2003 15:32:37 +0000 (15:32 +0000)]
Create a second GPtrArray for the keys. We don't want to use

2003-05-20  Jeffrey Stedfast  <fejj@ximian.com>

* camel-object.c (camel_object_bag_destroy): Create a second
GPtrArray for the keys. We don't want to use bag->free_key() on
the object. Fixes bug #43406.

21 years agoOnly display a warning if errno != EINVAL (EINVAL means there was a
Jeffrey Stedfast [Tue, 20 May 2003 15:59:55 +0000 (15:59 +0000)]
Only display a warning if errno != EINVAL (EINVAL means there was a

2003-05-19  Jeffrey Stedfast  <fejj@ximian.com>

* camel-folder-summary.c (camel_folder_summary_load): Only display
a warning if errno != EINVAL (EINVAL means there was a version
mismatch, so we don't really need to warn here since we are
supposed to be able to handle this situation).
(perform_content_info_load): Turn off debugging message.
(summary_assign_uid): Turn off uid-clash warning.
(camel_folder_summary_decode_token): Turn off debug messages.
(summary_header_load): Turn off version-mismatch warning.

21 years agofixed ChangeLog
Jeffrey Stedfast [Mon, 19 May 2003 17:20:04 +0000 (17:20 +0000)]
fixed ChangeLog

21 years agoEnsure cmdbuf is initialized to avoid compiler warning.
Jeremy Katz [Fri, 16 May 2003 18:47:59 +0000 (18:47 +0000)]
Ensure cmdbuf is initialized to avoid compiler warning.

2003-05-16  Jeremy Katz  <katzj@redhat.com>

  * providers/smtp/camel-smtp-transport.c (smtp_helo): Ensure cmdbuf
  is initialized to avoid compiler warning.

  * providers/local/camel-mbox-summary.c (summary_header_load): Cast
  folder_size to uint32 to fix warning.
  (camel_mbox_summary_sync_mbox): Update for new API.

  * providers/imap/camel-imap-utils.c (parse_params):
  imap_parse_nstring expects size_t, not int
  (imap_body_decode): Likewise.

  * camel-stream-filter.c (do_read): presize needs to be size_t
  instead of int.
  (do_write): Likewise.

  * camel-seekable-substream.c (stream_read): Return ssize_t instead
  of int to match prototypes.
  (stream_write): Likewise.

  * camel-mime-part.c (construct_from_parser): len needs to be
  size_t instead of int.

  * camel-mime-parser.c (folder_scan_step): datalength needs to be a
  size_t* instead of an int* to make dependent APIs safe.
  (camel_mime_parser_step): Likewise.
  (folder_scan_content): Likewise.
  * camel-mime-parser.h: Likewise.

  * camel-mime-message.c (camel_mime_message_class_init): Use glib
  macro for pointer/int conversions.
  (construct_from_parser): Update for new API.

  * camel-folder-summary.c
  (camel_folder_summary_info_new_from_parser): Update len to be
  size_t for new API.
  (summary_build_content_info): Likewise.
  * camel-http-stream.c (http_get_headers): Likewise.
  * camel-mime-part-utils.c
  (simple_data_wrapper_construct_from_parser): Likewise.
  * camel-multipart-signed.c (signed_construct_from_parser): Likewise.
  * camel-multipart.c (construct_from_parser): Likewise.

  * camel-folder-search.c (match_words_index): Use glib macros for
  pointer/int conversions.
  * camel-html-parser.c (tokenise_setup): Likewise.
  (convert_entity): Likewise.
  * camel-block-file.c (block_hash_func): Likewise.
  (camel_block_file_get_block): Likewise.
  (camel_block_file_get_block): Likewise.
  (camel_block_file_detach_block): Likewise.
  * camel-session.c (session_thread_msg_new): Likewise.
  (session_thread_msg_free): Likewise.
  (session_thread_wait): Likewise.
  * camel-text-index.c (text_index_compress_nosync): Likewise.
  (text_index_compress_nosync): Likewise.
  (camel_text_index_validate): Likewise.
  * camel-vee-folder.c (vee_folder_remove_folder): Likewise.
  (unmatched_check_uid): Likewise.
  (folder_added_uid): Likewise.
  (vee_folder_build_folder): Likewise.
  (folder_changed_add_uid): Likewise.
  (folder_changed_remove_uid): Likewise.
  * providers/imap/camel-imap-search.c (imap_body_contains): Likewise.
  * providers/pop3/camel-pop3-folder.c (cmd_list): Likewise.
  (cmd_uidl): Likewise.

  * camel-data-cache.c (data_cache_init): Cast to CamelCopyFunc.
  Fixes a warning.

21 years agoSimplistic quick-fix for #42710. Add a charset argument. If the data is 8
Not Zed [Thu, 15 May 2003 23:30:06 +0000 (23:30 +0000)]
Simplistic quick-fix for #42710. Add a charset argument. If the data is 8

2003-05-15  Not Zed  <NotZed@Ximian.com>

        * camel-mime-utils.c (header_decode_word): Simplistic quick-fix
        for #42710.  Add a charset argument.  If the data is 8 bit, try to
        decode into the passed in charset (i.e. the message header?), at
        worst, fall back to latin1, similarly to header_decode_text.
        Fixed all callers to pass in a charset if available, or NULL.

21 years agoOnly delete the index if it is non-NULL. Fixes bug #42763.
Jeffrey Stedfast [Thu, 15 May 2003 18:05:14 +0000 (18:05 +0000)]
Only delete the index if it is non-NULL. Fixes bug #42763.

2003-05-13  Jeffrey Stedfast  <fejj@ximian.com>

* providers/local/camel-local-folder.c (local_delete): Only delete
the index if it is non-NULL. Fixes bug #42763.

21 years ago** See bug #42540
Not Zed [Thu, 15 May 2003 01:00:56 +0000 (01:00 +0000)]
** See bug #42540

2003-05-14  Not Zed  <NotZed@Ximian.com>

        ** See bug #42540

        * camel-service.c (camel_gethostbyname): pre-initialise the error
        return to invalid, so if we cancel, we definetly get a bad result.
        Also, for a valid return against result, not herr.
        (camel_gethostbyaddr): Likewise.

21 years agoAdded a note to the comment block for this function noting that uids MUST
Jeffrey Stedfast [Wed, 14 May 2003 17:53:47 +0000 (17:53 +0000)]
Added a note to the comment block for this function noting that uids MUST

2003-05-13  Jeffrey Stedfast  <fejj@ximian.com>

* providers/imap/camel-imap-utils.c (imap_uid_array_to_set): Added
a note to the comment block for this function noting that uids
MUST be in sorted order.

* providers/imap/camel-imap-folder.c (imap_expunge_uids_online):
Sort the uids before splitting them into UID set chunks. The code
that splits the uid array into sets expects the uids to be in
sorted order. Fixes bug #42854.

* providers/imap/camel-imap-command.c (imap_command_start):
Re-enable camel_verbose_debug logging.

21 years agoIf host->h_name is NULL or an empty string, default back to using the IP
Jeffrey Stedfast [Tue, 13 May 2003 18:10:12 +0000 (18:10 +0000)]
If host->h_name is NULL or an empty string, default back to using the IP

2003-05-12  Jeffrey Stedfast  <fejj@ximian.com>

* providers/smtp/camel-smtp-transport.c (smtp_helo): If
host->h_name is NULL or an empty string, default back to using the
IP address rather than the hostname in the EHLO command.

21 years agoDisable debugging.
Ettore Perazzoli [Tue, 13 May 2003 16:03:19 +0000 (16:03 +0000)]
Disable debugging.

* providers/imap/camel-imap-folder.c (d): Disable debugging.

* providers/imap/camel-imap-command.c (d): Disable debugging.

* providers/imap/camel-imap-search.c (d): Disable debugging.

21 years agoDon't call g_utf8_validate on an empty message. Fixes a warning.
Dan Winship [Sun, 4 May 2003 19:11:39 +0000 (19:11 +0000)]
Don't call g_utf8_validate on an empty message. Fixes a warning.

* camel-mime-part-utils.c
(simple_data_wrapper_construct_from_parser): Don't call
g_utf8_validate on an empty message. Fixes a warning.

21 years agoWorkaround for POS mailers like the one in bug #42045.
Jeffrey Stedfast [Fri, 2 May 2003 17:37:11 +0000 (17:37 +0000)]
Workaround for POS mailers like the one in bug #42045.

2003-05-01  Jeffrey Stedfast  <fejj@ximian.com>

Workaround for POS mailers like the one in bug #42045.

* camel-mime-utils.c (header_decode_date): Make sure the numeric
timezone is between -1200 and 1200, otherwise it is invalid.

* broken-date-parser.c (get_tzone): Make sure the numeric timezone
is between -1200 and 1200, otherwise it is invalid.

21 years agoRemove camel-lock.c from here because it upsets automake 1.6 for some
Dan Winship [Wed, 30 Apr 2003 15:35:28 +0000 (15:35 +0000)]
Remove camel-lock.c from here because it upsets automake 1.6 for some

        * Makefile.am (camel_lock_helper_SOURCES): Remove camel-lock.c
        from here because it upsets automake 1.6 for some reason
        (camel_lock_helper_LDADD): Add camel-lock.o here

21 years agobody_len needs to be a size_t instead of int
Jeremy Katz [Wed, 30 Apr 2003 03:09:15 +0000 (03:09 +0000)]
body_len needs to be a size_t instead of int

2003-04-29  Jeremy Katz  <katzj@redhat.com>

* providers/imap/camel-imap-folder.c (parse_fetch_response):
body_len needs to be a size_t instead of int

21 years agoDisable debugging printfs.
Jeffrey Stedfast [Thu, 24 Apr 2003 17:41:28 +0000 (17:41 +0000)]
Disable debugging printfs.

2003-04-22  Jeffrey Stedfast  <fejj@ximian.com>

* providers/imap/camel-imap-utils.c: Disable debugging printfs.

21 years agoonly free the host if it was set. removes a run-time warning.
Not Zed [Thu, 24 Apr 2003 01:51:20 +0000 (01:51 +0000)]
only free the host if it was set. removes a run-time warning.

2003-04-23  Not Zed  <NotZed@Ximian.com>

        * providers/smtp/camel-smtp-transport.c (smtp_helo): only free the
        host if it was set.  removes a run-time warning.

21 years agofixed merge conflicts in ChangeLog
Jeffrey Stedfast [Wed, 23 Apr 2003 18:16:23 +0000 (18:16 +0000)]
fixed merge conflicts in ChangeLog

21 years ago#if 0 out some code that parsed the namespaces since the results aren't
Jeffrey Stedfast [Wed, 23 Apr 2003 18:14:54 +0000 (18:14 +0000)]
#if 0 out some code that parsed the namespaces since the results aren't

2003-04-22  Jeffrey Stedfast  <fejj@ximian.com>

* providers/imap/camel-imap-store.c (imap_connect_online): #if 0
out some code that parsed the namespaces since the results aren't
used and they generate debugging output.

21 years agooops, the service owns the url after we construct it.
Not Zed [Wed, 23 Apr 2003 04:44:58 +0000 (04:44 +0000)]
oops, the service owns the url after we construct it.

2003-04-23  Not Zed  <NotZed@Ximian.com>

        * camel-session.c (get_service): oops, the service owns the url
        after we construct it.

21 years ago** Should fix #41629, #41448, et al.
Not Zed [Wed, 23 Apr 2003 01:34:03 +0000 (01:34 +0000)]
** Should fix #41629, #41448, et al.

2003-04-22  Not Zed  <NotZed@Ximian.com>

        ** Should fix #41629, #41448, et al.

        * tests/folder/test10.c: a new torture test for object bag
        creation/unreffing.

        * camel-url.c (camel_url_copy): new function to copy a url.

        * camel-object.c (camel_object_bag_new): add arguments for key
        copy and key free functions.  Fixed all callers.
        (camel_object_bag_destroy): fix a memleak, free the bag key.
        (camel_object_bag_get, camel_object_bag_reserve)
        (camel_object_bag_abort, save_bag, save_object): Make the key a
        void type, rather than char *.
        (camel_object_bag_add): As above, and also copy the key.
        (camel_object_bag_remove_unlocked): free the key using
        bag->free_key.

        * camel-session.c (register_provider)
        (camel_session_destroy_provider, get_service): Changed to use an
        object bag instead of a hash table for the service 'cache'.
        (service_cache_remove): Removed, no longer required.

21 years agoDon't set seen_eof1 here anymore once we get a trust metric.
Jeffrey Stedfast [Tue, 22 Apr 2003 18:13:49 +0000 (18:13 +0000)]
Don't set seen_eof1 here anymore once we get a trust metric.

2003-04-21  Jeffrey Stedfast  <fejj@ximian.com>

* camel-gpg-context.c (gpg_ctx_parse_status): Don't set seen_eof1
here anymore once we get a trust metric.
(gpg_ctx_new): Init seen_eof1 to TRUE here.
(gpg_ctx_set_ostream): Change seen_eof1 to FALSE here this way we
only ever have to set this if we are expecting output.
(gpg_ctx_parse_status): Don't set seen_eof1 for importing either.
(gpg_ctx_op_step): Only FD_SET() those fd's that we have not yet
finished reading.

21 years ago** for #40989
Not Zed [Sat, 19 Apr 2003 03:15:54 +0000 (03:15 +0000)]
** for #40989

2003-04-17  Not Zed  <NotZed@Ximian.com>

        ** for #40989

        * camel-text-index.c (text_index_delete): delete the block
        file/key file directly, not just its files.

        * providers/local/camel-local-folder.c (local_delete): implement,
        just delete the index file if it exists.
        camel-store-delete-folder will delete other data (maybe it all
        should be done here).

        * camel-block-file.c (camel_key_file_finalise): keep lock around
        decrementing key file use count.
        (camel_key_file_delete): new function to delete the key file (&
        close it off).
        (key_file_use): if we've been deleted, always fail.
        (camel_block_file_finalise): only close the file if its a valid
        fd.
        (block_file_use): if we've been deleted, always fail.
        (camel_block_file_delete): delete the block file & close.
        (struct _CamelBlockFilePrivate): fix the !ENABLE_THREADS case to
        still compile this.

21 years agoUrls are unlikely to end with punctuation or closing braces, so strip any
Jeffrey Stedfast [Thu, 17 Apr 2003 16:43:07 +0000 (16:43 +0000)]
Urls are unlikely to end with punctuation or closing braces, so strip any

2003-04-16  Jeffrey Stedfast  <fejj@ximian.com>

* camel-url-scanner.c (camel_url_web_end): Urls are unlikely to
end with punctuation or closing braces, so strip any of those off
the end of the url. Fixes bug #41461.

* tests/mime-filter/data/html.1.out: Removed a trailing \n at the
end of the file, this is incorrect (since out input file does not
contain one).

* camel-mime-filter-tohtml.c (html_convert): Only output a <br> if
we found an eoln in the input. Along the same lines, don't write a
'\n' to the output buffer unless we've encountered that eoln in
the input. Fixes bug #41407.

21 years agoDont assume adding 1 to line length will go to the next line. e.g. for dos
Not Zed [Thu, 17 Apr 2003 03:09:37 +0000 (03:09 +0000)]
Dont assume adding 1 to line length will go to the next line. e.g. for dos

2003-04-17  Not Zed  <NotZed@Ximian.com>

        * camel-multipart-signed.c (parse_content): Dont assume adding 1
        to line length will go to the next line.  e.g. for dos lines that
        end in \r\n.  Fix for #41163.
        (parse_boundary): util function to help above.

21 years agoThe gpg child process exiting does not necessarily mean that we've ready
Jeffrey Stedfast [Wed, 16 Apr 2003 17:02:10 +0000 (17:02 +0000)]
The gpg child process exiting does not necessarily mean that we've ready

2003-04-15  Jeffrey Stedfast  <fejj@ximian.com>

* camel-gpg-context.c (gpg_verify): The gpg child process exiting
does not necessarily mean that we've ready everything that we can
from its pipes, so don't use gpg_ctx_op_exited() as a loop-exit
condition. If for some reason the gpg child process does exit (due
to an error), gpg_ctx_op_step() will eventually fail (as soon as
it has finished reading any data in the pipes) and cause the loop
to be terminated anyway. This fixes truncation of the gpg --verify
stderr output that Evolution displays as the "validity report"
(for lack of a better description).
(gpg_ctx_op_step): Removed some debugging printf's
(gpg_ctx_get_diagnostics): Use a bitfield to decide if we've
already flushed the diagnostics stream, much simpler (plus we have
plenty of bits to spare so might as well use them).
(gpg_ctx_new): Fixed another logic bug which determined whether or
not we should convert the gpg output from the locale charset into
UTF-8.

21 years agoTurn off debugging if CAMEL_VERBOSE_DEBUG isn't set.
Jeffrey Stedfast [Thu, 10 Apr 2003 17:13:15 +0000 (17:13 +0000)]
Turn off debugging if CAMEL_VERBOSE_DEBUG isn't set.

2003-04-10  Jeffrey Stedfast  <fejj@ximian.com>

* providers/smtp/camel-smtp-transport.c: Turn off debugging if
CAMEL_VERBOSE_DEBUG isn't set.

21 years agofilter data test cases.
Not Zed [Thu, 10 Apr 2003 11:35:00 +0000 (11:35 +0000)]
filter data test cases.

2003-04-10  Not Zed  <NotZed@Ximian.com>

        * tests/mime-filter/data: filter data test cases.

        * tests/mime-filter/test-tohtml.c: New test for html filter.

        ** See bug #40969

        * camel-mime-filter-tohtml.c (html_convert): Change the logic
        slightly, scan a whole line within the main loop.

21 years agoRemoved. I just realised that we ALWAYS want to get gpg's stderr in UTF-8,
Jeffrey Stedfast [Wed, 9 Apr 2003 23:21:15 +0000 (23:21 +0000)]
Removed. I just realised that we ALWAYS want to get gpg's stderr in UTF-8,

2003-04-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-gpg-context.c (gpg_ctx_get_utf8_diagnostics): Removed. I
just realised that we ALWAYS want to get gpg's stderr in UTF-8, so
the better solution (plus less code duplication) is to just have
the diagnostics buffer be a CamelStreamMem with a charset filter.
(gpg_ctx_new): Setup a filter stream to convert our gpg stderr
diagnostics into UTF-8 if needed.
(gpg_ctx_free): Unref the diagnostcis stream.
(gpg_ctx_parse_status): Flush the diagnostics stream if we are
going to use the data in an exception.
(gpg_ctx_op_step): Use camel_stream_write() to append to the
diagnostics stream rather than appending the buffer to a
GByteArray like we used to do.
(gpg_ctx_op_step): Flush the diagnostics stream here too.
(gpg_verify): Use a const char *diagnostics string here, forcing a
slight rearrangement of calls - use the const char *dignostics
before we gpg_ctx_free, otherwise we'll be passing free'd memory
off to the cipher_validity.
(gpg_ctx_get_diagnostics): Do the work here to flush the
diagnostics stream and nul-terminate the buffer. Also return const
now, instead of strdupping. No need for that.
(gpg_ctx_parse_status): Convert the user hint to UTF-8.

21 years agoChanged my mind a bit on how I wanted this to work. Instead of aborting on
Jeffrey Stedfast [Wed, 9 Apr 2003 22:04:33 +0000 (22:04 +0000)]
Changed my mind a bit on how I wanted this to work. Instead of aborting on

2003-04-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-gpg-context.c (gpg_ctx_get_utf8_diagnostics): Changed my
mind a bit on how I wanted this to work. Instead of aborting on an
illegal sequence, do like we do with camel-mime-filter-charset and
just skip over invalid sequences. Also, in the noop failure case,
close the iconv_t so we don't leak it.

21 years agoSave the content size in a temp variable until after we've successfully
Jeffrey Stedfast [Wed, 9 Apr 2003 21:02:34 +0000 (21:02 +0000)]
Save the content size in a temp variable until after we've successfully

2003-04-09  Jeffrey Stedfast  <fejj@ximian.com>

* providers/imap/camel-imap-utils.c (imap_body_decode): Save the
content size in a temp variable until after we've successfully
parsed all of the body_type_1part expr. Also fixed a type-o in the
body_type_mpart case that caused the parser to fail.
(imap_parse_body): On failure to parse the body, as we iterate
through the child nodes, set their children/parent/next nodes to
NULL so that content_info_free() won't double-free any of the
other nodes we have already free'd (or are about to free).

21 years agoNew function to get the UTF-8 version of the diagnostics (if the locale
Jeffrey Stedfast [Wed, 9 Apr 2003 19:27:17 +0000 (19:27 +0000)]
New function to get the UTF-8 version of the diagnostics (if the locale

2003-04-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-gpg-context.c (gpg_ctx_get_utf8_diagnostics): New function
to get the UTF-8 version of the diagnostics (if the locale isn't
already UTF-8 safe).
(gpg_verify): Use the utf-8 diagnostics here. Also fixed a memleak
in the exception case.

21 years agoNULL-protect the to/from charsets in the g_warning call.
Jeffrey Stedfast [Wed, 9 Apr 2003 05:55:45 +0000 (05:55 +0000)]
NULL-protect the to/from charsets in the g_warning call.

2003-04-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-filter-charset.c
(camel_mime_filter_charset_new_convert): NULL-protect the to/from
charsets in the g_warning call.

21 years agoFixes for bug #40778. Probably others as well.
Jeffrey Stedfast [Wed, 9 Apr 2003 00:09:24 +0000 (00:09 +0000)]
Fixes for bug #40778. Probably others as well.

2003-04-08  Jeffrey Stedfast  <fejj@ximian.com>

Fixes for bug #40778. Probably others as well.

* providers/imap/camel-imap-utils.c (parse_params): Return 0 on
success, or -1 on fail. The other small utility functions should
probably do the same, but I was too lazy to fix them. Setting
inptr to NULL is just lame.
(imap_body_decode): New function to correctly parse a BODY
response.
(imap_parse_body): Rewritten to use the above function.

21 years agoMake sure host->h_name is not an empty string too? Apparently for one user
Jeffrey Stedfast [Tue, 8 Apr 2003 20:20:04 +0000 (20:20 +0000)]
Make sure host->h_name is not an empty string too? Apparently for one user

2003-04-08  Jeffrey Stedfast  <fejj@ximian.com>

* providers/smtp/camel-smtp-transport.c (smtp_helo): Make sure
host->h_name is not an empty string too? Apparently for one user
his resolved hostname is "". Go figure. Probably a broken system,
but easy enough to work around.

21 years agoIf store->engine->line is NULL, use _("Unknown error") as the error string
Jeffrey Stedfast [Tue, 8 Apr 2003 17:48:12 +0000 (17:48 +0000)]
If store->engine->line is NULL, use _("Unknown error") as the error string

2003-04-08  Jeffrey Stedfast  <fejj@ximian.com>

* providers/pop3/camel-pop3-store.c (pop3_try_authenticate): If
store->engine->line is NULL, use _("Unknown error") as the error
string instead.

21 years agouse g_strerror, and use errno directly rather than ferrno().
Not Zed [Tue, 8 Apr 2003 00:55:02 +0000 (00:55 +0000)]
use g_strerror, and use errno directly rather than ferrno().

2003-04-08  Not Zed  <NotZed@Ximian.com>

        * camel-folder-summary.c (camel_folder_summary_load): use
        g_strerror, and use errno directly rather than ferrno().
        (summary_header_load): set errno when we get a version mismatch.

21 years agoDisable debug printf's
Jeffrey Stedfast [Mon, 7 Apr 2003 21:59:27 +0000 (21:59 +0000)]
Disable debug printf's

2003-04-07  Jeffrey Stedfast  <fejj@ximian.com>

* camel-gpg-context.c: Disable debug printf's

21 years ago(e_passwords_remember_password): Removed debugging
Ettore Perazzoli [Sun, 6 Apr 2003 00:18:09 +0000 (00:18 +0000)]
(e_passwords_remember_password): Removed debugging
message that would spit out the actual password, albeit in encoded
form.

21 years agoDelete all Passwords-* sections. Should make this work.
Not Zed [Wed, 2 Apr 2003 04:48:22 +0000 (04:48 +0000)]
Delete all Passwords-* sections. Should make this work.

2003-04-02  Not Zed  <NotZed@Ximian.com>

* e-passwords.c (e_passwords_forget_passwords): Delete all
Passwords-* sections.  Should make this work.

21 years agoIf we are using the heimdal krb5 implementation, don't free outbuf ever.
Jeffrey Stedfast [Tue, 1 Apr 2003 23:42:58 +0000 (23:42 +0000)]
If we are using the heimdal krb5 implementation, don't free outbuf ever.

2003-03-28  Jeffrey Stedfast  <fejj@ximian.com>

* camel-sasl-gssapi.c (gssapi_challenge): If we are using the
heimdal krb5 implementation, don't free outbuf ever. Seems to
segfault if we do.

21 years agoUse strncasecmp for matching the nofill and /nofill tags. Also fixed to
Jeffrey Stedfast [Mon, 31 Mar 2003 21:54:46 +0000 (21:54 +0000)]
Use strncasecmp for matching the nofill and /nofill tags. Also fixed to

2003-03-31  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-filter-enriched.c (enriched_to_html): Use strncasecmp
for matching the nofill and /nofill tags. Also fixed to handle
enriched tags that require <param> tags.

21 years agoremove unused var.
Not Zed [Fri, 28 Mar 2003 00:13:46 +0000 (00:13 +0000)]
remove unused var.

2003-03-28  Not Zed  <NotZed@Ximian.com>

* providers/local/camel-spool-store.c (get_folder): remove unused
var.

* providers/imap/camel-imap-utils.c (imap_parse_list_response):
moved an unused var to the #if 0'd out block.

* providers/imap/camel-imap-store.c
(parse_list_response_as_folder_info): remove unused vars.

* camel-index-control.c (main): deifne camel_init() before using
it.

* camel-vee-store.c (vee_get_folder): removed unused var.

* camel-stream-mem.c (stream_write): warning -> fixme.

* camel-stream-filter.c (_CamelStreamFilterPrivate): Add a ; to
the end of the struct definition.

* camel-store.c (camel_store_delete_folder): remove unused var.
(camel_store_rename_folder): "
(camel_store_unsubscribe_folder): "

* camel-mime-part.c (write_to_stream): Changed warning into a
TODO.
(construct_from_parser): removed warning, the code was fixed.

* camel-mime-message.c (construct_from_parser): Remove the
warning, the code was fixed.

* camel-mime-filter-enriched.c (camel_mime_filter_enriched_new):
Remove unused var.

* camel-medium.c (camel_medium_get_header): Removed warning, and
changed docs to reflect it.

* camel-gpg-context.c (gpg_ctx_parse_status): handle MODE_EXPORT
as a noop (warning).

* camel-folder.c (get_message_user_tag): changed a warning into a
fixme (its not that important).
(camel_folder_get_message_user_tag): "

* camel-data-cache.c (data_cache_finalise): removed unused vars.
(data_cache_expire): "

21 years agoGet rid of debug printf's that are no longer needed. In the case of
Jeffrey Stedfast [Thu, 27 Mar 2003 21:13:13 +0000 (21:13 +0000)]
Get rid of debug printf's that are no longer needed. In the case of

2003-03-27  Jeffrey Stedfast  <fejj@ximian.com>

* camel-sasl-gssapi.c (gssapi_challenge): Get rid of debug
printf's that are no longer needed. In the case of errors, don't
release the outbuf gss_buffer_t since it shouldn't be set. Also
g_free (str) when we are done with it.

21 years agoWhen we remove a node from the list, make sure to g_list_free_1().
Jeffrey Stedfast [Thu, 27 Mar 2003 20:16:57 +0000 (20:16 +0000)]
When we remove a node from the list, make sure to g_list_free_1().

2003-03-27  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-utils.c (header_encode_phrase_merge_words): When we
remove a node from the list, make sure to g_list_free_1().
(header_encode_phrase_merge_words): Don't use
CAMEL_FOLD_PREENCODED as the upper-bound for merged-word length if
the merged-word will not be an rfc2047 encoded word. Instead, use
CAMEL_FOLD_SIZE-8 (which is the value we use in other places for
determining upper-bound lengths). Solves bug #38659.

21 years agoLike camel_url_encode, but works directly on an existing GString.
Dan Winship [Thu, 27 Mar 2003 15:37:47 +0000 (15:37 +0000)]
Like camel_url_encode, but works directly on an existing GString.

* camel-url.c (append_url_encoded): Like camel_url_encode, but
works directly on an existing GString.
(camel_url_to_string, output_param): Use it.
(camel_url_encode): Likewise. Remove "escape_unsafe" arg since the
"unsafe" chars are the ones that should *always* be escaped, and
the places we were passing FALSE were wrong.
(camel_url_decode): replace with a new version

* camel-file-utils.c (camel_file_util_safe_filename): Remove extra
arg to camel_url_encode.

* tests/misc/url.c (main): Add tests of basic URL parsing and
unparsing

21 years agoDo a more thorough check for html marked as text/plain. Check that the
Not Zed [Thu, 27 Mar 2003 06:05:46 +0000 (06:05 +0000)]
Do a more thorough check for html marked as text/plain. Check that the

2003-03-28  Not Zed  <NotZed@Ximian.com>

*
camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser):
Do a more thorough check for html marked as text/plain.  Check
that the text starts with <html or <!doctype.  More complete fix
for #16817.

21 years agoPass in the addresses of the second args to the delete/release functions.
Jeffrey Stedfast [Wed, 26 Mar 2003 17:58:48 +0000 (17:58 +0000)]
Pass in the addresses of the second args to the delete/release functions.

2003-03-26  Jeffrey Stedfast  <fejj@ximian.com>

* camel-sasl-gssapi.c (camel_sasl_gssapi_finalize): Pass in the
addresses of the second args to the delete/release functions.

21 years agoIgnore whether or not the bag is currently reserved. We'll miss the new
Not Zed [Wed, 26 Mar 2003 07:01:38 +0000 (07:01 +0000)]
Ignore whether or not the bag is currently reserved. We'll miss the new

2003-03-27  Not Zed  <NotZed@Ximian.com>

* camel-object.c (camel_object_bag_list): Ignore whether or not
the bag is currently reserved.  We'll miss the new potential
object, but only 1.  Should address #40103.

21 years agoDon't use setv when we don't need the stdarg interface.
Jeffrey Stedfast [Tue, 25 Mar 2003 16:44:34 +0000 (16:44 +0000)]
Don't use setv when we don't need the stdarg interface.

2003-03-25  Jeffrey Stedfast  <fejj@ximian.com>

* camel-service.c (camel_gethostbyaddr): Don't use setv when we
don't need the stdarg interface.
(camel_gethostbyname): Same.

21 years agoThis expression can never return an array. We must always return a string
Jeffrey Stedfast [Mon, 24 Mar 2003 19:34:11 +0000 (19:34 +0000)]
This expression can never return an array. We must always return a string

2003-03-24  Jeffrey Stedfast  <fejj@ximian.com>

* camel-folder-search.c (search_user_tag): This expression can
never return an array. We must always return a string value.

21 years agoremoved invisible 8bit chars
Jeffrey Stedfast [Mon, 24 Mar 2003 19:09:28 +0000 (19:09 +0000)]
removed invisible 8bit chars

21 years agocamel-imap-command.c (imap_read_untagged) Integer overflow fix. If server
  Timo Sirainen  [Mon, 24 Mar 2003 19:05:40 +0000 (19:05 +0000)]
camel-imap-command.c (imap_read_untagged) Integer overflow fix. If server

2003-03-24  Timo Sirainen  <tss@iki.fi>

* camel-imap-command.c (imap_read_untagged) Integer overflow fix.
If server sent a huge literal length, only a few bytes of memory
was allocated to it, but server could write as much data there as
it wanted.

21 years agoCamel part of the fix for Lewing's bug #39204. Second half of the patch
Jeffrey Stedfast [Fri, 21 Mar 2003 21:54:59 +0000 (21:54 +0000)]
Camel part of the fix for Lewing's bug #39204. Second half of the patch

2003-03-21  Jeffrey Stedfast  <fejj@ximian.com>

Camel part of the fix for Lewing's bug #39204. Second half of the
patch should also fix it so that text/plain parts that are
detected as being text/html are rendered as html (not that I agree
with this mind you, but corporate customers want this "feature").

* camel-mime-part-utils.c
(simple_data_wrapper_construct_from_parser): If the part is
text/html and doesn't have a charset and we are able to find a
charset in the meta tags of the html content, set the charset
param on the content-type.
(camel_mime_part_construct_content_from_parser): After setting the
content object on the mime part, re-set the content-type of the
content object back to the one from the parser (as it may contain
changes from the original - see changes made to the ct in
simple_data_wrapper_construct_from_parser for details).

21 years agocontinue reading the cert file until we've read it all or get a non-EINTR
Jeffrey Stedfast [Fri, 21 Mar 2003 18:54:52 +0000 (18:54 +0000)]
continue reading the cert file until we've read it all or get a non-EINTR

2003-03-21  Jeffrey Stedfast  <fejj@ximian.com>

* camel-tcp-stream-ssl.c (camel_certdb_nss_cert_get): continue
reading the cert file until we've read it all or get a non-EINTR
errno.

21 years ago#include <com_err.h> instead of #include <et/com_err.h>
Jeffrey Stedfast [Fri, 21 Mar 2003 15:59:40 +0000 (15:59 +0000)]
#include <com_err.h> instead of #include <et/com_err.h>

2003-03-21  Jeffrey Stedfast  <fejj@ximian.com>

* camel-sasl-gssapi.c: #include <com_err.h> instead of
#include <et/com_err.h>

21 years agoPlug in GSSAPI support.
Jeffrey Stedfast [Thu, 20 Mar 2003 19:37:38 +0000 (19:37 +0000)]
Plug in GSSAPI support.

2003-03-20  Jeffrey Stedfast  <fejj@ximian.com>

* camel-sasl.c: Plug in GSSAPI support.

* camel-sasl-gssapi.[c,h]: Various fixes to make it compile (fixed
type-o's mostly).

21 years ago Handle raw 8-bit From data "correctly". (The same way we handle
Dan Winship [Thu, 20 Mar 2003 16:43:29 +0000 (16:43 +0000)]
Handle raw 8-bit From data "correctly". (The same way we handle
raw 8-bit Subject data.)

* camel-mime-utils.c (header_decode_mailbox): Take a charset arg
and pass it to header_decode_string.
(header_decode_address): Take a charset arg and pass it to
header_decode_mailbox.
(header_mailbox_decode): Likewise.
(header_address_decode): Take a charset arg and pass it to
header_decode_address.

* camel-folder-summary.c (summary_format_address): Take a charset
arg and pass to header_address_decode.
(message_info_new, camel_message_info_new_from_header): Pass
charset to summary_format_address

* camel-internet-address.c (internet_decode): Update for
header_address_decode change. (Unfortunately we don't have a
charset to pass here.)

* camel-mime-message.c (camel_mime_message_build_mbox_from): Move
this here from camel-mbox-summary since the same functionality is
needed by evolution-mail too (and update for header_address_decode
change)

* providers/local/camel-mbox-summary.c
(camel_mbox_summary_build_from): Moved to CamelMimeMessage

21 years agoRemoved unused variable left over from my previous fix.
Jeffrey Stedfast [Mon, 17 Mar 2003 16:53:57 +0000 (16:53 +0000)]
Removed unused variable left over from my previous fix.

2003-03-17  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-part.c (process_header): Removed unused variable left
over from my previous fix.

* providers/smtp/camel-smtp-transport.c (smtp_send_to): Don't pass
'has_8bit_parts' to smtp_data() anymore.
(smtp_data): No longer takes 'has_8bit_parts' argument. Ignore
whether or not the message has 8bit parts when deciding what the
required encoding type we need to enforce. Fixes bug #39744.

21 years agoUse header_contentid_decode() as this new function should be safer than
Jeffrey Stedfast [Fri, 14 Mar 2003 17:44:20 +0000 (17:44 +0000)]
Use header_contentid_decode() as this new function should be safer than

2003-03-13  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-part.c (process_header): Use
header_contentid_decode() as this new function should be safer
than the hack that we had before.

* camel-mime-utils.c (header_contentid_decode): New function to
try and parse a content-id string in such a way as to work around
some of the known bugs in other MIME implementations. Try to be as
"safe" as we can - ie. don't allow for more than 1 @ (since the
mailer uses "@@@%d" as a fake content-id value for parts without
content-ids) and don't allow for invalid content-type chars.

21 years agoDon't add bogus uids to the uid array. Might fix bug #38868 (it's the only
Jeffrey Stedfast [Thu, 13 Mar 2003 22:16:50 +0000 (22:16 +0000)]
Don't add bogus uids to the uid array. Might fix bug #38868 (it's the only

2003-03-13  Jeffrey Stedfast  <fejj@ximian.com>

* camel-folder.c (get_uids): Don't add bogus uids to the uid
array. Might fix bug #38868 (it's the only way I can figure that
camel_folder_get_message_info() could possibly return NULL for the
Outbox folder).

21 years agoremoved, it was double-freeing the key. (save_object): Just save the
Not Zed [Wed, 12 Mar 2003 00:37:33 +0000 (00:37 +0000)]
removed, it was double-freeing the key. (save_object): Just save the

2003-03-12  Not Zed  <NotZed@Ximian.com>

        * camel-object.c (remove_bag): removed, it was double-freeing the
        key.
        (save_object): Just save the object in an array.
        (camel_object_bag_destroy): first save the object bag's objects in
        a list, then remove them one at a time from the bag since we can't
        remove hash table entries while we're in a foreach (PITA glib
        shit).  For #39486.
        (camel_object_bag_*): killed some warnings.

21 years agoadded a comment
Jeffrey Stedfast [Mon, 10 Mar 2003 16:48:47 +0000 (16:48 +0000)]
added a comment

21 years agoDoh! If inptr == pos + 1, then return FALSE (before it was checking for
Jeffrey Stedfast [Sun, 9 Mar 2003 17:57:34 +0000 (17:57 +0000)]
Doh! If inptr == pos + 1, then return FALSE (before it was checking for

2003-03-09  Jeffrey Stedfast  <fejj@ximian.com>

* camel-url-scanner.c (camel_url_addrspec_end): Doh! If inptr ==
pos + 1, then return FALSE (before it was checking for inptr ==
pos, but it will never be pos).

21 years agohandle a null return from search_by_expression, for bug #33786.
Not Zed [Fri, 7 Mar 2003 03:38:45 +0000 (03:38 +0000)]
handle a null return from search_by_expression, for bug #33786.

2003-03-07  Not Zed  <NotZed@Ximian.com>

        * camel-vee-folder.c (vee_search_by_expression): handle a null
        return from search_by_expression, for bug #33786.

21 years agos/g_strncasecmp/strncasecmp/g
Jeffrey Stedfast [Thu, 6 Mar 2003 21:42:11 +0000 (21:42 +0000)]
s/g_strncasecmp/strncasecmp/g

2003-03-06  Jeffrey Stedfast  <fejj@ximian.com>

* providers/imap/camel-imap-utils.c (imap_next_word):
s/g_strncasecmp/strncasecmp/g

* camel-iconv.c: Updated to be an exact copy of e-iconv (except
names changed). We don't use this yet, but we may in the future.

* camel-url.c (camel_url_new_with_base): Here too.

* camel-sasl-kerberos4.c (krb4_challenge): Use camel_strdown()
here instead of g_strdown() since the latter has been deprecated.

* providers/imap/camel-imap-utils.c (imap_parse_body): Use
camel_strdown() since g_ascii_strdown() does not do what we
thought.

* providers/imap/camel-imap-store.c:
s/strstrcase/camel_strstrcase/g

* providers/imap/camel-imap-folder.c (do_append):
s/strstrcase/camel_strstrcase/
(handle_copyuid): Same.

* string-utils.c (camel_strdown): New function because the
g_ascii_strdown interface sucks.
(camel_strstrcase): Renamed from strstrcase.

21 years agoSame as below.
Jeffrey Stedfast [Thu, 6 Mar 2003 15:22:56 +0000 (15:22 +0000)]
Same as below.

2003-03-06  Jeffrey Stedfast  <fejj@ximian.com>

* camel-text-index.c (camel_utf8_next): Same as below.

* camel-search-private.c (camel_utf8_getc): Updated since not all
platforms/compiles support __inline__.

21 years agoGet rid of the const return - it mismatches the prototype and we don't
Jeffrey Stedfast [Thu, 6 Mar 2003 15:15:41 +0000 (15:15 +0000)]
Get rid of the const return - it mismatches the prototype and we don't

2003-03-06  Jeffrey Stedfast  <fejj@ximian.com>

* camel-mime-part.c (camel_mime_part_get_encoding): Get rid of the
const return - it mismatches the prototype and we don't need it
anyway. Fixes bug #39173.

21 years agoFixes for bug #39170
Jeffrey Stedfast [Thu, 6 Mar 2003 14:47:05 +0000 (14:47 +0000)]
Fixes for bug #39170

2003-03-06  Jeffrey Stedfast  <fejj@ximian.com>

Fixes for bug #39170

* camel-store-summary.c (camel_store_info_set_string): Remove
return keyword.

* camel-session.c (camel_session_thread_msg_free): Remove return
keyword.
(camel_session_thread_wait): Same.

* camel-index.c (camel_index_cursor_reset): Remove return keyword.

21 years agoFixes for bug #39168
Jeffrey Stedfast [Thu, 6 Mar 2003 14:39:30 +0000 (14:39 +0000)]
Fixes for bug #39168

2003-03-06  Jeffrey Stedfast  <fejj@ximian.com>

Fixes for bug #39168

* providers/local/camel-spool-summary.c: Get rid of the empty
private struct.

* providers/local/camel-mbox-summary.c: Get rid of the empty
private struct.

* camel-vee-store.c: Get rid of the empty private struct.

* camel-mime-filter-from.c: Get rid of the empty private struct.

2003-03-05  Jeffrey Stedfast  <fejj@ximian.com>

* providers/sendmail/camel-sendmail-transport.c
(sendmail_send_to): Add a CRLF filter to the output stream to
convert any CRLF sequences into plain old line-feeds to work
around a bug in some local transport programs.