2004-06-15 Jeffrey Stedfast * providers/imap4/camel-imap4-store.c (imap4_build_folder_info): Make sure we have elements in the array, if not then we're done (return a NULL fi). 2004-06-17 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (imap4_get_uid_set): Fixed to work properly. It was getting ranges wrong before sometimes which was making me lose mail! Ugh. Thanks to Christian Kellner for pointing out these bugs (and submitting the original patch for service_setv) * camel-service.c (service_setv): Don't use (tag & CAMEL_ARG_IGNORE) to determine if we should ignore this tag. CAMEL_ARG_IGNORE is not a bit flag. * providers/imap/camel-imap-store.c (imap_setv): Don't use (tag & CAMEL_ARG_IGNORE) to determine if we should ignore this tag. CAMEL_ARG_IGNORE is not a bit flag. 2004-06-16 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (imap4_refresh_info): Only force a re-update of all FLAGS if this folder wasn't in the SELECTED state. Otherwise, simply send a NOOP. * providers/imap4/camel-imap4-summary.c: Added a 'first' member to the imap_fetch_all_t struct so we can use that as a base offset in our GPtrArray, allowing us to limit resource consumption which could otherwise get quite large. Also added a ChangeInfo member that was needed for changes to untagged_fetch_all(). (imap4_fetch_all_add): Use fetch->first as a base offset and change int i to guint32 i. Also updated to sue the fetch->changes. (imap4_fetch_all_update): Same. (untagged_fetch_all): Same - this is where it is really valuable, since we can avoid adding elements to the GPtrArray that we won't even use. Also needed to change code a big in case index < fetch->first (which could happen if a server notified us of a FLAGS change for a message we didn't request info about). (imap4_fetch_all_free): Free the ChangeInfo. (imap4_summary_fetch_all): Init fetch->changes and fetch->first. (imap4_summary_fetch_flags): Same. (camel_imap4_summary_flush_updates): Only request envelope info if first <= summary->exists. Avoids needless queries. (info_uid_sort): #if 0'd (camel_imap4_summary_flush_updates): No need to sort the summary - this should never have been needed. I can't remember why I did this... 2004-06-15 Jeffrey Stedfast * camel-gpg-context.c (gpg_verify): Use camel_multipart_signed_get_content_stream() rather than getting the first part and canonicalising it ourselves. Fixes bug #60159. * providers/imap4/camel-imap4-stream.c (camel_imap4_stream_init): Init have_unget to FALSE. Don't set unget to NULL, it's no longer a pointer. (camel_imap4_stream_finalize): No need to g_free() unget anymore. (camel_imap4_stream_next_token): Check have_unget rather than unget != NULL. Set have_unget to FALSE if we get an unget'd token. (camel_imap4_stream_unget_token): Don't malloc space for an unget token. The unget token is no longer a pointer. 2004-06-14 Not Zed * providers/smtp/camel-smtp-transport.c (smtp_data): use g_ascii_strcasecmp. (smtp_connect): do not re-helo after we've authenticated. This was a misreading of the spec. We must only re-helo after a transport layer negotiation. 2004-06-12 Jeffrey Stedfast * providers/imap4/camel-imap4-store.c (imap4_noop): Flush summary updates for the currently selected folder. (imap4_noop): Sync the currently selected folder before sending NOOP. * providers/imap4/camel-imap4-summary.c (camel_imap4_summary_set_exists): Don't bother with exists_changed. We don't need it afterall. (camel_imap4_summary_flush_updates): Instead of updating flags if update_flags or exists_changed is set, only bother if update_flags is set or if exists is smaller than the summary count (since updating flags is also sueful for determining which messages have been removed). 2004-06-11 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (imap4_refresh_info): Force updating of the emsage flags (normally this only happens if something has changed that warrants rescanning them). 2004-06-11 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (untagged_fetch): Handle getting FLAGS even though we didn't request it (server can send us FLAGS info if another client changed them recently, for example). Also fixed to handle the fact that not every bit of info has to be in a single untagged FETCH response - it may come in several untagged responses. * providers/imap4/camel-imap4-summary.c (envelope_decode_address): Decode the email address name token. (envelope_decode_nstring): rfc2047 decode strings if requested. (decode_envelope): Request that the subject string be rfc2047 decoded. 2004-06-11 Not Zed * providers/local/camel-maildir-store.c (get_folder_info): if we scan from "" or top == NULL, then we really want to scan from "." instead. * camel-url.c (camel_url_new_with_base): don't check the for a character after the # before extracting the fragment. An empty fragment is still allowed and # should never be added to the path. See Rfc1808 2.4.1. 2004-06-11 Jeffrey Stedfast * providers/imap4/camel-imap4-summary.c (camel_imap4_summary_flush_updates): Don't bother scanning summary info if EXISTS was 0. (camel_imap4_summary_set_uidvalidity): Emit the folder_changed event after clearing the summary. (camel_imap4_summary_expunge): Emit the folder_changed event after removing the message from the summary. (camel_imap4_summary_set_exists): Only set exists_changed if the new and old exists values are different. (imap4_fetch_all_add): Emit a folder_changed signal if any new info's were added. (imap4_fetch_all_update): Emit a folder_changed event if any uids were removed or otherwise updated. (camel_imap4_summary_expunge): Use seqid-1 to determine the actual summary index. (camel_imap4_summary_flush_updates): Added some logic to distinguish between EXISTS value changing because it changed and because messages got expunged. * providers/imap4/camel-imap4-folder.c (imap4_sync): Flush updates after an EXPUNGE and don't unset expunge if we didn't delete anything (the logic was wrong anyway). (imap4_refresh_info): Implemented. 2004-06-10 Jeffrey Stedfast * providers/imap4/camel-imap4-engine.c (camel_imap4_engine_handle_untagged_1): Don't always try and parse a RESP-CODE in the BYE case as the RESP-CODE is optional. 2004-06-10 Jeffrey Stedfast * providers/imap4/camel-imap4-store.c (imap4_build_folder_info): If flags does not include FOLDER_INFO_FAST, get the total/unread counts for the folder-info. * providers/imap4/camel-imap4-engine.c (engine_parse_status): Removed. (camel_imap4_engine_handle_untagged_1): Don't handle untagged STATUS responses anymore. Let the STATUS requestor handle them instead. * providers/imap4/camel-imap4-utils.c (camel_imap4_untagged_status): New function to parse untagged status events. 2004-06-10 Not Zed * camel-filter-driver.c (camel_filter_driver_filter_message): add some :filter debug. (open_folder): only ignore the get_folder exception if we have a default folder, otherwise don't. See #59727. * providers/nntp/camel-nntp-store.c (camel_nntp_command): move the stream based checking into the loop, after we connect. Fixes a crash. 2004-06-09 Jeffrey Stedfast * camel-gpg-context.c (gpg_verify): Fixed a case where it was possible to double-free the gpg context. (gpg_verify): If we don't have a public key, then the signature is just BAD always. 2004-06-09 Jeffrey Stedfast * camel-gpg-context.c (gpg_verify): Don't assign trust to be UNKNOWN if gpg sent us a NODATA status. 2004-06-07 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (camel_imap4_folder_new): Load the entire summary, not just the summary header. This way when the user opens the folder, we don't do a complete re-sync with the server unnecessarily. * providers/imap4/camel-imap4-summary.c (untagged_fetch_all): New info's always have a uid of "", so don't checkagainst NULL. 2004-06-06 Jeffrey Stedfast * providers/imap4/camel-imap4-store.c (imap4_rename_folder): Implemented, mostly. Still need to update state on the renamed folder object if instantiated and rename the on-disk cache directory. * providers/imap4/camel-imap4-folder.c (camel_imap4_folder_finalize): Free the cachedir. (camel_imap4_folder_new): Init the cachedir and load the saved summary before updating it against the server summary. * providers/imap4/camel-imap4-store.c (camel_imap4_store_finalize): Free the storage_path. (imap4_construct): Init the storage_path. 2004-06-04 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (camel_imap4_folder_new): Initialise the folder->summary and force an update of the message info cache by selecting the folder and flushing the updates to the imap4 summary object. * providers/imap4/camel-imap4-store.c (imap4_get_folder_info): Lock the connect_lock before we check if the engine is NULL and after we connect (assuming we need to), initialise the engine pointer again. (imap4_get_folder_info): Doh. Need to escape the "'s in the LIST command string. (imap4_get_folder): Same. * camel-stream-buffer.c (stream_flush): Fixed to work properly. After we've flushed the buffer, we want to set sbf->ptr to sbf->buf, we don't want to do sbf->ptr += written, that'll just corrupt our next write. * providers/imap4/camel-imap4-store.c (connect_to_server_wrapper): Duh. If the user doesn't care about SSL/TLS - use USE_SSL_NEVER, not USE_SSL_ALWAYS. * camel-tcp-stream-ssl.c (stream_flush): Always just return 0, don't try to PR_Sync() - fsync on a socket causes an error. * providers/imap4/camel-imap4-command.c (camel_imap4_command_step): Set exceptions when write/flush fail. * providers/imap4/camel-imap4-engine.c (camel_imap4_engine_take_stream): Set an exception in the case where we get an unexpected greeting from the server. * providers/imap4/camel-imap4-store.c (imap4_create_folder): store->dir_sep no longer exists, so query the engine for the directory separator for the parent_folder. (imap4_build_folder_info): CamelFolderInfo no longer has a path component. 2004-06-03 Not Zed * providers/nntp/camel-nntp-store.c (camel_nntp_command): disconnect if we get an io error or user cancellation. (nntp_disconnect_online): reset current folder. (connect_to_server): and here too just to make sure. * providers/nntp/camel-nntp-folder.c (nntp_folder_sync_online): only save the summary, don't update from server, thats what refresh info does. (nntp_folder_download_message): fix exception handling. (nntp_folder_cache_message): same. (nntp_folder_get_message): ditto, plus major cleanup. (nntp_folder_download_message): take combined uid so it can cache and lookup properly. duh. * providers/nntp/camel-nntp-store.c (nntp_store_get_subscribed_folder_info): if not fast, then open the folder, and update it. Yeah i've given up trying to worry about performance vs usability. * providers/nntp/camel-nntp-summary.c (camel_nntp_summary_check): update the storesummary if we update the folder summary. Hmm, isn't duplicated data meant to be a bad thing? :P * providers/nntp/camel-nntp-store.c (camel_nntp_store_set_folder): removed, now handled by nntp_command. (nntp_connected): removed, now handled by nntp_command. * camel-string-utils.c (camel_tolower): added ascii to-lower function. (camel_toupper): and upper, for completeness. * camel-store-summary.c (CAMEL_STORE_SUMMARY_VERSION): bumped file version by 1. This is a mess, version 1 files treated the bitfield 'flags' with bit number values not bits. Messy. * providers/nntp/camel-nntp-store-summary.c (store_info_save): write last/first count. (CAMEL_NNTP_STORE_SUMMARY_VERSION): bump version to 1. (store_info_load): if we're loading >= version 1, then load last/first counts. * providers/nntp/camel-nntp-store.c (nntp_store_get_folder_info_all): pass the whole line to store_info_from_line, dont strip last/first info. (nntp_store_info_update): renamed from info_new_from_line. only add if not present. handle updates, try and handle unread counts and readonly status. 2004-06-02 Not Zed * providers/nntp/camel-nntp-store.c: setup xover once we've started. * providers/nntp/camel-nntp-summary.c: (xover_setup): moved to nntp store. * providers/nntp/camel-nntp-folder.c (folder_check) (folder_check_free, camel_nntp_folder_new): remove async summary stuff. * providers/nntp/camel-nntp-store.c (camel_nntp_command): take exception argument again, and folder argument. do retry logic and auth logic differently. (camel_nntp_raw_command): raw command interface, dont try reconnect or anything fancy. pass i/o errors straight out, etc. (camel_nntp_try_authenticate): change to return return codes & take exception. * providers/nntp/camel-nntp-summary.c (camel_nntp_summary_new): just take path argument. (camel_nntp_summary_check): take a store, and a folder name. (add_range_head, add_range_xover): remove the time based update events, they never had any effect anyway. Take store argument. (xover_setup): take store argument. * camel-folder-search.c (search_match_threads): remove debug. 2004-06-01 Not Zed ** A few fixes for better rfc compliance, and cleaner code. * camel-mime-utils.c (header_encode_param): a bunch of logic cleanups with new util functions. (header_decode_init): setup a new type ATTR_CHAR, for attribute-char. * tests/misc/test2.c (main): new test for rfc2184 stuff. * camel-mime-utils.c (header_convert): helper to convert between charsets. (rfc2184_decode): fix a bunch of logic problems and use the helper above to simplify code. (decode_param_token): removed, not needed. (header_decode_rfc2184_param): removed, not needed. (header_decode_param): removed, not needed. ugh. (header_decode_param_list): completely rewritten, hence lack of need of above. 2004-05-31 Jeffrey Stedfast * camel-mime-filter-gzip.c (camel_mime_filter_gzip_finalize): Don't leak the zlib stream internals. 2004-05-27 Jeffrey Stedfast Fixes bug #59191. * providers/imap/camel-imap-store.c (camel_imap_store_readline): Don't override the exception with g_strerror() since presumably camel_imap_store_is_connected() will have set that for us (and will have a much more accurate exception anyway). 2004-05-26 Dan Winship * camel-store.c (camel_store_get_trash): If the store is not a vtrash store, just invoke the virtual method. (In particular, don't assume that the trash folder's name is CAMEL_VTRASH_NAME). If it is a vtrash store, just let camel_store_get_folder() do the work since it's duplicated there anyway. #57114 (camel_store_get_junk): Likewise. 2004-05-25 Sivaiah nallagatla * providers/groupwise/came;-gw-listener.c (add_addressbook_sources): set port property on e-source 2004-05-25 Sivaiah Nallagatla * providers/groupwise/camel-gw-listener.c (lookup_account_info): return NULL when there is no existing gw account with same uid 2004-05-25 Not Zed * camel-store.c (camel_folder_info_build_path): removed. * camel-store.h (CamelFolderInfo): removed 'path', fixed all callers. (struct _CamelStore): removed 'dir_sep' fixed all uses. 2004-05-25 Sivaiah Nallagatla * providers/groupwise/camel-gw-listener.c (add_addressbook_sources) (add_calendar_tasks_sources) : remove /soap part from uri. make port as a e-source property instead of putting it in uri to amek the uris same as that of mail for password sharing 2004-05-24 Not Zed * camel-mime-utils.c (camel_header_newsgroups_decode): fix/rearrange logic. It was using the wrong working pointer. 2004-05-24 Sivaiah Nallagatla * camel-provider.h: Added CAMEL_PROVIDER_CONF_HIDDEN to conf item type enum. This is used by groupwise provider * providers/groupwise/camel-groupwise-provider.c: Added CAMEL_PROVIDER_CONF_HIDDEN to groupwise_conf_entries to pass the auth-domain value * providers/imap/camel-imap-store.c (imap_auth_loop): Read the auth-domain property from imap url and pass it camel sesstion apis 2004-05-22 Not Zed * camel-store.c: oops, forgot folder_created. Removed locking, the object bag serialises stuff for us. (camel_store_get_junk): same. ** Another unread count bug, #58814. * camel-store.c (camel_store_class_init): added a folder_opened event. (camel_store_get_folder): emit a folder_opened event whenever we [re] open the physical folder. 2004-05-21 Jeffrey Stedfast * providers/imap/camel-imap-store.c (connect_to_server): Added a workaround for broken IMAP. Hopefully by forcing only the IMAP4 command subset, we can work around their suckage. 2004-05-21 Not Zed * camel-session.c (camel_session_get_password): added a 'domain' argument, and rearragned arguments to be prettier and more consistent. Fixed all callers. (camel_session_forget_password): added a domain argument. Fixed all callers. ** See #58376. * camel-folder.c (set_message_flags): if system flags change, then don't trigger a folder changed event. * camel-folder-summary.h (CAMEL_MESSAGE_SYSTEM_MASK): added this to indicate which flags are internal/apps not interested in. * camel-folder.c (filter_free): rearrange and use some helpers. (folder_changed): if we're frozen, dont go firing off threads to do any processing on each change, wait until we're called unfrozen. Slight code rearragnement. (filter_filter): add progress to junk learn/unlearn, and separate them. 2004-05-19 Suresh Chandrasekharan Fix for #58738 ja_JP.UTF-8: Evolution crashes when certain ASCII/non-ASCII combination is used in mail subject * camel-mime-utils.c (camel_header_encode_string) Use camel_mime_is_lwsp for determining word separators, according to rfc822 (which's also same for rfc2047). g_unichar_isspace as word separator is illegal. 2004-05-20 Jeffrey Stedfast Fixes bug #42295 and the infinite loop part of bug #58766 (these 2 bugs are almost identical, except the server responses are broken in different ways). * providers/imap/camel-imap-folder.c (imap_update_summary): Remove the kludge to re-SELECT the folder to force a re-FETCH of message-info's. This 1) doesn't do what it was meant to do and 2) has a tendency to cause infinite loops with broken servers such as Courier-IMAP. (imap_update_summary): Rework the loop that adds messages to the summary such that if we encounetr an error, we break out and set an exception (we can keep the messages up to the point of failure, but none after that because otherwise our uid-to-seqid mapping would be inconsistant with that of the server and could potentially cause data loss). 2004-05-20 Not Zed * camel-exception.c (camel_exception_setv): re-arrange the code so exception debug will print the expanded description. (camel_exception_set): print exception debug. * providers/pop3/camel-pop3-folder.c (pop3_get_message): same. * providers/local/camel-mh-folder.c (mh_get_message) * providers/local/camel-mbox-folder.c (mbox_get_message): * providers/local/camel-maildir-folder.c (maildir_get_message): Don't use INVALID_UID for errors which are more system related. And sync up all the error messages. * providers/nntp/camel-nntp-folder.c (nntp_folder_get_message): oops, poke the right uid to get the article number. (nntp_folder_cache_message): & here too. Somehow fixes #58700, but i don't know why. 2004-05-19 Not Zed * camel-folder.c (transfer_message_to): copy the messageinfo rather than reference the source folder's one, we poke its flags variable if the message is deleted. And dont re-delete the message if its already deleted. * providers/nntp/camel-nntp-folder.c (nntp_folder_get_message): use the article number instead of the messageid. Some servers are just broken. (nntp_folder_cache_message): same. See #58655. * camel-smime-context.c (sm_verify_cmsg): import the certs as UsageEmailRecipient as well as signer, and also save the certs always. ** See #58641. * camel-vee-folder.c (vee_sync): don't rebuild auto-type vfolders. they should always be consistent and it saves a lot of unecessary work. * camel-store.c (store_sync): we don't want to sync any vfolders as part of the store sync call. its used for a different purpose in vFolders. oh well its a hack. 2004-05-18 Not Zed * providers/local/camel-maildir-store.c (fill_fi): do the same load of mailbox if we're in slow mode as we did for maildir. #58294. * camel-disco-folder.c (disco_expunge_uids): check for NULL implementation before calling it. (disco_sync): similar. Fixes crash #58278. 2004-05-17 Jeffrey Stedfast * camel-folder-search.c (search_match_threads): Fixed a string type-o. * camel-smime-context.c (sm_verify_cmsg): Fixed some spelling mistakes. * camel-gpg-context.c (gpg_decrypt): If the encrypted block was also signed, set the signature verification status on the Validity structure as well. 2004-05-17 Not Zed ** Bug #56050. * providers/imap/camel-imap-store.c (imap_get_trash) (imap_get_junk): similar to below. * providers/local/camel-local-store.c (local_get_trash) (local_get_junk): set state file on trash/junk to something we know about. 2004-05-17 Not Zed * providers/local/camel-mh-store.c (fill_fi): if we have no folder in-memory, load it if we're not doing it fast to get really up to date unread counts. #57616. 2004-05-14 Not Zed * camel-store.c (camel_store_create_folder): don't allow creation of Trash or Junk folders. (camel_store_rename_folder): similar for rename. #57250. 2004-05-13 Not Zed * camel-folder-summary.c (summary_build_content_info): set secure flag if we have a known security type. (flag_names[]): Added secure bit. (summary_build_content_info_message): set secure flag as appropriate. (summary_build_content_info_message): dont set attachments for simple types (non text), only base it on multipart/* stuff. (summary_build_content_info): same. * camel-folder-summary.h: added SECURE flag. Indicates signed or encrypted content. 2004-05-12 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): Ignore PERMANENTFLAGS if it gives us an empty set. Works around broken IMAP servers like the one in bug #58355. 2004-05-12 Not Zed * camel-folder-search.c (search_threads): changed to match_threads. (camel_folder_search_search): remove thread matching stuff from here. 2004-05-11 Jeffrey Stedfast * camel-smime-context.c (sm_signing_cmsmessage): Fixed a type-o. Fixes bug #58348. 2004-05-10 Jeffrey Stedfast * camel-mime-filter-gzip.[c,h]: New class for zipping/unzipping gzip streams. * camel-mime-filter-yenc.[c,h]: New class for encoding/decoding the crack known as YEncode. 2004-05-07 Not Zed * camel-folder-thread.c (thread_summary): properly set the parent nodes for the re-parented phantom-node children. Wasn't that fun to debug! * camel-folder-thread.h: make order and re bitfields, saves 4 bytes/node. 2004-05-06 Not Zed * camel-digest-folder.c (digest_search_by_expression) (digest_search_by_uids): * providers/nntp/camel-nntp-folder.c (nntp_folder_search_by_expression) (nntp_folder_search_by_uids): * providers/imap/camel-imap-folder.c (imap_search_by_expression) (imap_search_by_uids): * providers/local/camel-local-folder.c (local_search_by_expression) (local_search_by_uids): use camel_folder_search_search & some minor cleanups. * camel-folder-search.c (search_threads): keep track of the match threads option for this search. (camel_folder_search_match_expression): Removed, not used anymore. (camel_folder_search_search): new api entry point for searching, a bit easier to use and needed for thread matching. * camel-folder-search.c (camel_folder_search_search): new search api entry point, take a full summary and optionally a subset of uids to match against. (search_match_all): use the uids' passed in to only search a subset of uid's. * providers/imap/camel-imap-store.c (connect_to_server): set nodelay and keepalive on the socket. * camel-file-utils.c (camel_read): put a timeout on the select. Logic shuffle to match the ssl stuff. (camel_write): Similar. * camel-tcp-stream-ssl.c (stream_connect): remove timeout, use CONNECT_TIMEOUT directly. (stream_read): put a timeout on the poll. IO_TIMEOUT. And a little logic shuffle. (stream_write): similar. (CONNECT_TIMEOUT): make this 4 minutes === tcp-raw timeout. 2004-05-05 Not Zed * providers/imap/camel-imap-folder.c (get_message_simple): dont set X-Evolution-Source here anymore, set in caller. (imap_get_message): move the busted server get into the retry loop. only leave the simple cache test outside of it. 2004-05-04 Jeffrey Stedfast * camel-session.h: Get rid of #ifdef ENABLE_THREADS stuff, that was done away with a while back and is causing problems for 3rd parties trying to use camel unless they explicityly #define ENABLE_THREADS in their code. 2004-05-04 Christian Kellner * providers/groupwise/camel-gw-listener.h: Fixed typo. 2004-05-04 Not Zed ** See #57979. * camel-vee-folder.c (subfolder_renamed_update): when the subfolder gets renamed, remove all the old uid's and add the new ones (since the uid is based on the subordinate folder name). (subfolder_renamed): listen to renamed folder events. * providers/local/camel-mbox-store.c (get_folder_info): if we're getting a single folder with no children, make sure we fill the counts out. Fixes some rename strangeness. * camel-vee-folder.c (camel_vee_folder_add_folder): hook onto the folder renamed signal. (camel_vee_folder_finalise): unhook folder_renamed signal. (camel_vee_folder_remove_folder): same. 2004-05-03 Not Zed ** See bug #57881. * camel-vee-folder.c (camel_vee_folder_add_folder): use the folder's change log and frozen count, not our copy. (vee_thaw): dont maintain our frozen count. (vee_freeze): same. (folder_changed_change): if we get changed messages that dont match, make sure they're also propagated as a change too. * camel-private.h: remove the freeze_count from camelveefolderprivate. We already have that in the camel folder private. 2004-04-30 Priit Laes * providers/nntp/camel-nntp-summary.c: Fix typo. #53466. 2004-04-28 Jeffrey Stedfast * providers/imap/camel-imap-folder.c: Change the UID_SET_LIMIT to 768 (something <1000 octets as suggested by rfc2683). Fixes bug #57389. * camel-smime-context.c: Mark exception strings for translation and fixed a spelling mistake. 2004-04-27 Not Zed ** See bug #57659. * camel-vee-store.c (vee_get_folder_info): translate Unmatched in the folder display name. * camel-store.c (add_special_info): dont translate full_name or path, only do that for name. 2004-04-23 Sarfraaz Ahmed * camel-provider.h: Added HAS_LICENSE flag for allowing camel providers to display license file. 2004-04-22 Not Zed * providers/imap/camel-imap-folder.c (imap_expunge_uids_online) (imap_expunge_uids_resyncing): Send a flag list rather than \Deleted by itself. See #57381. 2004-04-21 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (imap4_sync_changes): Don't bother doing any work if perm_flags is 0. 2004-04-21 Sivaiah Nallagatla * providers/groupwise/Makefile.am: added CAMEL_LIBS to LIBADD * providers/groupwise/camel-gw-listener.c (get_addressbook_names_from_server): added new function to get addres book names from server. (add_addressbook_sources) (modify_addressbook_sources) : changed the implementation to use address book returned form above call while creating e-sources 2004-04-21 Not Zed * providers/nntp/camel-nntp-folder.c (camel_nntp_folder_new): set the meta data file on nntp folders. * camel-disco-folder.c (disco_refresh_info_online): implement dummy virtual method. * providers/nntp/camel-nntp-folder.c (nntp_folder_refresh_info_online): implement. Fixes #57280. 2004-04-19 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (untagged_fetch): Fixed to not expect ]'s as part of the BODY atom token. 2004-04-16 Jeffrey Stedfast * camel-vee-store.c (change_folder): (flags & 0) will never be true. Fixes bug #56982. 2004-04-15 Jeffrey Stedfast * providers/imap/camel-imap-utils.c: Changed imap_atom_specials[] to not treat ']' as an atom char (as per rfc3501). Fixes bug #50985. 2004-04-14 Jeffrey Stedfast * providers/imap/camel-imap-store.c (imap_check_folder_still_extant): Use %F instead of %S so that the folder name gets properly converted from using '/' path delimeters to whatever the native character the server uses. Should fix bug #56715. 2004-04-14 Not Zed * camel-disco-store.c (set_status): do offline mail syncing (only for open folders so far). If we fail doing syncing or store sync, don't abort. 2004-04-13 Jeffrey Stedfast * providers/imap/camel-imap-store.c (get_folder_online): Rearranged some error checking code to fix bug #56405. 2004-04-13 Not Zed * camel-folder.c (folder_getv): implement the new counts, and get them all atomically so they're only calculated once and can return consistent results. * camel-folder.h: Added CAMEL_FOLDER_DELETED, CAMEL_FOLDER_JUNKED, and CAMEL_FOLDER_VISIBLE args, to support client display of various values. 2004-04-13 Jeffrey Stedfast * camel-folder.h (camel_folder_delete_message): Fix NotZed's fix to not mark messages as unseen. Fixes bug #56879. 2004-04-13 Sivaiah Nallagatla * providers/groupwise/camel-gw-listener.c (add_addressbook_sources): change the auth type string from Password to plain/password 2004-04-12 Jeffrey Stedfast Fix for bug #56878. * camel-gpg-context.c (gpg_verify): Don't rely on the exit code of gpg, we already save enough state to decide if the sig is valid without it. Modified to only set BAD if gpg->validsig and gpg->nopubkey are both FALSE. If we get a NO_PUBKEY status message, then it simply means that the the sender could not be verified. (gpg_ctx_parse_status): Listen for NO_PUBKEY status messages. 2004-04-11 Sivaiah Nallagatla * providers/groupwise/camel-gw-listener.c (add_addressbook_sources) (modify_addressbook_sources) : add bookname to uri 2004-04-09 Jeffrey Stedfast * providers/imap/camel-imap-provider.c: Fix capitalisation of the "mailcheck" section title and move it to the top (so it matches with the UI). * providers/imap/camel-imap-folder.c (imap_get_message): Fetch the entire message in one fell swoop even if the message size is larger than 5k *if* the message is a single part. Fixes bug #56686. 2004-04-08 Jeffrey Stedfast * providers/imap/camel-imap-store.c (get_folder_online): Changed (!flags & _CREATE) to (!(flags & _CREATE)) (get_folder_online): Do what create_folder() does and if the parent folder has \NoInferiors set but contains no messages, delete the parent folder and recreate it before creating the child folder. Fixes bug #56651. 2004-04-08 Not Zed * camel-folder.h (camel_folder_delete_message): always set the seen flag when we delete a message. I demand that this may, or may not, fix #56549. * camel-folder.c: include camel-debug.h (camel_folder_set_message_flags): fixed the doco slightly (well reversed the flag and set explanation) and give an example. * providers/local/camel-mbox-folder.c (mbox_set_message_user_flag): message changed to folder_changed. (mbox_set_message_user_tag): ditto. * camel-vee-folder.c (camel_vee_folder_finalise): dont hook onto message_changed. (camel_vee_folder_add_folder): or unhook. (camel_vee_folder_remove_folder): " (message_changed): or proxy. * camel-folder.c (camel_folder_class_init): removed the message_changed event - its redundant, and covered by folder_changed, and just makes life difficult for everything using message_changed/folder_changed (and nothing uses it anyway). (message_changed): removed. (set_message_user_flag): emit a folder_changed event instead of message changed. (set_message_user_tag): " (set_message_flags): " * camel-object.h: revered the CAMEL_OBJECT_TYPE change, it should be a global variable access. 2004-04-06 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c: Added mutex locking where appropriate. * providers/imap4/camel-imap4-store.c: Added mutex locking where appropriate. * camel-folder-summary.h (CAMEL_FOLDER_SUMMARY_TYPE): Defined. * camel-object.h (CAMEL_OBJECT_TYPE): Fixed. * providers/imap4/camel-imap4-folder.c (camel_imap4_folder_utf7_name): Implemented. * providers/imap4/camel-imap4-store.c (imap4_build_folder_info): Use camel_folder_info_build() to build the folder-info tree. 2004-04-06 Not Zed * camel-debug.c: #if 0 out the sys/debugreg stuff. ** See bug #56110. * providers/imap/camel-imap-folder.c (get_content): more debug! (get_content): if we have no content-type header set on a sub-part of a multipart/digest, then we need to set it to message/rfc822 as in the multipart/digest rfc (2046 or so?). * camel-folder.c (camel_folder_get_message): output this stuff as folder debug. * providers/imap/camel-imap-folder.c (imap_get_message): add some imap:folder debug. (get_content): get xx.TEXT rather than xx if we're from a message parent part. ** See bug #56464. * camel-folder.c (camel_folder_transfer_messages_to): do not lock the source here. (transfer_message_to): call the main entry point for get message and append message. ** See bug #56050. * camel-vee-store.c (vee_delete_folder): delete the state file if it exists. * camel-object.c (camel_object_state_write): create the parent dir if we need to. Also spit a warning if we fail in the end. * camel-vee-folder.c (camel_vee_folder_new): set the persistent state file location. (vee_sync): write the state file when we sync. 2004-04-05 Not Zed * providers/imap/camel-imap-store.c (get_folders): dont add folders to folders_out here, only in get_folders_add_folders. * camel-store.c (camel_folder_info_build): simplify 'list append' since we have next pointer at the head of the struct. * providers/imap/camel-imap-store.c (create_folder): fixed "containes" spelling count. (get_folder_online): " (get_folders_add_folders): duh, add the folder info to the folders_out array. 2004-04-02 Jeffrey Stedfast * camel-folder.c (camel_folder_get_deleted_message_count): New function to get the deleted message count (used for Outbox count which is total minus deleted). 2004-04-02 Not Zed * camel-exception.c (w): turn this on, this should always be on, it points to real bugs in the code. * camel-folder-summary.c (summary_build_content_info): dont set attachments if its a signature block. (summary_build_content_info_message): same. 2004-04-01 Jeffrey Stedfast * camel-exception.c (w): Wrap annoying exeption warnings with w(). 2004-03-30 Jeffrey Stedfast * providers/imap4/camel-imap4-store.c (imap4_get_folder_info): Partially implemented. (imap4_get_folder): Implemented. 2004-03-30 Jeffrey Stedfast * camel-mime-utils.c (header_decode_param): Use header_decode_text() rather than rfc2047_decode_word() to decode the brokenly rfc2047 encoded param value string in case it is actually made up of multiple rfc2047 encoded words. Thanks to Andrei Nigmatulin for the fix. 2004-03-30 Not Zed * camel-store.c (camel_store_get_folder_info): added some debug to dump the whole folderinfo tree if store:folder_info is set. * providers/imapp/camel-imapp-driver.c: #if 0 out some code, to kill warnings. * camel-url-scanner.c: include ctype.h for isspace (wonder if it should use utf8 funcs?). 2004-03-29 Not Zed ** See #56146. * providers/imap/camel-imap-store.c (get_folders): check the top-level folders list for duplicates as well. (get_folders_add_folders): split out the folder return merging code from get_folders. Absolute mess of crap to deal with more busted servers. * camel-debug.c (camel_debug_start, camel_debug_end): some helpers to wrap debug output for atomicicity. 2004-03-29 Jeffrey Stedfast * providers/imap4/camel-imap4-folder.c (camel_imap4_folder_new): Implemented. * providers/imap4/camel-imap4-engine.c (engine_parse_namespace): If the namespace begins with "INBOX", canonicalise the INBOX portion (ie, make it all caps). * providers/imap4/camel-imap4-store.c (imap4_noop): Implemented. 2004-03-29 Jeffrey Stedfast * providers/imap/camel-imap-store.c (get_folder_status): Updated to actually parse the STATUS response into a list of item/value pairs. (create_folder): Updated for above change. (get_folder_counts): Only call get_folder_status() once (we can get both values with a single call now). Fixes a FIXME and might also fix bug #55784. * providers/local/camel-mbox-store.c (get_folder_info): Removed debugging printfs. * providers/local/camel-local-folder.c (camel_local_folder_construct): Since the folder was *just* created, it shouldn't have any subfolders so set the CAMEL_FOLDER_NOCHILDREN flag (altho, ideally, we wouldn't be guessing these flags at all - rather we'd call get_folder_info() or some such). Fixes bug #56010. 2004-03-29 Radek Doulik * camel-mime-filter-tohtml.c (html_convert): close pre tag in case we just flush. I am not sure if it's still worth to check for inlen == 0 and handle it specially, but didn't want to make too big changes. Fixes #55817 2004-03-29 Not Zed * providers/imap/camel-imap-store.c (get_folder_counts): use object_bag_peek instead of _get, since we dont want to clash/wait for reservations. More #56045 related fixes. (get_folder_counts): revert the lookup/hashtable stuff for the folder, and use object_bag_peek. * camel-object.c (camel_object_bag_peek): new method to get an object bag entry without worrying about if its reserved or not. * camel-gpg-context.c (gpg_verify): get the content-type off of the multipart-signed, not its container. This seems wrong ... but might fix #56084. * providers/imap/camel-imap-store.c (get_folder_counts): remove locking here, we're locked whne we enter. (fill_fi): call refresh_info unlocked. More for #56045. 2004-03-28 Jeffrey Stedfast * providers/imap4/camel-imap-folder.[c,h]: New source files implementing the CamelFolder class for the new IMAP4 implementation. * providers/imap4/camel-imap-summary.[c,h]: New source files implementing the CamelFolderSummary class for the new IMAP4 implementation. 2004-03-26 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (imap_update_summary): Reverted imap.web.de fix. 2004-03-26 Not Zed * providers/local/camel-maildir-summary.c (flagbits[]): Added new maildir flags D for draft and commented P for forwarded. * providers/imap/camel-imap-store.c (get_folder_counts): Instead of looking up the folder in the object bag which will handle reservations and perhaps deadlock, just get the list of opened folders and use them if they're available. Should fix #56045. 2004-03-25 Jeffrey Stedfast * providers/imap4/camel-imap-engine.c (camel_imap_engine_parse_resp_code): No longer need to split ']' tokens from atom tokens due to a fixup in the ABNF grammar in rfc3501. * providers/imap4/camel-imap-specials.c: Changed ATOM_SPECIALS to include ']' (this is an addition in rfc3501). * providers/imap4/camel-imap-store.[c,h]: New Store class for IMAP. Implemnted a bunch of but still got a ways to go. 2004-03-25 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (imap_update_summary): If the server is imap.web.de, just ask for all the headers rather than "HEADER.FIELDS.NOT (RECEIVED)". Actually, maybe we should always just query for the entire header block? (imap_get_message): If the server is brain-damaged (that's a technical term), always fetch the message in whole, never bother to try and fetch partial messages (Courier-IMAP gives us the wrong BODY responses fairly often). * providers/imap/camel-imap-store.c (connect_to_server): Set store->braindamaged to TRUE if we find the string "Courier-IMAP" in the greeting. 2004-03-25 Jeffrey Stedfast Fix for bug #55018. * providers/imap/camel-imap-store.c (create_folder): Don't allow the user to create folders with #, %, * or the directory separator in the folder name (added the checks for %, * and #). (get_folder_online): Add a check to make sure the folder name is sane before sending a CREATE (ie. we want to allow getting of folders with discouraged characters in them if they exist, but we don't want to allow the user to create them). 2004-03-25 Martyn Russell * providers/smtp/camel-smtp-provider.c: Removed newline character from the provider description 2004-03-25 Sivaiah Nallagatla * providers/groupwise/camel-groupwise-provider.c: removed ldap seetings and added a settng for SOAP port * providers/groupwise/camel-gw-listner.h (add_calendar_tasks_sources, remove_calendar_tasks_sources), (modify_calendar_tasks_sources): read port number from url instead of hardcoding. Also removed code for adding e-sources for ldap address book and adding now e-sources for groupwise address book 2004-03-24 Jeffrey Stedfast * providers/imap4/camel-imap-engine.c (camel_imap_engine_literal): New convenience wrapper function. (engine_parse_status): Fixed to handle literal mailbox strings. * providers/imap4/camel-imap-command.c (camel_imap_command_newv): Changed how %L works - create the CamelIMAPLiteral for our caller instead of expecting them to create it for us. We can autodetect what type of object (stream vs data-wrapper) the literal is, so it's trivial to do. 2004-03-23 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (imap_get_message): Reworked the else bit to fix a bug where if we had the BODY structure, we simply wouldn't try fetching the actual message. 2004-03-23 Not Zed * camel-exception.c (camel_exception_setv): use camel debug to add some debug here. * camel.c (camel_init): call camel_debug_init(). * camel-debug.c (camel_debug_init, camel_debug): new util stuff for extended debug options. * providers/imap/camel-imap-folder.c (imap_get_message): if we're supposed to be online, check we are online before proceeding. Actually major restructure so we re-try the fetch a couple of times first before failing. i.e. silent reconnect. See #55381. * providers/imap/camel-imap-store.c (get_folder_info_online): connect lock around this. was getting a race with mem corruption otherwise. 2004-03-22 Not Zed * camel-mime-utils.c (camel_header_newsgroups_decode) (camel_header_newsgroups_free): decode newsgroups header into a list of newsgroups. ** See #55887. * providers/nntp/camel-nntp-summary.c (camel_nntp_summary_check): NOOP if we're offline. * providers/nntp/camel-nntp-store.c (nntp_connected): spit a warning if we try to do stuff whilst offline, rather than crash. 2004-03-19 Not Zed * camel-disco-store.c (disco_connect): ref the diary before replaying it. it could get unreffed during replay if there's an error and we disconnect. * camel-store.c (camel_store_get_folder): no longer use folder_lock, we already have adequate serialisation code here or below here. I hope. (camel_store_get_folder_info): same here. * providers/imap/camel-imap-store.h: remove async_thread thing. 2004-03-17 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (imap_get_message): Checking that mi->content->type != NULL is no longer good enough to tell if a ContentInfo is complete (ie. contains the parsed BODY response). We need to check that th ContentInfo has children if the part is a multipart of message/rfc822 part. Apparently Zucchi didn't test his camel-folder-summary changes using IMAP. 2004-03-17 Radek Doulik * camel-folder-summary.h: use 1<<30 for CAMEL_MESSAGE_JUNK_LEARN, 1<<17 was already used by imap provider and maybe others 2004-03-17 Not Zed * providers/local/camel-maildir-store.c: Reverted jeff's fix for #55018, since it wasn't tested, and doesn't work. Wrote an alternate implementation, and tested it at least once. Turns out it was a one line error, it still wasn't tested. Oh well, its rewritten now. * See bug #55618. * camel-disco-diary.c (camel_disco_diary_new): seek to the end of the file after we open it. c99 apparently says the file merely adds to the end of the file when you write, not that it is opened and positioned at the end of the file (linux's man pages are out of date). * camel-folder-summary.c (content_info_new): setup the content type as well, from the headers. * providers/imap/camel-imap-summary.c (camel_imap_summary_add_offline): copy size from the source info. 2004-03-15 Jeffrey Stedfast * camel-folder.c (folder_rename): Always use '/' to derive the basename of the folder. folder->full_name is always the UNIX-path evrsion of the folder name, no matter what the actual path delimeter is on the underlying store. * providers/imap/camel-imap-store.c (get_folder_online): If the initial SELECT fails and CREATE is specified, clear the exception before attempting to CREATE, this way we don't have an exception set even if the CREATE passes. Fixes bug #55607. 2004-03-16 Not Zed * camel-mime-message.c (camel_mime_message_init): rever previous patch. We don't want to set the default mime/type, it'll break stuff. 2004-03-15 Not Zed * providers/local/camel-mbox-store.c (delete_folder): same as below for path. * providers/local/camel-local-store.c (delete_folder): NULL out str before looking up the state file using it. Otherwise we double-free str. * camel-mime-parser.c (folder_scan_skip_line): we want to scan till in-end-1. If we've been called we're either at the end of data, or we know we have an end of line character within memory. Another case in Bug #53355. * providers/imap/camel-imap-folder.c (get_content, get_message): set the mime-type field on the content the same way as construct_from_stream does. Bug #55472. * camel-mime-message.c (camel_mime_message_dump): utility function to dump message content to stdout. (camel_mime_message_init): default mime type to message/rfc822. * camel.c (camel_init): change camel verbose debug to be an int, a bitmask of debug options. * camel-mime-utils.c (camel_header_location_decode): drop embedded whitespace characters, and don't do unquoting, etc. See rfc2557 4.4.2 and rfc2017 3.1. 2004-03-12 Jeffrey Stedfast * providers/smtp/camel-smtp-transport.c (smtp_set_exception): Now takes an argument to specify whether disconnecting when respbuf is NULL is safe (to prevent us from recursively disconnecting or disconnecting during a connect). 2004-03-12 Jeffrey Stedfast Fix for bug #53497. * providers/smtp/camel-smtp-transport.c (smtp_helo): Instead of unreffing the streams, call camel_service_disconnect(). (smtp_mail): Same. (smtp_rcpt): Same. (smtp_data): Same. (smtp_send_to): Ignore exceptions for smtp_rset(). Also, check that we are connected before we try to send (in Evolution's current usage scenario, this isn't a problem but in the future if we ever try to fire off several messages via the same smtp connection, it may be - especially if RSET failed during the previous send). 2004-03-12 Radek Doulik * camel-folder.c (folder_changed): clearn the learn bit only if set. 2004-03-12 Jeffrey Stedfast Fixes for bug #55018. * providers/local/camel-local-folder.c (camel_local_folder_construct): Use camel_url_to_string() here too, so we properly encode the fragment. * providers/local/camel-mbox-store.c (get_folder_info): Use CamelURL to properly encode the fi->uri. Pass the CamelURL into scan_dir() so that scan_dir() can re-use it (rather than having to malloc/parse/free for each file/dir) (scan_dir): Use camel_url_to_string(). * providers/local/camel-maildir-store.c (get_folder_info): Same as mbox. (scan_dir): Same as mbox. We also need to set the CAMEL_FOLDER_NOSELECT flag if appropriate. * providers/local/camel-mh-store.c (get_folder_info): Same as mbox and maildir. (folders_scan): Now takes a url argument which we pass off to folder_info_new(). (recursive_scan): Same. (folder_info_new): Use camel_url_to_string(). 2004-03-11 Radek Doulik * camel-folder.c (camel_folder_set_message_flags): watch for setting JUNK flag, if JUNK_LEARN is not set as well then reset JUNK_LEARN bit (folder_changed): look for junk changes in uid_changed's messages, if these changes request junk filter learning (CAMEL_MESSAGE_JUNK_LEARN bit set) then prepare junk and nonjunk uid arrays, clear CAMEL_MESSAGE_JUNK_LEARN bit so that we don't process it again (folder_changed): start filter thread if there's junk and/or nonjunk arrays (filter_filter): if junk/nonjunk arrays are non-NULL, call junk filter report to learn junk/non-junk messages (filter_free): free junk/nonjunk uids and arrays * camel-folder-summary.h: added CAMEL_MESSAGE_JUNK_LEARN to CamelMessageFlags, used when setting CAMEL_MESSAGE_JUNK flag to say that we request junk plugin to learn that message as junk/non-junk 2004-03-12 Jeffrey Stedfast * providers/smtp/camel-smtp-transport.c (connect_to_server): If errno is EINTR, set USER_CANCEL instead of SERVICE_UNAVAILABLE or whatever. (smtp_helo): Same. (smtp_auth): Same. (smtp_mail): Same. (smtp_rcpt): Same. (smtp_data): Same. (smtp_rset): Same. (smtp_quit): Same. (smtp_set_exception): Here too. (smtp_auth): If the AUTH response code is not 334, then use smtp_set_exception() to get the most accurate error report we can. 2004-03-11 Jeffrey Stedfast * camel-object.c (cobject_state_read): Sanity check that count is <1024 and also use g_try_malloc so that we can recover if malloc fails. 2004-03-11 Not Zed * providers/imap/camel-imap-store.c (no_such_folder): removed this. not sure what it was doing there, a 1 line funciton used once. (get_folder_online): pass exception to camel_imap_command. if we got a user cancel, pass it up. See #55388. (hash_folder_name, compare_folder_name): more g_ascii_strcasecmp stuff. 2004-03-11 Not Zed * camel-vee-store.c (vee_get_folder_info): we need to add the folderinfo always if we're recursive from top. Should fix #52965 and maybe the other vfolders not showing on startup bug. * providers/imap/camel-imap-store.c (get_one_folder_offline): (parse_list_response_as_folder_info): turn off NOINFERIORS always, translate to nochildren. (imap_store_refresh_folders): check we're updating an imap folder, we could also have trash folders in the store too. 2004-03-08 Not Zed * camel-vee-store.c (vee_get_folder_info): setup virtual/system flags as appropriate. (change_folder): setup flags properly. * providers/nntp/camel-nntp-store.c (nntp_store_get_subscribed_folder_info): mark all folders as system folders. * providers/local/camel-mh-store.c (fill_fi): add this to setup folderinfo. (folder_info_new): call fill_fi to fill unread/total. (recursive_scan, folders_scan): ahh yeah, so wtf was i thinking, store->flags != get_folder_info flags!!!! * providers/local/camel-maildir-store.c (camel_folder_info_new): remove unread count arg & setup total. (fill_fi): setup total field. (scan_dir): remove the code that checked the directory directly - use fill_fi instead. It will more accurately reflect what you get when you visit the folder. (camel_folder_info_new): mark "." as a system folder. (scan_dir): try to setup children/no children flags properly. * providers/local/camel-mbox-store.c (fill_fi): setup total field. (scan_dir): init total. (get_folder_info): " 2004-03-05 Not Zed * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): mark INBOX as a system folder. Can't be renamed/deleted. (fill_fi): setup total field. (get_folder_counts): ditto. * camel-store.c (add_special_info): set the system folder flag. * camel-store.h: time to fix up the camelfolderinfo mess. fix some member names, and add some type fields. Fixed all uses. 2004-03-04 Not Zed ** See bug #53355. * providers/imap/camel-imap-folder.c (get_content): if the parent isn't a message/rfc822 type, we don't want to get the section.TEXT for multipart/signed, we just want to get section. 2004-03-03 Rodrigo Moya * providers/groupwise/camel-gw-listener.c (add_calendar_tasks_sources): use "Calendar" and "Checklist" for the folder names, instead of "Default". (remove_calendar_tasks_sources): remove the correct folder. 2004-03-03 Not Zed * camel-operation.c (camel_operation_uncancel): attempt at uncancelling a cancelled operation. * camel-stream-filter.c (do_write, do_write): fun dun diddley un fun. Since we're writing a const buffer, we need to copy it first. See #54937. 2004-02-27 Not Zed ** See bug #54755. * camel-vtrash-folder.c (vtrash_append_message) (vtrash_transfer_messages_to): error/fail out if the user tries to copy messages to the trash. (vtrash_transfer_messages_to): use the destination bit not the source bit for moving messages to a vtrash folder. * camel-gpg-context.c (gpg_ctx_parse_status): ignore NODATA response, otherwise we abort in a meaningless way. See #52939. * providers/imap/camel-imap-utils.c: use g_ascii_str[n]casecmp everywhere. * providers/imap/camel-imap-utils.c (imap_body_decode): fix the sense of the nil check for the subtype of a mutlipart. See #53355. 2004-02-26 Not Zed * camel-session.c (camel_session_check_junk_for_imap) (camel_session_set_check_junk_for_imap): removed. * providers/imap/camel-imap-provider.c: Add filter_junk and filter_junk_inbox options to the receive option page. * providers/imap/camel-imap-store.c (imap_setv, imap_getv): handle FILTER_JUNK and FILTER_JUNK_INBOX parameters. (imap_setv): conver to switch rather than if statement. (construct): handle url args for filter_junk and filter_junk_inbox. * providers/imap/camel-imap-folder.c (camel_imap_folder_new): Set the folder's flags based on the stores junk settings. (imap_update_summary): remove the test for session_check_junk_for_imap, its handled per-store now. * camel-folder.c (folder_changed): only check for FILTER_RECENT or FILTER_JUNK to see if we need to do filtering. * camel-folder.h (CAMEL_FOLDER_FILTER_JUNK): renamed from CAMEL_FOLDER_SUPRESS_JUNK_TEST (and obviously inverted logic). 2004-03-02 Jeffrey Stedfast * providers/local/camel-mbox-summary.c (mbox_summary_encode_x_evolution): Overrides the parent method. We don't want to encode user flags/tags or the size of the header will change and force a complete rewrite of the mbox file. 2004-02-25 Jeffrey Stedfast * camel-filter-driver.c (camel_filter_driver_filter_folder): Free the uids *after* reporting Complete, otherwise we get an FMR if our caller didn't pass in the uids. * camel-sasl-gssapi.c (gssapi_challenge): #ifdef out another gss_release_buffer() call as this function causes memory corruption if using Heimdal's implementation of Kerberos5. Yay Heimdal. 2004-02-25 Radek Doulik * camel-folder.c (get_unread_message_count): do not avoid junk mails in unread count 2004-02-25 Not Zed * providers/imap/camel-imap-summary.h (CAMEL_IMAP_MESSAGE_RECENT): moved the RECNET flag into the folder area (bit 17-30). 2004-02-24 Not Zed ** See bug #53876. * providers/imap/camel-imap-command.c (camel_imap_command): ref the folder before unreffing store->current_folder, incase they're the same. Do a select anyway. * providers/imap/camel-imap-folder.c (imap_refresh_info): keep the connect_lock for longer, imap_rescan for one assumes its locked. Fixes a race selecting the folder for refresh. 2004-02-24 Jeffrey Stedfast * camel-stream-process.c: #include , we don't need limits.h * camel-store.c (camel_store_folder_uri_equal): New function to do what camel_store_uri_cmp() was supposed to do. 2004-02-23 Rodney Dawes * providers/local/camel-mbox-store.c (scan_dir): If our folder has a subdir, but no actual children, then we need to unset the flag for CAMEL_FOLDER_CHILDREN Fixes #54470 2004-02-23 Jeffrey Stedfast * camel-store.c (camel_store_uri_cmp): Removed. Useless/broken function. 2004-02-20 Jeffrey Stedfast * camel-provider.c (camel_provider_list): Init list to NULL to prevent the crash in bug #54574. 2004-02-19 Chris Toshok * camel-smime-context.c: wrap this file with #ifdef ENABLE_SMIME instead of #ifdef HAVE_NSS. 2004-02-19 Jeffrey Stedfast * providers/local/camel-mbox-store.c (get_folder): Add sanity checking to the folder name if we are going to create it, just like we do in create_folder(). 2004-02-19 Not Zed * providers/*/camel-*-provider.c (camel_provider_module_init): Fixes for api changes. * camel-provider.c (camel_provider_load): no longer take session argument. the providers are global resources. (camel_provider_init): dont return anything anymore. (error?) call from camel_init now. Use a recursive lock too. * camel-session.c (camel_session_register_provider) (camel_session_list_providers, camel_session_get_provider): Moved to camel-provider, camel_provider_register/list/get. (vee_provider): moved to camel-provider.c 2004-02-18 Jeffrey Stedfast * camel-gpg-context.c (gpg_verify): Use the trust to decide the validity signature status. (Better way of solving yesterday's problem) * camel-cipher-context.h: Revert change from yesterday. 2004-02-17 Jeffrey Stedfast * camel-gpg-context.c (gpg_verify): Set the trust. * camel-cipher-context.h: Add a trust metric to signatures. 2004-02-17 Not Zed * camel-vee-store.c (vee_sync): implment, make sync a noop on vee stores. Speeds up exit, so we don't try and sync and re-sync folders multiple times. ** See bug #53861. * providers/nntp/camel-nntp-summary.c (add_range_xover): Fix from Edd Dumbill to avoid aborting on irrelevently truncated lines. 2004-02-16 Not Zed ** See bug #51045. * providers/imap/camel-imap-store.c (fill_fi): similar to mbox version. (get_folder_counts): use fill_fi to try and get folder counts if we're not doing the hard slog. (get_one_folder_offline): use fill_fi to try to get folder counts from open folders or summaries. * providers/local/camel-maildir-store.c (fill_fi): similar to mbox version. (scan_dir): use fill_fi to get the unread count now. * providers/local/camel-mbox-store.c (fill_fi): helper to lookup unread count either from active folder or from summary file, if it's available. (scan_dir, get_folder_info): use helper above to get folder info. * devel-docs/camel-folder-summary.txt: New document describing the format/conventions in the CamelFolderSummary file. * providers/nntp/camel-nntp-summary.c (summary_header_load/save): * providers/imapp/camel-imapp-summary.c (summary_header_load/save): * providers/imap/camel-imap-summary.c (summary_header_load/save): Handle versions, per-class version number (1). * providers/local/camel-mbox-summary.c (summary_header_load/save): Handle versions properly, add a per-class version (1). Write out the folder size as a size_t rather than 32 bit int. * providers/local/camel-local-summary.c (summary_header_load/save): read/write the per-class version number (1). * camel-folder-summary.c (summary_header_load): do version checking differently, allow the version to be bumped without aborting the load. Added unread/deleted/junk counts to base header. (summary_header_save): Save out the new-format header. Version bumped to 13. * camel.c (camel_init): return 0 rather than spit a compiler warning. * camel-file-utils.c (camel_file_util_encode_*_t): macro-ise the type encoder/decoders. Also add size_t encoder/decoder. 2004-02-13 Jeffrey Stedfast * providers/imap/camel-imap-store.c (get_folder_online): Same. * providers/local/camel-mh-store.c (get_folder): Same as maildir changes. * providers/local/camel-maildir-store.c (get_folder): Make exceptions strings consistanmt with the mbox exception strings (and vise versa). Also handle the CAMEL_STORE_FOLDER_EXCL flag. * providers/local/camel-mbox-store.c (get_folder): Check CAMEL_STORE_FOLDER_EXCL flag. * providers/local/camel-local-store.c (get_folder): Simplified by using camel_mkdir instead of doing it manually. * camel-store.c (camel_store_get_folder): If the folder exists in the cache and the O_EXCL flag was passed, return NULL and set an exception. * camel-store.h: Added a new CAMEL_STORE_FOLDER_EXCL flag for use with get_folder(). 2004-02-12 Jeffrey Stedfast * camel-file-utils.c (camel_file_util_encode_string): Since decoding a string doesn't allow strings longer than 65536, truncate strings that are longer than 65536 here. Fixes bug #54319. 2004-02-09 Not Zed ** See bug #53978. * providers/local/camel-mbox-store.c: added ".lock" to the list of ignored extensions. (ignore_file): ignore anything ending in ~ too. ** See bug #51319. * providers/local/camel-local-folder.c (camel_local_folder_construct): re-enable indexing when folder is first opened. Also load defaults if no meta-data present. 2004-02-06 Jeffrey Stedfast * camel-provider.c (camel_provider_init): Use strrchr, not strchr. * camel-gpg-context.c (gpg_ctx_parse_status): Forget the need_id, not the userid. Fixes bug #53908. * camel-store.c (add_special_info): Set a CAMEL_FOLDER_VIRTUAL bit on the special folder info so our UI can know if it is virtual or not (meant for vTrash/vJunk). (camel_store_get_folder_info): Don't add vTrash/vJunk if the NO_VIRTUAL flag bit is set. Used by the subscriptions editor. 2004-02-06 Not Zed * camel-folder.c (camel_folder_set_message_flags): changed to return a boolean to indicate if the flags were actually changed or not. Fixed all implementors. 2004-02-05 Not Zed * providers/local/camel-mbox-store.c (rename_folder): rename .cmeta file too, and fix the recovery order. * providers/local/camel-local-store.c (rename_folder): rename the .cmeta file too. * providers/local/camel-local-folder.c (local_rename): fix this to use local_get_full_path stuff. * camel-store.c (camel_store_rename_folder): fix umr's comparing old and new names. 2004-02-05 Not Zed ** See bug #53553. * camel-provider.c (camel_provider_init): changed to return a hashtable of url protocols to CamelProviderModule structs, rather than simple strings. * camel-session.c (get_provider): if we load a provider module, mark it as loaded. (ensure_loaded): Check the module loaded flag before trying to load it. * providers/local/libcamellocal.urls: Remove spoold from the list, since it doesn't exist anymore. Actually fixes #53553, the rest is to robustify the code. 2004-02-05 Not Zed * camel-session.c (CS_CLASS): dont typecheck cast. * camel-store.c (camel_vjunk_folder_new): removed, use vtrash_new(junk). (setup_special): changed to get_special, with a type now, and dont add vtrash folders to the sources. (get_trash, get_junk): down to 1 liners, call get_special * camel-vtrash-folder.c (CF_CLASS): dont use cast typecheck macros here, makes debugging easier and removes redundant checks. (camel_vtrash_folder_init): dont set flags here. (camel_vtrash_folder_new): takes a new argument, type, for junk folders too, removed name arg (taken from type). (vtrash_transfer_messages_to): parameterise flag processing. 2004-02-04 Jeffrey Stedfast * providers/imap/camel-imap-store.c: Get rid of some unnecessary CAMEL_OBJECT() casts. * providers/imap/camel-imap-folder.c (camel_imap_folder_new): The folder_name argument passed to this function is ALWAYS in the UNIX path form (ie. using '/' as the dir sep) and so when getting the short_name, don't use imap_store->dir_sep as the dir sep, always use '/'. Fixes bug #53755 for the IMAP case. 2004-02-04 Jeffrey Stedfast Fix for bug #53755 (local folders case) * providers/local/camel-mbox-store.c (xrename): No longer takes an exception arg, we just set errno. Our caller can take care of setting an exception. (rename_folder): Don't pass an exception to xrename(), we always overwrote if an error occured it anyway. (rename_folder): Rename the .sbd as well. 2004-02-04 Not Zed * camel-store.c (camel_store_sync): added 'expunge' parameter, easier 'empty trash on exit' call. (store_sync): duh, actually pass expunge to folder_sync. 2004-02-04 Not Zed * providers/imap/camel-imap-store.c (parse_list_response_as_folder_info): if we can't add the folder to the summary {i.e. duplicate}, then ignore it. See #53836. * camel-store.c (camel_store_delete_folder): changed order around, first try to delete and only remove from the object bag if the delete worked. If vjunk/vtrash enabled, don't allow those to be deleted. (cs_delete_cached_folder): helper to delete the folder if its in the cache, remove it from trash/junk, etc. (camel_store_unsubscribe_folder): changed similarly to delete_folder. * camel-vee-store.c (vee_delete_folder): dont do any trash/junk processing anymore. 2004-02-03 Not Zed * camel-store.c: use the folders object bag to manage the trash folder and junk folders. Remove the init_trash and init_junk stuff, just use get_trash and get_junk to mean the same thing. Get rid of the hacked up vjunk and vtrash "uri" stuff too. * camel-object.c (camel_object_bag_add): null out the pair->func, otherwise we get an uninitalised memory read during unhook event. 2004-02-03 Jeffrey Stedfast * camel-vee-folder.c (vee_folder_build_folder): Use macro casts from int to pointer. * camel-folder.c (camel_folder_change_info_add_source): Same as below. (camel_folder_change_info_add_source_list): Same. * camel-folder-search.c (camel_folder_search_execute_expression): Use GINT_TO_POINTER() to cast 1 to a pointer for g_hash_table_insert(). * camel-vee-folder.c (vee_folder_remove_folder): 64bit fixes. (folder_added_uid): Same. (vee_folder_build_folder): Here too. (folder_changed_add_uid): And here. (folder_changed_remove_uid): Same. 2004-02-03 Jeffrey Stedfast * tests/misc/url-scan.c: New test suite for url scanning. * camel-url-scanner.c: Added single/double quotes to url_braces[] in case the user is quoting the url. (camel_url_web_end): Add "-;:" to list of punctuation to strip off the end of urls. Also fixed to handle user@domain's (camel_url_addrspec_start): Strip open brace characters from the beginning of the addr. (camel_url_web_start): Make sure "www" wasn't part of something not a url (like "Ewww.Gross") by check that pos[-1] is either an open brace or whitespace. (camel_url_addrspec_end): Don't allow toplevel domain addr-specs (if we encounter something that looks like it is a toplevel domain addr, it is more likely to be bogus than correct). 2004-02-02 Jeffrey Stedfast Fixes for bug #53091. * providers/imap/camel-imap-store.c (create_folder): Set the new folder's fi->flags to CAMEL_FOLDER_NOCHILDREN since we know it doesn't have any (we just created it!). (subscribe_folder): Same. * camel-store.c (folder_info_clone_rec): Copy the flags too. Fixes the local folder case of bug #53091. 2004-01-31 Jeffrey Stedfast * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): Same as below. * providers/smtp/camel-smtp-transport.c (smtp_connect): Instead of using the form %s@%s in the password prompt, use %s on host %s, hopefully this will be less confusing to users who have usernames of the form user@vhost. 2004-01-30 Rodney Dawes * providers/groupwise/camel-groupwise-provider.c: Fix spelling error in provider description Fixes #53572 2004-01-30 Jeffrey Stedfast * camel.c (camel_init): Protect against multiple camel_init() calls. Remember if we've already been called. 2004-01-30 Not Zed * camel-store.c (camel_store_rename_folder): copy the old_name that comes in, since it might be the actual folder_name, which will go away during processing. Related to #53123. 2004-01-29 Not Zed ** See bug #53269. * providers/nntp/camel-nntp-store.c (nntp_store_get_cached_folder_info): don't dereference last before checking if its null. * camel-object.c (camel_object_bag_rekey): added a doc comment. ** See bug #53520. * camel-session.c (get_service): free the url once done, it now gets copied by the service. * camel-service.c (construct): copy the url that comes in, don't just '0Wn34z' it. clena up exception handling too. 2004-01-29 Not Zed * camel-object.c (cobject_state_write): output scan->name and scan->value for writing metadata, rather than meta->name/value which just duplicates the last entry, related to #53195. * camel-url.c (camel_url_free): zero out passwd/user/host before freeing them. 2004-01-28 Jeffrey Stedfast * providers/local/camel-mbox-store.c (rename_folder): Make sure the new dir path exists before trying to rename files to avoid ENOENT errors. Also save errno when we encounter errors so that we can report the true error later rather than an error we may get while reverting changes. Also, it is OK if the ibex files don't exist, so check for that errno case. 2004-01-28 Sivaih Nallagatla * providers/groupwise/camel-gw-listener.c (account_changed) (modify_esources) : add port and /soap parts to relative uri while updating it when account changed 2004-01-27 Radek Doulik * providers/imap/camel-imap-folder.c (camel_imap_folder_new): set CAMEL_FOLDER_SUPRESS_JUNK_TEST flag (imap_update_summary): update CAMEL_FOLDER_SUPRESS_JUNK_TEST flag * camel-folder.c (folder_changed): use CAMEL_FOLDER_SUPRESS_JUNK_TEST flag instead of check_junk_for_imap * camel-folder.h (CAMEL_FOLDER_SUPRESS_JUNK_TEST): added new flag 2004-01-27 Not Zed ** See bug #53373. * camel-store.c (camel_store_rename_folder): use object_bag_rekey to rename the object. object_bag use was broken. * camel-object.c (camel_object_bag_rekey): new api to atomically re-key 2004-01-23 Radek Doulik * camel-folder.c (folder_changed): use check_junk_for_imap flag * camel-session.c: add check_junk_for_imap flag 2004-01-23 Not Zed * providers/imap/camel-imap-folder.c (IMAP_SMALL_BODY_SIZE): removed the meaningless fixme, a butt-retrieved-number is as good as any in this case. (imap_get_message): revert peterw's change of 2002-07-15, instead of checking for online mode here, let get_message do it when you retrieve the parts. This lets a multi-fetch (i.e. large) message work more betterer in offline mode. 2004-01-22 Jeremy Katz * camel-mime-part.c: Fix prototype to be consistent. 2004-01-21 Not Zed ** See bug #52996. * camel-data-cache.c (camel_data_cache_add): put a do-loop around the object_bag_reserve stuff, otherwise we can add/abort out of sync (i.e. when object_bag_reserve returned a pointer we mustn't call add/abort). * camel-object.c (camel_object_bag_*): Added some inline doco. 2004-01-20 Not Zed ** See bug #52817. * camel-session.c (camel_session_get_password): merged reprompt and secret into a flags field, and add more options. Fixed all callers. ** See bug #52899. * camel-gpg-context.c (gpg_ctx_parse_status): use need_id as the password key, not userid. 2004-01-19 Siviaah Nallagatla * providers/groupwise/camel-gw-listener.c (remove_calender_tasks_sources, modify_calender_tasks_sources): add port and "/soap" to source uri 2004-01-19 Not Zed * camel-vee-store.c (change_folder): use a CamelURL to properly encode the url we generate. (vee_get_folder_info): ditto. (vee_get_folder_info): removed unused variable/warning. * camel-session.c (vee_provider): Update the provider flags for URL_FRAGMENT_IS_PATH. * providers/imapp/camel-imapp-utils.c (imap_parse_addfress_list): namespaces fixes for HEADER_ADDRESS* 2004-01-17 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (imap_sync_online): Limit the flags we set (or unset) to the folder's permanent flags. 2004-01-16 Jeffrey Stedfast * camel-mime-message.c (camel_mime_message_build_mbox_from): Same. * camel-internet-address.c (internet_decode): Same. * camel-mime-utils.[c,h]: Namespaced camel_header_address_t enums s/HEADER_ADDRESS_/CAMEL_HEADER_ADDRESS_/g 2004-01-16 Sivaiah Nallagatla * providers/groupwise/camel-groupwise-provider.c: added some conf entries for LDAP address setup * providers/groupwise/camel-gw-listener.c: added add_ldap_addressbook_source, modify_ldap_addressbook_source, remove_ldap_addressbook_source functions for setting up LDAP address book. Also setting "username" property on cal/tasks ESources 2004-01-15 Not Zed ** See bug #52881. * camel-object.c (camel_object_bag*): Support reserving different keys from the same thread. Oh the pain. * camel-vee-store.c (vee_get_folder_info): implement child flags properly. Changed to build tree itself rather than calling camel_folder_info_build. (vee_get_folder): if we're adding a folder with dummy parents, create and add the dummy parent folders too (as real folder objects). We are the only owner of the ref, so this sort of leaks the folder, but they're small. 2004-01-14 Rodrigo Moya * providers/groupwise/camel-groupwise-provider.c (camel_provider_module_init): missing renaming. 2004-01-14 Rodrigo Moya * providers/groupwise/camel-gw-listener.c (add_esource): set the "auth" property on the ESource's we create, to get authentication. 2004-01-14 Not Zed ** Patch from Timo Sirainen to honour read-only status for imap folders. * providers/imap/camel-imap-folder.c (camel_imap_folder_selected): check for read-only status response. (imap_sync_online): only call sync_offline if we're read-only. (imap_expunge_uids_resyncing): NOOP for read-only. 2004-01-14 Not Zed * providers/local/camel-local-folder.c (local_getv): remove debug printf. ** See bug #52835. * camel-smime-context.c (sm_get_passwd): removed debugging. If we get called multiple times in a row, then forget the old password and re-prompt - it was a bad password. Zero out password memory too. 2004-01-13 Rodrigo Moya * providers/groupwise/camel-groupwise-provider.c: missing renaming. 2004-01-13 Sivaiah Nallagatla * providers/groupwise/camel-gw-listener.[ch]: renamed groupwise-config-listener.[ch] to these file names and also changed code to use CamelURL insted of EUri * providers/groupwise/camel-groupwise-provider.c: use the renamed config listener apis * providers/groupwise/Makefile.am: changed the source file names 2004-01-12 Meilof Veeningen * providers/nntp/camel-nntp-folder.[ch]: now based on discofolder, cache_message and append_message implemented, only retrieve messages when we are subscribed, some stubs * providers/nntp/camel-nntp-provider.c: newsgroup name display settings, password authentication, fix for check_equal where the protocols wouldn't be checked * providers/nntp/camel-nntp-store.[ch]: base on discostore with online/offline support, subscriptions, downloading changed parts of the newsgroup list, some stubs, authentication, automatic reconnect * providers/nntp/camel-nntp-store-summary.[ch]: NNTP store summary based on IMAP code * providers/nntp/camel-nntp-summary.c: save summary after xover * providers/nntp/camel-nntp-grouplist.h: added CamelNNTPGroupList structs * providers/nntp/Makefile.am: added store summary 2004-01-12 Not Zed ** See bug 52725. * providers/imap/camel-imap-folder.c (get_content): pass in transfer encoding when setting up wrapper part. * providers/imap/camel-imap-wrapper.c (camel_imap_wrapper_new): Added an encoding type parameter, set on data wrapper. 2004-01-10 Jeffrey Stedfast * camel-mime-utils.c (rfc2047_encode_word): if e_iconv() returns -1, check that errno != EINVAL - if errno *is* EINVAL, it just means that our convlen wasn't long enough to include the whole sequence. This is fine, we'll just start where we left off next loop thru. Fixes bug #52593 (the buffer was duplicated because state wasn't flushed). (camel_header_encode_string): Fixed a type-o in loop where encoding=0, don't g_string_append_len starting at 'word' inptr-start bytes long - 'word' could be NULL and/or inptr-start could be longer than inptr-word. 2004-01-09 Rodney Dawes * providers/groupwise/Makefile.am (EXTRA_DIST): libcamelgroupwise.urls instead of libcamelimap.urls 2004-01-09 Sivaiah Nallagatla * providers/groupwise/groupwise-config-listener.c (is_groupwise_account): added null check for source url to take care of accounts with Server Types as "None" 2004-01-09 Not Zed * providers/imap/camel-imap-store.c (imap_forget_folder): fix removal of journal file, and remove the cmeta state file too. * providers/imap/camel-imap-folder.c (imap_getv): count up so we call parent class if we missed any, rather than only if we didn't miss any. (imap_rename): rename the object state file. (camel_imap_folder_new): set the object state file for persistent properties. * camel-disco-folder.c (disco_getv): support (PERSISTENT_)PROPERTIES & OFFLINE_SYNC. (disco_setv): implement OFFLINE_SYNC. (camel_disco_folder_get_type): setup disco properties list. (cdf_folder_changed): honour the offline_sync setting on the current folder. (disco_sync): save object state. (disco_setv): save object state if it changed. * camel-data-wrapper.c (camel_data_wrapper_set_mime_type_field): move assertions here. (set_mime_type_field): change order slightly to properly handle setting the same object. removed assertions from internal method. 2004-01-08 Sivaiah Nallagatla * providers/Makefile.am : add groupwise to SUBDIRS * providers/groupwise/camel-groupwise-provider.c: new camel provider for groupwise * providers/groupwise/groupwise-config-listener.[ch] : new class to add e-sources for groupwise calender and tasks * providers/groupwise/Makefile.am : added new files to Makefile.am 2004-01-05 JP Rosevear * camel-utf8.c: include sys/types.h for freebsd 2004-01-05 Rodrigo Moya * providers/groupwise/camel-groupwise-provider.c (groupwise_url_hash, groupwise_url_equal): renamed from imap_*. 2004-01-05 Not Zed * camel-tcp-stream-raw.c (socket_connect): check the right return of the socket call, dont set fd to the value of the -1 check! 2003-12-27 Jeffrey Stedfast * camel-tcp-stream-raw.c (socket_connect): Save errno and check the return of the socket() call. 2003-12-24 Rodrigo Moya * providers/groupwise/camel-groupwise-provider.c (camel_provider_module_init): g_module_open the other providers we depend on to avoid load ordering problems. 2003-12-22 Rodrigo Moya * providers/groupwise/camel-groupwise-provider.c: removed useless configuration options, and added other options from the IMAP provider. (camel_provider_module_init): register a SMTP transport object also, and removed SASL registration, since we don't support it. 2003-12-19 Sivaiah Nallagatla * providers/groupwise/Makefile.am: * providers/groupwise/libcamelgroupwise.urls: * providers/groupwise/camel-groupwise-provider.c: added Camel provider for Groupwise accounts, based on the IMAP one. 2003-12-11 Jeffrey Stedfast * camel-gpg-context.c (gpg_ctx_parse_status): Don't immediately prompt for a passwd after receiving the NEED_PASSPHRASE status message. Instead, parse the userid that gpg needs a passwd for and store it on our context. Wait for a GET_HIDDEN status message before prompting, this way if the user has their gpg configured to use gpg-agent, the user won't get 2 passwd prompts. (gpg_sign): Fixed to not free a gpg context that we have not allocated (could happen in a fail case). 2003-12-10 Jeffrey Stedfast * providers/local/camel-mbox-store.c (get_folder_info): Use CAMEL_FODLER_NOCHILDREN rather than NOINFERIORS because NOINFERIORS means the folder cannot contain subfolders. However, our mbox structure always allows subfolders. (scan_dir): Same. * camel-store.h: Add a CAMEL_FOLDER_NOCHILDREN flag. * providers/imap/camel-imap-store.c (get_folders): Same as below. * providers/imap/camel-imap-utils.c (imap_parse_list_response): s/CAMEL_IMAP_FOLDER_NOCHILDREN/CAMEL_FOLDER_NOCHILDREN/ 2003-12-10 Not Zed * camel-smime-context.c (sm_verify): also check application/pkcs7-signature (bloody applemail). (camel_smime_context_describe_part): as above, and fix the logic. dont think its used anyway. Bug #51750. 2003-12-10 Not Zed * Makefile.am (libcamel_la_SOURCES): put the camel-smime-context.[ch] back in here, remove it from EXTRA_DIST, and make it compile optinally the same way camel-tcp-stream-ssl.c does (#ifdef ...). * camel-smime-context.c (sm_verify_cmsg): add signer info to certvalidity. * camel-cipher-context.c (CamelCipherValidity): Added certinfo to validity for signing and encrypting, so we can find the keys later for a gui. (camel_cipher_validity_add_certinfo): add signer or encrypter info to the validity. (camel_cipher_validity_envelope): add sign/encrypt keys. 2003-12-10 Not Zed * camel-stream-process.c (do_exec_command): remove dthe clearenv stuff, not sure why its there. s/setenv/putenv/ for portability. See Bug #51767. 2003-12-08 Jeffrey Stedfast * camel-mime-utils.c (header_decode_rfc2184_param): Revert the s/is_/camel_mime_is_/ changes or we get really long variable names. (header_decode_param): Same. (header_decode_param_list): Here too. 2003-12-09 Not Zed ** See bug #51899. * providers/imap/camel-imap-store.c (get_folders): add the first fi to the info's hash, so we dont duplicate it if we come across it again (which we generally will). 2003-12-09 Not Zed * providers/smtp/camel-smtp-transport.c (smtp_send_to): encode the address before sending it out, rather than using the raw/utf8 version. * camel-internet-address.c (camel_internet_address_encode_address): check for quoting the local part of the address before outputting it. (cia_encode_addrspec): quote local part if need be. (camel_internet_address_encode_address): make folding optional based on whether inlen is null or not. * camel-mime-utils.[ch]: rename is_* to camel_mime_is_* and export the type functions. 2003-12-08 Not Zed * providers/imap/camel-imap-utils.c (imap_atom_specials): add } to the atom specials list. This isn't correct, but some busted servers expect it. Bug #50728. 2003-12-08 Jeffrey Stedfast Fixes bug #51881 * providers/local/camel-mbox-store.c (delete_folder): Same. * providers/local/camel-local-store.c (delete_folder): Unlink the cmeta file too. 2003-12-06 JP Rosevear * */Makefile.am: Remove hard coded disable deprecated flags 2003-12-05 Jeffrey Stedfast * camel-gpg-context.c (gpg_ctx_op_start): Properly set the O_NONBLOCK flag along with any previously set flags. * camel-filter-search.c (run_command): Don't set O_NONBLOCK on the pipe (1. we don't need to, and 2. we should have been setting O_NONBLOCK|prev_flags but we weren't, and so the pipe got O_RDONLY|O_NONBLOCK even tho we wanted to write to it). * camel-filter-driver.c (pipe_to_system): Same. 2003-12-04 Radek Doulik * camel-folder.c (folder_changed): check recent messages for junk mail * camel-session.c (camel_session_check_junk): new wrapper method for check_junk flag (camel_session_set_check_junk): ditto * camel-session.h (struct _CamelSession): added check_junk flag (if to check incoming mail for junk messages) 2003-12-03 Jeffrey Stedfast * camel-store.c (add_special_info): Free fi->path if we are gonna replace it with the vinfo path. * providers/local/camel-mbox-store.c (create_folder): Treat parent_name == NULL and parent_name == "" the same. * camel-store.c (camel_store_get_folder_info): Only add vTrash/vJunk info's if we've requested the toplevel folder tree, otherwise we get vTrash/vJunk folders in odd places in the folder tree. (add_special_info): Use the provider->url_flags to determine if the url uses the fragment or not for the path. 2003-12-02 Not Zed * providers/local/camel-local-folder.c (local_setv): remove some debug printfs. ** See bug #51576. * camel-url.c (camel_url_decode): robustify url decoding for bad input, addresses a crash. ** See bug #51478. * camel-gpg-context.c (gpg_ctx_free): handle the context being NULL, and exit silently. 2003-12-01 Jeffrey Stedfast * camel-process.c (camel_process_fork): Start at fd = 3. * camel-gpg-context.c (gpg_ctx_op_start): Same. * camel-filter-driver.c (pipe_to_system): Same. * camel-filter-search.c (run_command): Start at fd = 3. 2003-12-01 Jeffrey Stedfast * camel-stream-process.c (do_exec_command): Same. * camel-process.c (camel_process_fork): Same. * camel-filter-search.c (run_command): Same as below. * camel-filter-driver.c (pipe_to_system): Same as below. * camel-gpg-context.c (gpg_ctx_op_start): Use fcntl() to set FD_CLOEXEC on each fd rather than close()ing it. Apparently Linux's older pthread implementations use sockets and so this fouls threading up. GO LINUX! GO! 2003-12-01 Radek Doulik * camel-store.c (add_special_info): set SUBSCRIBED and NOINFERIORS flags to special folders, renamed method from add_vtrash_or_vjunk_info (camel_store_get_folder_info): call add_special_info directly 2003-11-28 Radek Doulik * camel-store.c (camel_store_get_folder_info): use old code from mail-ops to add vtrash/vjunk info 2003-11-28 David Woodhouse * providers/imap/camel-imap-provider.c: Enable GUI option for 'custom command' connection. * providers/imap/camel-imap-store.c: Don't g_free strings in .rodata. It's considered rude. 2003-11-28 Not Zed * camel-mime-utils.c (mail_list_magic[]): Added list-unsubscribe header match. 2003-11-26 Not Zed * camel-smime-context.c (sm_verify_cmsg): take a stream rather than a part for the content. (sm_verify): get the content directly as a stream. * camel-multipart-signed.c (camel_multipart_signed_get_content_stream): new api to get the content stream which will match the signed version. 2003-11-26 JP Rosevear * Makefile.am: make sure we always dist the smime stuff 2003-11-26 JP Rosevear * Makefile.am (libcamelinclude_HEADERS): conditionally compile s/mime support 2003-11-25 Not Zed * camel-cipher-context.c (camel_cipher_validity_envelope): change args to make it suit storing the validity in a tree. (camel_cipher_validity_init): init the list header. (camel_cipher_validity_clone): call validity_new so it gets init properly. (camel_cipher_validity_free): free any children nodes recursively. * camel-cipher-context.h (CamelCipherValidity): added next/prev and list header. 2003-11-18 Jeffrey Stedfast * camel-session.c (camel_session_finalise): Don't destroy the providers as it mans we can never ever have more than 1 session object. See bug #51119 for details. 2003-11-17 Jeffrey Stedfast * providers/local/camel-mbox-store.c (create_folder): Fixed an exception to give a more meaningful description. (get_folder_info): We always want to scan at least one level deep. 2003-11-14 Jeffrey Stedfast * providers/local/camel-mbox-store.c (get_folder_info): Handle "" as a request for a full dirscan as well (not just NULL). Makes it consistant with other stores. (scan_dir): Set the fi->flags appropriately. * providers/local/camel-mbox-folder.c (camel_mbox_folder_get_full_path): Removed temporary hack. 2003-11-14 Not Zed * camel-mime-parser.c (SCAN_BUF): oops, put the mempool stuff back, we don't use e-memory afterall. Damn plane hacking. 2003-11-13 Not Zed * camel-mime-parser.c: Remove mempool code, we use the stuff in e-util. (PRESERVE_HEADERS): new compile option, if on, we preserve headers and folding exactly rather than unfolding all input. THIS BREAKS EVERYTHING right now, so don't turn it on. * camel-gpg-context.c (gpg_decrypt): reset the input memstream before passing it to the gpg engine. * tests/smime/pgp-mime.c (main): redirect /dev/null to stdin so it doesn't hang waiting for input. (main): removed from build - this tests multipart/signed explictly, but now the details of this is handled directly by the cipher context. * tests/smime/pgp.c (main): fixes for api changes. (main): redirect /dev/null to stdin so it doesn't hang waiting for input. * tests/message/test1.c (main): update for api changes. * camel-smime-context.c (sm_verify): look at the content object's mime type, not the container's type. 2003-11-11 Not Zed * camel-cipher-context.c (camel_cipher_validity_set_valid): take into account the @valid argument and set validity properly. (camel_cipher_validity_clone): new method to copy validities. * camel-smime-context.c (sm_signing_cmsmessage): removed a todo. 2003-11-10 Not Zed * camel-smime-context.c (sm_verify_cmsg): split out the CMSMessage verification code so it can be used from enveloped or externally signed data. * camel-cipher-context.c (camel_cipher_verify): only take a mimepart, internally handle multiparts and the hash. 2003-11-07 Not Zed * camel-cipher-context.c: make ciphervalidity a public structure, added encrypt status. (camel_cipher_decrypt): changed to return a ciphervalidity. fixed implementations. (camel_cipher_validity_*): Fixed implementations to match new structure, some of this is now redundant. 2003-11-06 Not Zed * camel-smime-context.c (camel_smime_context_describe_part): implement. 2003-11-05 Jeffrey Stedfast * providers/smtp/camel-smtp-transport.c (connect_to_server): Don't bother trying to see if the server advertises EHLO. Simply always try EHLO and fall back to HELO if EHLO fails. Fixes bug #50535. 2003-11-05 Not Zed * camel-smime-context.c (camel_smime_context_describe_part): new (unfinished) api to peek inside smime parts to tell us whats in it. 2003-11-03 Not Zed * camel-gpg-context.c (gpg_encrypt): Make this output the full multipart/encrypted part, not just the encrypted content part. * camel-cipher-context.c (camel_cipher_sign): change to output full mime part, not just a stream. (camel_cipher_canonical_to_stream): utility function to canonicalise a mimepart to a stream. * camel-smime-context.c (sm_encode_cmsmessage): removed. (sm_sign): change interface to output a full mime-part, not just a part of a mime part in multipart/signed mode. 2003-11-04 Jeffrey Stedfast * camel-gpg-context.c (gpg_ctx_parse_status): We might need to convert the passwd from UTF-8 into the locale charset. Fixes bug #50485. 2003-10-31 Not Zed * camel-cms-context.[ch]: removed, now redundant. 2003-10-30 Not Zed * camel-smime-context.c (sm_get_passwd): implement something basic. * camel-cipher-context.h: Added a note about api inconsistencies. 2003-10-30 Not Zed * camel-multipart-encrypted.c (camel_multipart_encrypted_decrypt): fix for cipher_decrypt changes. * camel-gpg-context.c, camel-cipher-context.c: moved all the init code to the end to save having to keep forward declarations around. (camel_cipher_decrypt): changed to take mimepart input and return a mimepart. (gpg_decrypt): fix for changed args. 2003-10-29 Not Zed * camel-smime-context.[ch]: replaced entirely with a new implementation which inherits from camel-cipher-context, and add to build. * camel-multipart-encrypted.c (camel_multipart_encrypted_encrypt): fix for cipher_encrypt api changes. (camel_multipart_encrypted_decrypt): use g_ascii_strcasecmp. * camel-gpg-context.c (gpg_encrypt): Fix to handle input/output as parts not streams * camel-cipher-context.c (camel_cipher_encrypt): change to take mimeparts rather than streams as input/output. And remove the 'sign' argument, it is implied if userid is supplied. 2003-10-28 Not Zed * tests/smime/pgp.c (main): fix for ciphercontext api changes. * camel-multipart-signed.c (camel_multipart_signed_verify): pass in the part to cipher_verify directly. (camel_multipart_signed_sign): let the cipher context setup the part details. * camel-gpg-context.c (gpg_sign): put the signature stream into a mimepart, with appropriate headers/encoding. (swrite): write out a mimepart rather than a stream. (gpg_verify): handle changed args. * camel-cipher-context.c (camel_cipher_sign): write the signature to a mimepart rather than a simple stream. (camel_cipher_verify): take the signature as a mimepart not a stream. 2003-10-22 Not Zed * camel-utf8.c (camel_ucs2_utf8, camel_utf8_ucs2): helpers for ucs2 stuff. ucs2 is 16 bit truncated unicode. 2003-10-28 Jeffrey Stedfast * camel-mime-utils.c: We should check List-Post before List-Id (List-Post has to contain the mailing-list posting address, whereas List-Id does not.) WAlso moved X-Loop to after List-Id to make FreeBSD lusers happy. Fixes bug #32297. 2003-10-28 Jeffrey Stedfast * Fixes bug #35083 * providers/imap/camel-imap-store.c (connect_to_server): Same here. * providers/pop3/camel-pop3-store.c (connect_to_server): Same as the smtp changes. * providers/smtp/camel-smtp-transport.c (connect_to_server): If HAVE_SSL is undefined, don't default to raw connections if the option to connect via ssl is set. Instead set an exception and return fail. 2003-10-27 Frederic Crozat * camel-mime-utils.c: (camel_header_decode_date): better detection of broken date to give to broken_date_parser. 2003-10-24 Jeffrey Stedfast * camel-text-index.c (text_index_name_add_buffer): If a word is longer than CAMEL_TEXT_INDEX_MAX_WORDLEN, then ignore it. This fixes bug #50096. 2003-10-23 Jeffrey Stedfast * *.c: Removed unneeded CAMEL_OBJECT() casts. 2003-10-21 Not Zed * providers/local/camel-local-folder.c (local_getv, local_setv): use the right tag name for the index_body arg. (local_sync): write any persistent metadata - to make it persistent. (camel_local_folder_construct): turn off indexing, for now, it should be done in local_setv. * providers/local/camel-local-folder.h: change body_index to a bool type. * camel-object.c (cobject_state_read, cobject_state_write): handle bool types (cobject_state_write): make sure we free all arg types. * camel-arg.c (camel_argv_build): (camel_arggetv_build): handle bool type. * camel-arg.h: Added BOO (bool) type. 2003-10-15 Not Zed * camel-store.c (camel_folder_info_build): Fix so we output the tree in sorted depth-first order, rather in reverse. 2003-10-16 Jeffrey Stedfast * camel-sasl-kerberos4.c: Fixed a #include. 2003-10-10 Not Zed * providers/local/camel-local-provider.c: set the url fragment flag for local providers. * camel-provider.h: Move the URL_PART_NEED bits to the high 16 bits, to allow for easier changes in the future. Added a URL_PART_FRAGMENT flag for providers that use fragment = folder path. 2003-10-09 Jeffrey Stedfast * camel-mime-utils.c (header_decode_date): Allow timezone offsets to be up to 14 hours ahead of UTC. Fixes bug #49357. * broken-date-parser.c (get_tzone): Same. 2003-10-08 Jeffrey Stedfast * providers/smtp/camel-smtp-transport.c (smtp_helo): Removed an unused variable. 2003-09-25 Jeffrey Stedfast * providers/smtp/camel-smtp-transport.c (smtp_helo): If the localhost lookup results in a numeric IPv6 host, use the form "[IPv6:]" as specified in rfc2821. Fixes bug #46006. 2003-09-23 Ettore Perazzoli * providers/local/camel-local-provider.c: Set the IS_STORAGE bit in the mbox provider, since it can now contain a hierarchy of folders. 2003-09-23 Jeffrey Stedfast * providers/local/camel-mbox-store.c (get_folder): If the CREATE flag is set and the parent .sbd directory does not exist, create it. * camel-mime-utils.c (append_8bit): Don't forget to flush the iconv conversion. * tests/message/test4.c (main): Don't try dot-files. 2003-09-22 Not Zed ** See bug #41610 * providers/pop3/camel-pop3-folder.c (cmd_tocache): protect a divide by 0 for 0 length messages. 2003-09-22 Not Zed * providers/imap/camel-imap-provider.c: Added "offline_sync" option, which lets you synchronise all mail to local storage automagically. * camel-disco-folder.c (cdf_folder_changed): hook onto the folder changed single, for all new messages, check that they are online using another thread, if the offline_sync option has been enabled for this store. 2003-09-21 Not Zed * camel-session.c (session_thread_destroy): call proper entry point for freeing the message. 2003-09-18 Not Zed * camel-folder.c (filter_filter): register the filtering process for progress, and do progress of the filtering process. 2003-09-17 Not Zed * camel.c (camel_init): init camel operation. * camel-operation.c (camel_operation_reset): removed, not used, not worth it. (camel_operation_mute): new method to stop all status updates permanently. (*): Changed to use thread specific data and a list rather than a hashtable. (cancel_thread): removed. (camel_operation_register): return the previously registered op. 2003-09-22 Jeffrey Stedfast * providers/nntp/camel-nntp-store.c (connect_to_server): Fix the code that creates a new ssl stream to pass the correct arguments and the proper flags. * providers/imapp/camel-imapp-folder.c (imap_sync): Cast the CamelFolder to a CamelIMAPPFolder to hush some compiler warnings. * camel-mime-utils.h: Define a struct _CamelContentDisposition (allows the imapp code to compile) * providers/imapp/camel-imapp-driver.c: #include 2003-09-18 Jeffrey Stedfast * camel-mime-utils.c (camel_transfer_encoding_to_string): New function to replace the one from camel-mime-part.c (camel_transfer_encoding_from_string): Same. (camel_content_transfer_encoding_decode): Renamed from camel_header_content_encoding_decode(). * camel-mime-part.c (camel_mime_part_encoding_to_string): Removed. (camel_mime_part_encoding_from_string): Removed. * camel-data-wrapper.[c,h]: updated for CamelTransferEncoding namespace changes * camel-folder-summary.c: updated for CamelTransferEncoding namespace changes * camel-mime-filter-bestenc.[c,h]: updated for CamelTransferEncoding namespace changes * camel-mime-message.c: updated for CamelTransferEncoding namespace changes * camel-mime-part-utils.c: updated for CamelTransferEncoding namespace changes * camel-multipart-signed.c: updated for CamelTransferEncoding namespace changes * camel-smime-context.c: updated for CamelTransferEncoding namespace changes * providers/imapp/camel-imapp-utils.c: updated for CamelTransferEncoding namespace changes * tests/lib/messages.c: updated for CamelTransferEncoding namespace changes * tests/message/test1.c: updated for CamelTransferEncoding namespace changes 2003-09-18 Jeffrey Stedfast * camel-mime-utils.[c,h]: namespaced the encoding/decoding routines. * camel-mime-filter-basic.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c * camel-multipart.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c * camel-sasl-digest-md5.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c * camel-sasl.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c * camel-vee-folder.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c * providers/imap/camel-imap-search.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c * providers/pop3/camel-pop3-folder.c: updated for namespace changes to the encoding/decoding routines in camel-mime-utils.c 2003-08-26 Jeffrey Stedfast * camel-mime-parser.[c,h]: s/HSCAN_/CAMEL_MIME_PARSER_STATE_/g and s/_header_state/_camel_mime_parser_state/g * camel-filter-driver.c: Same. * camel-folder-summary.c: Here too. * camel-http-stream.c: And here. * camel-mime-message.c: ... * camel-mime-part-utils.c: ... * camel-mime-part.c: ... * camel-movemail.c: ... * camel-multipart-signed.c: ... * camel-multipart.c: ... * providers/local/camel-mbox-folder.c: ... * providers/local/camel-mbox-summary.c: ... * providers/local/camel-mh-summary.c: ... * providers/nntp/camel-nntp-summary.c: ... * providers/pop3/camel-pop3-folder.c: ... 2003-08-25 Jeffrey Stedfast * camel-mime-utils.[c,h]: Namespaced. * camel-data-wrapper.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-digest-folder.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-filter-driver.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-filter-search.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-folder-search.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-folder-summary.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * camel-http-stream.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-http-stream.h: updated for namespace changed made to camel-mime-utils.[c,h] * camel-internet-address.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-medium.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * camel-mime-message.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-mime-parser.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * camel-mime-part-utils.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-mime-part.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * camel-movemail.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-multipart-encrypted.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-multipart-signed.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-multipart.c: updated for namespace changed made to camel-mime-utils.[c,h] * camel-search-private.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * camel-types.h: updated for namespace changed made to camel-mime-utils.[c,h] * providers/imap/camel-imap-folder.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/imap/camel-imap-store-summary.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/imap/camel-imap-utils.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/imapp/camel-imapp-utils.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * providers/local/camel-local-summary.[c,h]: updated for namespace changed made to camel-mime-utils.[c,h] * providers/local/camel-maildir-summary.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/local/camel-mbox-summary.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/local/camel-spool-summary.h: updated for namespace changed made to camel-mime-utils.[c,h] * providers/nntp/camel-nntp-summary.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/nntp/camel-nntp-utils.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/pop3/camel-pop3-folder.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/sendmail/camel-sendmail-transport.c: updated for namespace changed made to camel-mime-utils.[c,h] * providers/smtp/camel-smtp-transport.c: updated for namespace changed made to camel-mime-utils.[c,h] 2003-09-16 Jeffrey Stedfast * providers/local/camel-mbox-folder.c (camel_mbox_folder_get_full_path): Implement a temp hack so trunk works until we merge in new-ui-branch. * camel-stream-filter.c (do_flush): Don't warning about how we haven't written anything to the stream, this is not an error. fflush() doesn't care if you try to fflush() a stream without writing to it, so we shouldn't care either. 2003-09-15 Not Zed * providers/imapp/camel-imapp-store.c (store_resp_list) (imap_login, try_sasl, imap_try_authenticate): removed dead code. * providers/imapp/camel-imapp-stream.c: return -1 if stream not set. * providers/imapp/camel-imapp-engine.c (iterate_completion): put done request on the done queue, so all requests are always somewhere. (camel_imapp_engine_command_free): just spit warnings of active messages being freed, but abort if the item isn't in any list. Also remove the node from its list before going on. (iterate_untagged, iterate_continuation, iterate_completion): staticifiy. * providers/imapp/camel-imapp-provider.c (camel_imapp_module_init): move camel_exception_setup call here. * providers/imapp/camel-imapp-driver.c (camel_imapp_driver_get_type): remove execption setup here, it isn't early enough. (camel_imapp_driver_list): handle exceptions. 2003-09-12 Jeffrey Stedfast * providers/local/camel-mbox-folder.c (camel_mbox_folder_get_full_path): Implements CamelLocalFolder::get_full_path() (publicly namespaced so that CamelMboxStore can re-use them). (camel_mbox_folder_get_meta_path): Same. * providers/local/camel-mbox-store.c (get_folder): Changed the way the path is constructed, since we now handle subdirectories and stuff. (delete_folder): Try deleting the Folder.sbd directory. We also need to manage our own meta files since CamelLocalStore's impl constructs paths differently than what we need. (create_folder): Implemented. (rename_folder): Implemented. (scan_dir): Scan an mbox tree (get_folder_info): Implemented using scan_dir(). * providers/local/camel-local-store.c (delete_folder): Set fi->url to the correct value, meaning we need to prefix it with the protocol and the folder_name is not actually part of the path, it is a separate component to the url. * providers/local/camel-local-folder.c (camel_local_folder_construct): Use the new class virtual method to construct the full folder path and all the meta files. (local_get_full_path): Implemented default get_full_path method. (local_get_meta_path): Implemented default get_meta_path method. 2003-09-11 Dan Winship * Makefile.am (noinst_LTLIBRARIES): Remove libcamel-static.la 2003-09-05 Not Zed * providers/imap/camel-imap-store.c (imap_noop): call camel_folder_sync bypassing the folder lock. See imap_store_refresh_folders too. 2003-09-04 David Woodhouse * providers/camel-imap-store.[ch]: Add PREAUTH handling and pine/mutt/etpan/etc.-style 'ssh exec imapd' support. 2003-09-03 David Woodhouse * camel-stream-process.[ch]: New stream implementation for running commands. * Makefile.am: Compile the above 2003-08-20 Not Zed ** See bug #47765. * camel-folder-search.h: Removed match1 member. * camel-folder-search.c (camel_folder_search_match_expression): use current directly rather than match1. This method isn't used anywhere anyway. (search_not): remove match1 stuff. (search_match_all): properly handle the match-all against 1 message as a scalar result, not an array result. 2003-09-03 Not Zed * camel-http-stream.c (camel_http_stream_set_proxy): handle NULL proxy_url - unset the proxy. 2003-08-29 Not Zed * camel-object.c (camel_object_state_write): (cobject_getv): (cobject_setv, cobject_state_read, cobject_state_read) (cobject_state_write): removed debug printfs. * providers/local/camel-local-folder.c (local_getv): Need to copy the local properties list before passing it out, since it's freed. 2003-08-27 Not Zed * providers/local/camel-local-folder.c (local_getv): implement PERSISTENT_PROPERTIES, for index mode. * camel-object.c (cobject_state_read): Also add property reading, and bump version to 1. (cobject_state_write): add persistent property writing. 2003-08-26 Not Zed * camel-folder.c (folder_getv): chain up properly. * camel-file-utils.c (camel_file_util_savename): helper to create a .#filename filename. * providers/local/camel-local-folder.c (camel_local_folder_construct): init meta-data for local folders. (local_getv): chain up properly, if args are not processed, rather than don't if they aren't. 2003-08-23 Not Zed * camel-object.c (cobject_class_init): added a new event, meta_changed. (camel_object_meta_set, camel_object_meta_get): meta-data api. (camel_object_free_hooks): Free meta-data if it is set on the object. * providers/local/camel-local-folder.c (camel_local_folder_get_type): setup a property list for local folders, just 'index_body' at present. 2003-08-25 Jeffrey Stedfast * camel-filter-driver.c (pipe_to_system): Added some more error checking for reading/writing to the pipes. Fixes bug #47880. 2003-08-21 Jeffrey Stedfast * camel-data-wrapper.c (decode_to_stream): Don't poke wrapper->stream directly, use camel_data_wrapper_write_to_stream() instead as this simplifies things and makes the imap data wrapper implementation Just Work (tm). * providers/imap/camel-imap-wrapper.c: changed prototype of write_to_stream() to return ssize_t. 2003-08-20 Jeffrey Stedfast * providers/imap/camel-imap-wrapper.c (imap_wrapper_hydrate): Make sure to ref the stream. Fixes bug #47749. 2003-08-18 Not Zed * camel-http-stream.c: Various fixes to make it work. * tests/smime/pgp-mime.c (main): added missing 'ret' variable. * providers/smtp/camel-smtp-transport.c (connect_to_server): * providers/imapp/camel-imapp-store.c (connect_to_server: * providers/imap/camel-imap-store.c (connect_to_server): * providers/pop3/camel-pop3-store.c (connect_to_server): * camel-http-stream.c (http_connect): change service->session for tcp_stream_ssl_new. * camel-tcp-stream-ssl.c: Changed service to session, and fix some refcounting of it. include camel-operation.h 2003-08-15 Not Zed ** See bug #47634. * tests/lib/messages.c (test_message_compare): check write_to_stream returns. (message_dump_rec): helper to dump message structure. * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): dont set content encoding here. (camel_mime_part_construct_content_from_parser): set it here instead, on every part. basically same as setting the mime_type_field always. * camel-multipart-signed.c (camel_multipart_signed_class_init): * camel-mime-message.c (camel_mime_message_class_init): * camel-multipart.c (camel_multipart_class_init): override decode_to_stream to always do the same as write_to_stream, since we can never be encoded. 2003-08-15 Jeffrey Stedfast * providers/imap/camel-imap-wrapper.c (imap_wrapper_hydrate): Don't attach any filters to decode base64/qp/etc. 2003-08-14 Jeffrey Stedfast * camel-mime-part.c (write_to_stream): Save errno when flushing/unreffing the filter stream. 2003-08-13 Jeffrey Stedfast * camel-mime-part.c (write_to_stream): If the content is base64/qp/uu/etc encoded but the part is 7bit/8bit/(or otherwise non-encoded), set reencode to TRUE so that we decode the original content stream. Fixes a bug noticed on evolution-patches@ximian.com where a patch had a Content-Transfer-Encoding of 7bit but was base64 encoded. 2003-08-13 Not Zed * camel-folder-summary.c (camel_folder_summary_remove_range): Fix the range check, we were stopping removal of 1 or 2 removals, for some odd and completely uncomprehensible reason. Perhaps debug left in? 2003-08-13 Not Zed ** See bug #47517. * camel-vee-folder.c (vee_sync): Always rebuild folder on any sync, not just expunge ones. 2003-08-11 Not Zed * providers/imapp/camel-imapp-store.c (imap_get_folder_info): force connect manually so basics work. ** See bug #45505. * camel-service.c (camel_gethostbyname): duh, pthread_create returns the error code directly, not via errno. (camel_gethostbyaddr): Same, also properly handle the failure case. 2003-08-01 Not Zed ** See bug #47208. * camel-filter-search.c (match_all): match-all with no arguments should always return TRUE. * camel-folder-search.c (camel_folder_search_execute_expression): print a warning when we get an invalid result type & fixed a leak for that case. 2003-08-08 Jeffrey Stedfast * tests/message/test4.c: New test suite for the mime parser (which is where the below 2 fixes were noticed). * camel-mime-parser.c (folder_boundary_check): Calculate 'len' by subtracting the boundary start from inend rather than 'atleast'. (folder_scan_content): Calculate 'inend' differently depending on the EOF state. 2003-08-08 Jeffrey Stedfast * camel-mime-filter-tohtml.c (html_convert): Rather than checking *inptr == '\n', check inptr >= inend - this gets rid of an Invalid Read report from valgrind. * camel-mime-part.c (write_to_stream): Don't necessarily re-encode just because the encodings differ. Need to look into making it so that message/rfc822 and multipart parts ignore the Content-Transfer-Encoding header and just keep their 'encoding' bits set to DEFAULT. 2003-08-05 Jeffrey Stedfast * providers/imap/camel-imap-folder.c (get_content): Updated. * camel-mime-message.c (camel_mime_message_init): Don't override the mime_type here. (process_header): Updated to use CamelDataWrapper's mime_type field. (find_best_encoding): Same. (best_encoding): Here too. * camel-digest-folder.c (camel_digest_folder_new): Updated for CamelMimePart::content_type change. * camel-mime-part.c (camel_mime_part_init): Override our parent class's default mime_type. (camel_mime_part_finalize): Don't need to unref the content_type anymore. (process_header): Updated to use CamelDataWrapper's mime_type field. (camel_mime_part_set_filename): Same. (camel_mime_part_get_filename): Same. (camel_mime_part_get_content_type): Same. (set_content_object): Here too. (write_to_stream): Updated. (construct_from_parser): Updated. * camel-mime-part.h: Remove the content_type field. 2003-07-31 Jeffrey Stedfast * tests/lib/messages.c (test_message_compare_content): If the chunks differ, perform a hexdump on the data being compared so that we may analyse it easier. * camel-multipart-signed.c (write_to_stream): Return ssize_t. * camel-mime-utils.h: Added the CamelMimePartEncodingType enum here. * camel-mime-part.h: Removed the CamelMimePartEncodingType enum from here. * camel-mime-part.c (write_to_stream): Updated to return ssize_t. Also minor changes to only re-encode the content stream if the charset or encoding changed (this way we write it out in the original raw form if nothing changed). * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): Drastically simplify. We no longer scan html content to try and find the charset, nor do we care about converting the content to UTF-8 and handling broken windows charsets. * camel-mime-message.c (find_best_encoding): Use decode_to_stream() here. Also updated to not assume the content charset is UTF-8 since it is very likely not the case anymore since data-wrappers no longer are converted to UTF-8 at parse time. * camel-folder-summary.c (summary_build_content_info_message): Use decode_to_stream instead here too. * camel-folder-search.c (match_words_1message): Use decode_to_stream instead of write_to_stream so we can search the contents. * camel-data-wrapper.c (camel_data_wrapper_init): Set the default encoding to DEFAULT. (write_to_stream): Updated to return ssize_t (camel_data_wrapper_decode_to_stream): New virtual function to decode a data wrapper to a stream (results in nearly identical behaviour to the old write_to_stream method). (decode_to_stream): Default implementation of above virtual method. Decodes base64/qp/etc streams. * camel-data-wrapper.h: Removed the rawtext bit and added an encoding member. 2003-08-01 Jeffrey Stedfast * tests/smime/pgp-mime.c: Same. * tests/smime/pgp.c: Updated to build and to import some custom gpg keys for use with testing. 2003-07-30 Jeffrey Stedfast * camel-mime-message.c (write_to_stream): Also updated. * camel-data-wrapper.c (write_to_stream): This should return ssize_t * camel-multipart-signed.c (write_to_stream): Updated. * camel-multipart.c (write_to_stream): Same. * camel-mime-part.c (write_to_stream): Here too. 2003-07-11 Suresh Chandrasekharan * camel-iconv.c: Fix for #46168 'some additional locale aliases required for chinese support'. 2003-07-25 Jeffrey Stedfast * camel-mime-utils.c (header_decode_word): Revert NotZed's fix for bug #42170 - this causes even more problems than it solves. See bug #46331 for info. Basically, each address header would be converted to UTF-8 twice which means no raw 8bit address header would render correctly. (header_decode_mailbox): Perform a sanity check on the resultant addr->str to make sure that it is valid UTF-8, if not convert it to UTF-8. Fixes bug #42170. 2003-07-23 Ettore Perazzoli * camel-provider.c (camel_provider_init): Print the provider directory as well, for debugging. 2003-07-23 Jeffrey Stedfast * camel-mime-message.c (find_best_encoding): Revert my previous changes to this function. * camel-mime-utils.h: Revert previous changes. * camel-mime-part.h: Revert previous changes. * camel-mime-part-utils.c: Revert previous changes. * camel-data-wrapper.c (camel_data_wrapper_init): Revert previous changes. (write_to_stream): Revert previous changes. 2003-07-23 Dan Winship * camel-block-file.c: #include camel-file-utils.h for camel_read() * camel-uid-cache.c (camel_uid_cache_save): Remove unused variable and label. * camel-url.c: #include camel-string-utils.h for camel_strdown * providers/pop3/camel-pop3-store.c (pop3_try_authenticate): Cast an (unsigned char *) to (char *) to fix a warning 2003-07-17 Jeffrey Stedfast * camel-mime-message.c (find_best_encoding): Updated to convert to/from the correct charset (since content is no longer necessarily in UTF-8). (best_encoding): Free the charset string when we're done with it. * camel-stream-fs.c (stream_read): Increment the seekable stream position by the number of bytes read. Oops. (stream_write): Same here. 2003-07-17 Timo Sirainen ** See bug #42573 * providers/imap/camel-imap-folder.c (do_append): Only free the response after we have finished the literal request, otherwise we could try processing folder updates incorrectly. 2003-07-14 Jeffrey Stedfast * camel-mime-utils.h: Add the CamelMimePartEncodingType definition here. * camel-mime-part.h: Remove the CamelMimePartEncodingType definition. * camel-mime-part-utils.c (simple_data_wrapper_construct_from_parser): Don't do any of the auto-detection we used to do here anymore. Just read the content into a memory buffer and record the encoding type. (camel_mime_part_construct_content_from_parser): Don't mangle the Content-Type struct here anymore. * camel-data-wrapper.c (camel_data_wrapper_init): Init encoding to DEFAULT. (write_to_stream): If the stream needs to be decoded, decode it. 2003-07-15 Jeffrey Stedfast * camel-stream-fs.c (stream_read): If we read 0 bytes, then set eos to TRUE. 2003-07-09 Jeffrey Stedfast Get rid of the #ifdef ENABLE_THREADS since we no longer plan to support/maintain this. * providers/nntp/camel-nntp-store.c: Here. * providers/nntp/camel-nntp-newsrc.c: And here. * providers/nntp/camel-nntp-folder.c: Same. * providers/local/camel-local-folder.c: And here. * camel-block-file.c: Here too. * 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. 2003-07-09 Larry Ewing * camel.h: remove reference to camel-pgp-mime.h 2003-07-08 Jeffrey Stedfast * camel-pgp-mime.[c,h]: Removed. * 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(). 2003-07-07 Jeffrey Stedfast * 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. 2003-07-07 Jeffrey Stedfast * 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. 2003-07-01 Jeffrey Stedfast * camel-sasl-digest-md5.c (digest_response): Don't quote the charset value, the qop value, nor the response value. Fixes bug #45712. 2003-07-01 Jeffrey Stedfast * camel-mime-utils.c (header_format_date): Use gmtime_r() instead of using gmtime() and memcpy() to try and be "atomic". 2003-06-30 Dan Winship * camel-folder-search.c (camel_folder_search_finalize): free the summary hash 2003-06-24 David Woodhouse * camel-mime-utils.c (header_format_date): Put day of week into outgoing email. 2003-06-25 Jeffrey Stedfast * camel-folder-summary.h: Added prototype for camel_folder_summary_remove_range(). 2003-06-25 Not Zed ** See bug #45386 * camel-service.c (camel_gethostbyname, camel_gethostbyaddr): Make sure we have an exception that we test against. 2003-06-20 Not Zed ** See bug #43887 * camel-mime-filter-enriched.c (camel_enriched_to_html): simple wrapper to convert enriched to html in one go. 2003-06-18 Not Zed * 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. 2003-06-13 Jeffrey Stedfast * providers/pop3/camel-pop3-folder.c (pop3_finalize): Made static (to match the prototype). 2003-06-13 Larry Ewing * camel-folder-thread.c (camel_folder_thread_messages_apply): don't leak the summary when reloading it. Fixes a very large leak. 2003-06-17 Not Zed * 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. 2003-06-16 Not Zed ** 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. 2003-06-16 Not Zed ** 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. 2003-06-05 Jeffrey Stedfast 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. 2003-06-11 Jeffrey Stedfast 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. 2003-06-04 Jeffrey Stedfast * 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). 2003-06-11 Larry Ewing * camel-text-index.c (text_index_normalise): use g_utf8_strdown properly. 2003-06-09 Jeffrey Stedfast * 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.