is now functional. The signal proxy needs to be tested
though. The thread proxy folder is being implemented.
A rough summary :
1999-10-13 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-folder.c (camel_folder_close): the
folder->close method is now asynchronous.
* camel/camel-folder-pt-proxy.c (_folder_open_cb):
(_open):
(_folder_open_cb):
(_open):
open/close method implemented in the thread proxy
folder. More to come.
* camel/camel-exception.c (camel_exception_xfer):
new utility func.
* camel/camel-marshal-utils.c: some new marshallers
* camel/camel-folder-pt-proxy.c: Some explanations
on the thread proxy system.
1999-10-11 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-marshal-utils.c:
camel/camel-marshal-utils.h:
Handles operation marshalling.
* camel/camel-thread-proxy.c:
camel/camel-thread-proxy.h:
new files. Generic proxy system.
* camel/camel-folder-pt-proxy.c
moved all proxy related code in dedicated files.
(camel_folder_pt_proxy_init):
removed proxy initialisation code
(_finalize):
removed proxy finalization code
* camel/camel-exception.c
(camel_exception_new):
(camel_exception_set):
(camel_exception_free):
New funcs.
svn path=/trunk/; revision=1328
* Added corba-like exception system to CamelFolder.
The exception system is not finished but I wanted
the async and event implementation to get this right
from the beginning.
svn path=/trunk/; revision=1256
1999-09-21 bertrand <Bertrand.Guiheneuf@aful.org>
* configure.in:
Check pthreads.
* camel/Makefile.am:
camel-folder-pt-proxy.c is only compiled
when pthreads are available.
svn path=/trunk/; revision=1247
1999-09-08 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/ui-tests/store_listing.c (close_all):
close all open folders and stores. (necessary
for UID list saving ).
* camel/providers/MH/mh-uid.c (mh_generate_uid_list):
fix: store UID list in CamelMHFolder object.
* camel/providers/MH/camel-mh-folder.c (_open):
read or create UID list.
(_close): save UID list.
svn path=/trunk/; revision=1192
1999-09-05 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/ui-tests/store_listing.c (show_folder_messages):
re-enabled old CPU/Mem consumming
method. Use ful for pop3 for instance.
svn path=/trunk/; revision=1179
1999-09-05 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/ui-tests/store_listing.c (show_folder_messages):
use folder summary instead of opening all messages.
* camel/providers/MH/camel-mh-folder.c (_create_summary):
basic and highly non-efficient summary implementation.
Should be seen as a proof of concept only.
subfolder summary still has to be implemented.
* camel/providers/maildir/camel-maildir-folder.c (_init_with_store):
hasn't summary for the moment.
* camel/providers/maildir/camel-maildir-folder.c
cosmetic changes.
svn path=/trunk/; revision=1178
1999-08-28 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-mime-part.c (camel_mime_part_set_text):
text parameter declared const
* camel/camel-mime-part-utils.c (camel_mime_part_store_stream_in_buffer):
actually test correctly nb_bytes_read_chunk is >0
* camel/gstring-util.c:
* camel/gmime-content-field.c:
* camel/providers/MH/camel-mh-folder.c:
* camel/camel-stream-fs.c:
include string.h
* camel/camel-stream-mem.c (_write): return
the numver of written bytes.
* camel/camel-stream-buffered-fs.c (_eos):
return sthg
* camel/camel-stream.c (default_camel_seek):
return something.
* Started to work on new recipient code.
* fixed various leaks.
svn path=/trunk/; revision=1146
1999-08-26 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-mime-part.c (_get_content_object):
contruct the content from the buffer before calling
CamelMedium implementation.
(_construct_from_stream): Do not construct the content
by default, just store the content bytes in
a temporary buffer. Content will be constructed only
at caller request (when calling CamelMedium::get_content_object)
Providers with better access to the messages (mbox/MH ...)
will have to provider lighter implementation, that is
shall not read content at all unless the caller asks
for it (again with get_content).
* camel/camel-mime-part-utils.c: new file, groups
mime-part related utils. Meant to be used by providers
subclassing MimeMessage.
(camel_mime_part_construct_headers_from_stream):
(camel_mime_part_construct_content_from_stream):
no more useless temporary hash table.
* camel/camel-mime-part.c (_construct_from_stream): calls
mime-part-utils functions now.
* camel/gmime-utils.c (_store_header_pair_from_string):
do not use hash table to store header, use an array instead.
svn path=/trunk/; revision=1145
1999-08-25 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-mime-part.c: now descend from CamelMedium.
* tests/test1.c (main): all headers must be strdup'ed
(main): unref created objects
svn path=/trunk/; revision=1143
1999-08-17 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-stream.c (camel_stream_read):
return the number of bytes read.
How can this have ever worked ?
(camel_stream_flush): don't return anything.
* camel/gmime-utils.c (get_header_table_from_stream):
mem leak fixed.
+ various other mem leaks.
svn path=/trunk/; revision=1120
1999-08-16 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/ui-tests/store_listing.c:
Now has a popup menu on mailbox tree to allow
easier tests. Implemented the copy stuff.
Works well for the MH provider :)))
* camel/providers/MH/camel-mh-folder.c (_copy_message_to):
Test MH provider fast copy implemented.
svn path=/trunk/; revision=1119
1999-08-15 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-folder.c (camel_folder_expunge):
moved the active list readjustment code here.
Much saner, providers won't have to worry about
that.
* camel/providers/MH/camel-mh-folder.c (_expunge):
no more active list readjustment stuff.
* camel/camel-folder.h: the expunge virtual no more
returns a list of expunged messages. Now providers
only have to set the expunge flag on the expunged
messages.
* camel/camel-folder.c (camel_folder_get_message):
moved the caching code here. Finally, I don't want
providers to crash the libs with a bad message cache
implementation.
(_close): do not call the CamelFolder virtual expunge
method directly, use camel_folder_expunge() instead.
(camel_folder_expunge): added the want_list param.
The client can decide if it wants the expunged message
list or not. If yes, it'll have to unref the messages
itself.
svn path=/trunk/; revision=1117
1999-08-13 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-folder.c (_get_message): default implementation
to be called first by providers methods. It looks in the
folder message list (in memory) to see if the message has
not already been retrieved, and in this case, returns
the same CamelMimeMessage object
Caching messages in folder kworks. Providers will have to
do their job correctly.
Enough for today.
svn path=/trunk/; revision=1113
1999-08-12 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/ui-tests/store_listing.c (show_folder_messages):
remove stupid debug code.
(add_mail_store): use camel_session_get_store instead
of creating MH store directly.
(main): load MH provider.
* camel/camel-provider.c (camel_provider_register_as_module):
register new provider.
(camel_provider_get_for_protocol):
Now, implementation is correct.
* camel/camel-store.c (_finalize):
* camel/camel-store.h (struct _CamelStore):
further disabled url_name field use.
URL will be generated dynamically.
Ben dam don dieu, vla t'y pas que ya tout le bourier qui marche !
(Autoload store/protocol from URL works)
svn path=/trunk/; revision=1108
1999-08-12 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/providers/MH/camel-mh-provider.c:
new file. MH provider registration stuff.
* camel/camel-provider.c (camel_provider_register_as_module):
load a provider from a shared object (plugin).
(camel_provider_register): register a provider
"by hand". Used for statically defined providers.
* tests/test7.c: new test.
tests providers loading framework.
Providers modules loading Works !!! :))))
svn path=/trunk/; revision=1105
memory (maybe 2000k) and I really don't
understand why. As soon as I get gtk working
on SunOS, purify should hopefully give me the answer.
1999-08-11 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-service.c (_finalize):
* camel/camel-stream-fs.c (_finalize):
(_destroy): close file descriptor.
* camel/camel-stream-mem.c (_finalize):
* camel/camel-store.c (_finalize):
* camel/camel-folder.c (_finalize):
svn path=/trunk/; revision=1104
1999-08-09 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/ui-tests/store_listing.c:
* tests/ui-tests/store_listing.glade:
Quick and (very) dirty hack to test Camel more
easily.
* camel/camel-folder.c (camel_folder_append_message):
new method.
svn path=/trunk/; revision=1102
1999-08-08 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/providers/MH/camel-mh-folder.c (_get_message):
fix. In MH, message number is not related to
message position in folder.
svn path=/trunk/; revision=1097
1999-08-06 Robert Brady <rwb197@ecs.soton.ac.uk>
* tests/test6.c: encoder test.
* camel/gmime-rfc2047.c: Fixed decoder bug : sequence
?= is not always the terminator for an encoded-string.
svn path=/trunk/; revision=1091
1999-08-06 bertrand <Bertrand.Guiheneuf@aful.org>
* tests/test4.c (main): added real test for MH folder
provider. All tested things seem to work OK :)
svn path=/trunk/; revision=1089
1999-08-06 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/providers/MH/camel-mh-folder.c (_exists): add debug information
(_list_subfolders): test if first char in folder name is not '.'
before adding it to the folder list.
* camel/camel-store.c (_init):
disable session check temporarily
(_get_separator):
(_get_folder): new static func.
Fixed several oddities in class definition.
* camel/providers/MH/camel-mh-store.c (camel_mh_store_get_type):
parent type is CAMEL_STORE_TYPE not CAMEL_FOLDER_TYPE
* camel/camel-store.c:
prent class is CamelServiceClass not GtkObjectClass
* camel/url-util.c :
cosmetic changes + use of const when possible.
(find_host):
fix a bug: when there is no host and no port don't skip the '/'
all static find_* func are now named _func_*
(g_url_free): destructor func.
cache field has been disabled. Constructing the url string
won't be too slow and will occur rarely enough that we
do not need to add complexity to this code.
* camel/providers/MH/camel-mh-store.c:
parent class is CamelStorClass not GtkObjectClass
svn path=/trunk/; revision=1087
* camel/camel-mime-part.c (_set_content_object):
There is a probleme here. We can not allow mime part
content-type field and content_object mime-type to be
different. I thus chosed to set mime part object
content field to be freed (if necessary) and set
to be a pointer to content_object mime type
field.
(_construct_from_stream): set content_object mime type
to be the same as mime_part's one. This is necessary
because we use _set_content_type.
This two things are a bit hackish ansd may need
to be redesigned.
svn path=/trunk/; revision=1071
Making it work with other types is now just a matter of writing
the various data wrappers. And display them will just be a matter of writing
the good bonobo components.
1999-08-03 bertrand <Bertrand.Guiheneuf@aful.org>
* camel/camel-simple-data-wrapper.c (_construct_from_stream):
more debugging output + nb_bytes_read is now a signed int
to avoid bug when eos is encountered.
* camel/camel-mime-part.c (_construct_from_stream):
sync to data_wrapper_repository function name changes.
Use default "text/plain" type when conten-type field
is not found. (following RFC 2046 spec).
* camel/data-wrapper-repository.c (data_wrapper_repository_set_data_wrapper_type):
(data_wrapper_repository_get_data_wrapper_type):
change function name prefix (s/data_wrapper/data_wrapper_repository/)
* camel/camel-multipart.c (_read_part):
add `\n` at eol but not before boundary.
* camel/gmime-utils.c (get_header_table_from_stream):
correct implementation of end of stream detection.
svn path=/trunk/; revision=1070
1999-06-28 bertrand <Bertrand.Guiheneuf@inria.fr>
* tests/test2.c (main): now use
CamelDataWrapper::contruct_form_stream to test
message parsing
* camel/camel-data-wrapper.c:
* camel/camel-data-wrapper.h:
construct_from_stream no longer has maximimum size arg.
* camel/camel-mime-part.c (_construct_from_stream): new.
Construct the mime_part from a stream.
svn path=/trunk/; revision=1003
1999-05-31 bertrand <Bertrand.Guiheneuf@inria.fr>
* tests/test2.c (main):
use new stream code instead of raw file * stuff.
* camel/gmime-utils.c (get_header_table_from_stream):
new func. Will replace get_header_table_from_file and will
be used to parse headers from files as well as from
memory buffers.
* camel/camel-stream-fs.c:
CamelStream Subclass. File system based
stream.
svn path=/trunk/; revision=961
* camel/camel-mime-message.c (_set_recipient_list_from_string):
remove leading and trailing spaces in recipient addresses.
* camel/gmime-utils.c (_store_header_pair_from_gstring):
remove leading and trailing spaces from header values.
* camel/gstring-util.c (g_string_trim): new
func: remove leading or trailng chars from
a specified char set.
(g_string_split): allow trimming of substrings.
* tests/test1.c (main): remove gtk_main call
svn path=/trunk/; revision=958
* tests/test2.c (main): rewrite message obtained via
parsing into a file. Actually, it works pretty well :))
* camel/camel-mime-message.c (_set_recipient_list_from_string):
create recipient list form a comma separated string.
(_parse_header_pair): added recipient lists parsing.
* camel/camel-mime-part.c (_parse_header_pair):
new (protected) method. Parse a head pair and
decides what to do with it.
(_add_header): Call in _parse_header_pair
* camel/camel-mime-message.c (_parse_header_pair):
overload header parsing MimePart mthod.
* camel/gstring-util.c (g_string_split):
new func: split a gstring into a GList of
substring.
svn path=/trunk/; revision=950
