X-Git-Url: https://git.notmuchmail.org/git?p=notmuch;a=blobdiff_plain;f=lib%2Fthread.cc;h=c126aac8cc8b945e9018f7505f7910174c83535e;hp=b8be3e1b5e889f5b873fe21fe50c6a0c6038db44;hb=16aa65ba2575fd504c31d9671d8c5150f8e8adf1;hpb=cd19671f51b8f87303de6ab536afd6aa36efec61 diff --git a/lib/thread.cc b/lib/thread.cc index b8be3e1b..c126aac8 100644 --- a/lib/thread.cc +++ b/lib/thread.cc @@ -21,21 +21,25 @@ #include "notmuch-private.h" #include "database-private.h" -#include - #include #include /* GHashTable */ -struct _notmuch_thread { +struct visible _notmuch_thread { notmuch_database_t *notmuch; char *thread_id; char *subject; GHashTable *authors_hash; + GPtrArray *authors_array; + GHashTable *matched_authors_hash; + GPtrArray *matched_authors_array; char *authors; - char *nonmatched_authors; GHashTable *tags; + /* All messages, oldest first. */ notmuch_message_list_t *message_list; + /* Top-level messages, oldest first. */ + notmuch_message_list_t *toplevel_list; + GHashTable *message_hash; int total_messages; int matched_messages; @@ -47,16 +51,31 @@ static int _notmuch_thread_destructor (notmuch_thread_t *thread) { g_hash_table_unref (thread->authors_hash); + g_hash_table_unref (thread->matched_authors_hash); g_hash_table_unref (thread->tags); g_hash_table_unref (thread->message_hash); + if (thread->authors_array) { + g_ptr_array_free (thread->authors_array, TRUE); + thread->authors_array = NULL; + } + + if (thread->matched_authors_array) { + g_ptr_array_free (thread->matched_authors_array, TRUE); + thread->matched_authors_array = NULL; + } + return 0; } +/* Add each author of the thread to the thread's authors_hash and to + * the thread's authors_array. */ static void _thread_add_author (notmuch_thread_t *thread, const char *author) { + char *author_copy; + if (author == NULL) return; @@ -64,97 +83,90 @@ _thread_add_author (notmuch_thread_t *thread, author, NULL, NULL)) return; - g_hash_table_insert (thread->authors_hash, xstrdup (author), NULL); + author_copy = talloc_strdup (thread, author); - if (thread->authors) - thread->authors = talloc_asprintf (thread, "%s, %s", - thread->authors, - author); - else - thread->authors = talloc_strdup (thread, author); + g_hash_table_insert (thread->authors_hash, author_copy, NULL); + + g_ptr_array_add (thread->authors_array, author_copy); } -/* - * move authors of matched messages in the thread to - * the front of the authors list, but keep them in - * existing order within their group - */ +/* Add each matched author of the thread to the thread's + * matched_authors_hash and to the thread's matched_authors_array. */ static void -_thread_move_matched_author (notmuch_thread_t *thread, - const char *author) +_thread_add_matched_author (notmuch_thread_t *thread, + const char *author) { - char *authors_copy; - char *current_author; - char *last_pipe,*next_pipe; - int idx,nm_start,author_len,authors_len; + char *author_copy; - if (thread->authors == NULL || author == NULL) - return; - if (thread->nonmatched_authors == NULL) - thread->nonmatched_authors = thread->authors; - author_len = strlen(author); - authors_len = strlen(thread->authors); - if (author_len == authors_len) { - /* just one author */ - thread->nonmatched_authors += author_len; - return; - } - current_author = strcasestr(thread->authors, author); - if (current_author == NULL) + if (author == NULL) return; - /* how far inside the nonmatched authors is our author? */ - idx = current_author - thread->nonmatched_authors; - if (idx < 0) { - /* already among matched authors */ + + if (g_hash_table_lookup_extended (thread->matched_authors_hash, + author, NULL, NULL)) return; + + author_copy = talloc_strdup (thread, author); + + g_hash_table_insert (thread->matched_authors_hash, author_copy, NULL); + + g_ptr_array_add (thread->matched_authors_array, author_copy); +} + +/* Construct an authors string from matched_authors_array and + * authors_array. The string contains matched authors first, then + * non-matched authors (with the two groups separated by '|'). Within + * each group, authors are listed in date order. */ +static void +_resolve_thread_authors_string (notmuch_thread_t *thread) +{ + unsigned int i; + char *author; + int first_non_matched_author = 1; + + /* First, list all matched authors in date order. */ + for (i = 0; i < thread->matched_authors_array->len; i++) { + author = (char *) g_ptr_array_index (thread->matched_authors_array, i); + if (thread->authors) + thread->authors = talloc_asprintf (thread, "%s, %s", + thread->authors, + author); + else + thread->authors = author; } - /* are there any authors in the list after our author? */ - if (thread->nonmatched_authors + author_len < thread->authors + authors_len) { - /* we have to make changes, so let's get a temp copy */ - authors_copy = talloc_strdup(thread,thread->authors); - /* nm_start is the offset into where the non-matched authors start */ - nm_start = thread->nonmatched_authors - thread->authors; - /* copy this author and add the "| " - the if clause above tells us there's more */ - strncpy(thread->nonmatched_authors,author,author_len); - strncpy(thread->nonmatched_authors+author_len,"| ",2); - thread->nonmatched_authors += author_len+2; - if (idx > 0) { - /* we are actually moving authors around, not just changing the separator - * first copy the authors that came BEFORE our author */ - strncpy(thread->nonmatched_authors, authors_copy+nm_start, idx-2); - /* finally, if there are authors AFTER our author, copy those */ - if(author_len+nm_start+idx < authors_len) { - strncpy(thread->nonmatched_authors + idx - 2,", ",2); - strncpy(thread->nonmatched_authors + idx, authors_copy+nm_start + idx + author_len + 2, - authors_len - (nm_start + idx + author_len + 2)); - } - } - /* finally let's make sure there's just one '|' in the authors string */ - last_pipe = strchr(thread->authors,'|'); - while (last_pipe) { - next_pipe = strchr(last_pipe+1,'|'); - if (next_pipe) - *last_pipe = ','; - last_pipe = next_pipe; + + /* Next, append any non-matched authors that haven't already appeared. */ + for (i = 0; i < thread->authors_array->len; i++) { + author = (char *) g_ptr_array_index (thread->authors_array, i); + if (g_hash_table_lookup_extended (thread->matched_authors_hash, + author, NULL, NULL)) + continue; + if (first_non_matched_author) { + thread->authors = talloc_asprintf (thread, "%s| %s", + thread->authors, + author); + } else { + thread->authors = talloc_asprintf (thread, "%s, %s", + thread->authors, + author); } - } else { - thread->nonmatched_authors += author_len; - /* so now all authors are matched - let's remove the '|' */ - last_pipe = strchr(thread->authors,'|'); - if (last_pipe) - *last_pipe = ','; + + first_non_matched_author = 0; } - return; + + g_ptr_array_free (thread->authors_array, TRUE); + thread->authors_array = NULL; + g_ptr_array_free (thread->matched_authors_array, TRUE); + thread->matched_authors_array = NULL; } -/* clean up the uggly "Lastname, Firstname" format that some mail systems +/* clean up the ugly "Lastname, Firstname" format that some mail systems * (most notably, Exchange) are creating to be "Firstname Lastname" * To make sure that we don't change other potential situations where a * comma is in the name, we check that we match one of these patterns * "Last, First" * "Last, First MI" */ -char * +static char * _thread_cleanup_author (notmuch_thread_t *thread, const char *author, const char *from) { @@ -163,11 +175,19 @@ _thread_cleanup_author (notmuch_thread_t *thread, char *blank; int fname,lname; + if (author == NULL) + return NULL; clean_author = talloc_strdup(thread, author); if (clean_author == NULL) return NULL; + /* check if there's a comma in the name and that there's a + * component of the name behind it (so the name doesn't end with + * the comma - in which case the string that strchr finds is just + * one character long ",\0"). + * Otherwise just return the copy of the original author name that + * we just made*/ comma = strchr(author,','); - if (comma) { + if (comma && strlen(comma) > 1) { /* let's assemble what we think is the correct name */ lname = comma - author; fname = strlen(author) - lname - 2; @@ -187,7 +207,6 @@ _thread_cleanup_author (notmuch_thread_t *thread, /* we didn't identify this as part of the email address * so let's punt and return the original author */ strcpy (clean_author, author); - } return clean_author; } @@ -199,7 +218,8 @@ _thread_cleanup_author (notmuch_thread_t *thread, */ static void _thread_add_message (notmuch_thread_t *thread, - notmuch_message_t *message) + notmuch_message_t *message, + notmuch_string_list_t *exclude_terms) { notmuch_tags_t *tags; const char *tag; @@ -247,6 +267,15 @@ _thread_add_message (notmuch_thread_t *thread, notmuch_tags_move_to_next (tags)) { tag = notmuch_tags_get (tags); + /* Mark excluded messages. */ + for (notmuch_string_node_t *term = exclude_terms->head; term; + term = term->next) { + /* We ignore initial 'K'. */ + if (strcmp(tag, (term->string + 1)) == 0) { + notmuch_message_set_flag (message, NOTMUCH_MESSAGE_FLAG_EXCLUDED, TRUE); + break; + } + } g_hash_table_insert (thread->tags, xstrdup (tag), NULL); } } @@ -280,6 +309,10 @@ _thread_set_subject_from_message (notmuch_thread_t *thread, thread->subject = talloc_strdup (thread, cleaned_subject); } +/* Add a message to this thread which is known to match the original + * search specification. The 'sort' parameter controls whether the + * oldest or newest matching subject is applied to the thread as a + * whole. */ static void _thread_add_matched_message (notmuch_thread_t *thread, notmuch_message_t *message, @@ -290,13 +323,20 @@ _thread_add_matched_message (notmuch_thread_t *thread, date = notmuch_message_get_date (message); - if (date < thread->oldest || ! thread->matched_messages) + if (date < thread->oldest || ! thread->matched_messages) { thread->oldest = date; + if (sort == NOTMUCH_SORT_OLDEST_FIRST) + _thread_set_subject_from_message (thread, message); + } - if (date > thread->newest || ! thread->matched_messages) + if (date > thread->newest || ! thread->matched_messages) { thread->newest = date; + if (sort != NOTMUCH_SORT_OLDEST_FIRST) + _thread_set_subject_from_message (thread, message); + } - thread->matched_messages++; + if (!notmuch_message_get_flag (message, NOTMUCH_MESSAGE_FLAG_EXCLUDED)) + thread->matched_messages++; if (g_hash_table_lookup_extended (thread->message_hash, notmuch_message_get_message_id (message), NULL, @@ -304,39 +344,27 @@ _thread_add_matched_message (notmuch_thread_t *thread, notmuch_message_set_flag (hashed_message, NOTMUCH_MESSAGE_FLAG_MATCH, 1); } - _thread_move_matched_author (thread,notmuch_message_get_author(hashed_message)); - if ((sort == NOTMUCH_SORT_OLDEST_FIRST && date <= thread->newest) || - (sort != NOTMUCH_SORT_OLDEST_FIRST && date == thread->newest)) - { - _thread_set_subject_from_message (thread, message); - } + _thread_add_matched_author (thread, notmuch_message_get_author (hashed_message)); } static void -_resolve_thread_relationships (unused (notmuch_thread_t *thread)) +_resolve_thread_relationships (notmuch_thread_t *thread) { - notmuch_message_node_t **prev, *node; + notmuch_message_node_t *node; notmuch_message_t *message, *parent; const char *in_reply_to; - prev = &thread->message_list->head; - while ((node = *prev)) { + for (node = thread->message_list->head; node; node = node->next) { message = node->message; in_reply_to = _notmuch_message_get_in_reply_to (message); if (in_reply_to && strlen (in_reply_to) && g_hash_table_lookup_extended (thread->message_hash, in_reply_to, NULL, (void **) &parent)) - { - *prev = node->next; - if (thread->message_list->tail == &node->next) - thread->message_list->tail = prev; - node->next = NULL; - _notmuch_message_add_reply (parent, node); - } else { - prev = &((*prev)->next); - } + _notmuch_message_add_reply (parent, message); + else + _notmuch_message_list_add_message (thread->toplevel_list, message); } /* XXX: After scanning through the entire list looking for parents @@ -351,16 +379,17 @@ _resolve_thread_relationships (unused (notmuch_thread_t *thread)) */ } -/* Create a new notmuch_thread_t object for the given thread ID, - * treating any messages matching 'query_string' as "matched". +/* Create a new notmuch_thread_t object by finding the thread + * containing the message with the given doc ID, treating any messages + * contained in match_set as "matched". Remove all messages in the + * thread from match_set. * - * Creating the thread will trigger two database searches. The first - * is for all messages belonging to the thread, (to get the first - * subject line, the total count of messages, and all authors). The - * second search is for all messages that are in the thread and that - * also match the given query_string. This is to allow for a separate - * count of matched messages, and to allow a viewer to display these - * messages differently. + * Creating the thread will perform a database search to get all + * messages belonging to the thread and will get the first subject + * line, the total count of messages, and all authors in the thread. + * Each message in the thread is checked against match_set to allow + * for a separate count of matched messages, and to allow a viewer to + * display these messages differently. * * Here, 'ctx' is talloc context for the resulting thread object. * @@ -369,60 +398,38 @@ _resolve_thread_relationships (unused (notmuch_thread_t *thread)) notmuch_thread_t * _notmuch_thread_create (void *ctx, notmuch_database_t *notmuch, - const char *thread_id, - const char *query_string, + unsigned int seed_doc_id, + notmuch_doc_id_set_t *match_set, + notmuch_string_list_t *exclude_terms, notmuch_sort_t sort) { - notmuch_thread_t *thread; - const char *thread_id_query_string; + void *local = talloc_new (ctx); + notmuch_thread_t *thread = NULL; + notmuch_message_t *seed_message; + const char *thread_id; + char *thread_id_query_string; notmuch_query_t *thread_id_query; notmuch_messages_t *messages; notmuch_message_t *message; - notmuch_bool_t matched_is_subset_of_thread; - thread_id_query_string = talloc_asprintf (ctx, "thread:%s", thread_id); - if (unlikely (query_string == NULL)) - return NULL; + seed_message = _notmuch_message_create (local, notmuch, seed_doc_id, NULL); + if (! seed_message) + INTERNAL_ERROR ("Thread seed message %u does not exist", seed_doc_id); - /* Under normal circumstances we need to do two database - * queries. One is for the thread itself (thread_id_query_string) - * and the second is to determine which messages in that thread - * match the original query (matched_query_string). - * - * But under two circumstances, we use only the - * thread_id_query_string: - * - * 1. If the original query_string *is* just the thread - * specification. - * - * 2. If the original query_string matches all messages ("" or - * "*"). - * - * In either of these cases, we can be more efficient by running - * just the thread_id query (since we know all messages in the - * thread will match the query_string). - * - * Beyond the performance advantage, in the second case, it's - * important to not try to create a concatenated query because our - * parser handles "" and "*" as special cases and will not do the - * right thing with a query string of "* and thread:". - **/ - matched_is_subset_of_thread = 1; - if (strcmp (query_string, thread_id_query_string) == 0 || - strcmp (query_string, "") == 0 || - strcmp (query_string, "*") == 0) - { - matched_is_subset_of_thread = 0; - } + thread_id = notmuch_message_get_thread_id (seed_message); + thread_id_query_string = talloc_asprintf (local, "thread:%s", thread_id); + if (unlikely (thread_id_query_string == NULL)) + goto DONE; - thread_id_query = notmuch_query_create (notmuch, thread_id_query_string); + thread_id_query = talloc_steal ( + local, notmuch_query_create (notmuch, thread_id_query_string)); if (unlikely (thread_id_query == NULL)) - return NULL; + goto DONE; - thread = talloc (ctx, notmuch_thread_t); + thread = talloc (local, notmuch_thread_t); if (unlikely (thread == NULL)) - return NULL; + goto DONE; talloc_set_destructor (thread, _notmuch_thread_destructor); @@ -430,15 +437,23 @@ _notmuch_thread_create (void *ctx, thread->thread_id = talloc_strdup (thread, thread_id); thread->subject = NULL; thread->authors_hash = g_hash_table_new_full (g_str_hash, g_str_equal, - free, NULL); + NULL, NULL); + thread->authors_array = g_ptr_array_new (); + thread->matched_authors_hash = g_hash_table_new_full (g_str_hash, + g_str_equal, + NULL, NULL); + thread->matched_authors_array = g_ptr_array_new (); thread->authors = NULL; - thread->nonmatched_authors = NULL; thread->tags = g_hash_table_new_full (g_str_hash, g_str_equal, free, NULL); thread->message_list = _notmuch_message_list_create (thread); - if (unlikely (thread->message_list == NULL)) - return NULL; + thread->toplevel_list = _notmuch_message_list_create (thread); + if (unlikely (thread->message_list == NULL || + thread->toplevel_list == NULL)) { + thread = NULL; + goto DONE; + } thread->message_hash = g_hash_table_new_full (g_str_hash, g_str_equal, free, NULL); @@ -448,58 +463,53 @@ _notmuch_thread_create (void *ctx, thread->oldest = 0; thread->newest = 0; + /* We use oldest-first order unconditionally here to obtain the + * proper author ordering for the thread. The 'sort' parameter + * passed to this function is used only to indicate whether the + * oldest or newest subject is desired. */ notmuch_query_set_sort (thread_id_query, NOTMUCH_SORT_OLDEST_FIRST); for (messages = notmuch_query_search_messages (thread_id_query); notmuch_messages_valid (messages); notmuch_messages_move_to_next (messages)) { + unsigned int doc_id; + message = notmuch_messages_get (messages); + doc_id = _notmuch_message_get_doc_id (message); + if (doc_id == seed_doc_id) + message = seed_message; - _thread_add_message (thread, message); + _thread_add_message (thread, message, exclude_terms); - if (! matched_is_subset_of_thread) + if ( _notmuch_doc_id_set_contains (match_set, doc_id)) { + _notmuch_doc_id_set_remove (match_set, doc_id); _thread_add_matched_message (thread, message, sort); + } _notmuch_message_close (message); } - notmuch_query_destroy (thread_id_query); - - if (matched_is_subset_of_thread) - { - const char *matched_query_string; - notmuch_query_t *matched_query; - - matched_query_string = talloc_asprintf (ctx, "%s AND (%s)", - thread_id_query_string, - query_string); - if (unlikely (matched_query_string == NULL)) - return NULL; - - matched_query = notmuch_query_create (notmuch, matched_query_string); - if (unlikely (matched_query == NULL)) - return NULL; - - for (messages = notmuch_query_search_messages (matched_query); - notmuch_messages_valid (messages); - notmuch_messages_move_to_next (messages)) - { - message = notmuch_messages_get (messages); - _thread_add_matched_message (thread, message, sort); - _notmuch_message_close (message); - } - - notmuch_query_destroy (matched_query); - } + _resolve_thread_authors_string (thread); _resolve_thread_relationships (thread); + /* Commit to returning thread. */ + talloc_steal (ctx, thread); + + DONE: + talloc_free (local); return thread; } notmuch_messages_t * notmuch_thread_get_toplevel_messages (notmuch_thread_t *thread) +{ + return _notmuch_messages_create (thread->toplevel_list); +} + +notmuch_messages_t * +notmuch_thread_get_messages (notmuch_thread_t *thread) { return _notmuch_messages_create (thread->message_list); } @@ -549,23 +559,23 @@ notmuch_thread_get_newest_date (notmuch_thread_t *thread) notmuch_tags_t * notmuch_thread_get_tags (notmuch_thread_t *thread) { - notmuch_tags_t *tags; + notmuch_string_list_t *tags; GList *keys, *l; - tags = _notmuch_tags_create (thread); + tags = _notmuch_string_list_create (thread); if (unlikely (tags == NULL)) return NULL; keys = g_hash_table_get_keys (thread->tags); for (l = keys; l; l = l->next) - _notmuch_tags_add_tag (tags, (char *) l->data); + _notmuch_string_list_append (tags, (char *) l->data); g_list_free (keys); - _notmuch_tags_prepare_iterator (tags); + _notmuch_string_list_sort (tags); - return tags; + return _notmuch_tags_create (thread, tags); } void