* fix searching for strings with spaces (e.g. maildir:"sent items")

This commit is contained in:
djcb 2011-11-24 00:12:29 +02:00
parent cc6738c195
commit c9f72ba3c7
5 changed files with 118 additions and 69 deletions

View File

@ -294,6 +294,35 @@ resolve_bookmark (MuConfig *opts, GError **err)
}
gchar*
str_quoted_from_strv (const gchar **params)
{
GString *str;
int i;
g_return_val_if_fail (params, NULL);
if (!params[0])
return g_strdup ("");
str = g_string_sized_new (64); /* just a guess */
for (i = 0; params[i]; ++i) {
if (i > 0)
g_string_append_c (str, ' ');
g_string_append_c (str, '"');
g_string_append (str, params[i]);
g_string_append_c (str, '"');
}
return g_string_free (str, FALSE);
}
static gchar*
get_query (MuConfig *opts, GError **err)
{
@ -313,7 +342,7 @@ get_query (MuConfig *opts, GError **err)
return NULL;
}
query = mu_util_str_from_strv ((const gchar**)&opts->params[1]);
query = str_quoted_from_strv ((const gchar**)&opts->params[1]);
if (bookmarkval) {
gchar *tmp;
tmp = g_strdup_printf ("%s %s", bookmarkval, query);

View File

@ -203,7 +203,8 @@ get_query (MuQuery *mqx, const char* searchexpr, GError **err)
Xapian::QueryParser::FLAG_PURE_NOT |
Xapian::QueryParser::FLAG_WILDCARD |
Xapian::QueryParser::FLAG_AUTO_SYNONYMS |
Xapian::QueryParser::FLAG_BOOLEAN_ANY_CASE);
Xapian::QueryParser::FLAG_BOOLEAN_ANY_CASE
);
g_free (preprocessed);
return query;
@ -275,22 +276,32 @@ mu_query_destroy (MuQuery *self)
}
/* preprocess a query to make them a bit more permissive */
/* preprocess a query to make them a bit more promiscuous */
char*
mu_query_preprocess (const char *query)
{
gchar *my_query;
GSList *parts, *cur;
gchar *myquery;
g_return_val_if_fail (query, NULL);
my_query = g_strdup (query);
/* remove accents and turn to lower-case */
mu_str_normalize_in_place (my_query, TRUE);
/* escape '@', single '_' and ':' if it's not following a
* xapian-pfx with '_' */
mu_str_ascii_xapian_escape_in_place (my_query);
/* convert the query to a list of query terms, and escape them
* separately */
parts = mu_str_esc_to_list (query);
return my_query;
for (cur = parts; cur; cur = g_slist_next(cur)) {
/* remove accents and turn to lower-case */
cur->data = mu_str_normalize_in_place ((gchar*)cur->data, TRUE);
/* escape '@', single '_' and ':' if it's not following a
* xapian-pfx with '_' */
cur->data = mu_str_ascii_xapian_escape_in_place
((gchar*)cur->data, TRUE /*escape spaces too*/);
}
myquery = mu_str_from_list (parts, ' ');
mu_str_free_list (parts);
return myquery;
}

View File

@ -432,7 +432,7 @@ mu_str_subject_normalize (const gchar* str)
* specially; function below is an ugly hack to make it DWIM in most
* cases...*/
char*
mu_str_ascii_xapian_escape_in_place (char *query)
mu_str_ascii_xapian_escape_in_place (char *query, gboolean esc_space)
{
gchar *cur;
const char escchar = '_';
@ -444,10 +444,14 @@ mu_str_ascii_xapian_escape_in_place (char *query)
*cur = tolower(*cur);
switch (*cur) {
case ' ':
if (!esc_space)
break;
case '@':
case '-':
case '/':
case '.': {
case '.':
{
/* don't replace a final special char */
if (cur[1]== ' ' || cur[1]=='\t' || cur[1]== '.')
++cur;
@ -472,11 +476,11 @@ mu_str_ascii_xapian_escape_in_place (char *query)
}
char*
mu_str_ascii_xapian_escape (const char *query)
mu_str_ascii_xapian_escape (const char *query, gboolean esc_space)
{
g_return_val_if_fail (query, NULL);
return mu_str_ascii_xapian_escape_in_place (g_strdup(query));
return mu_str_ascii_xapian_escape_in_place (g_strdup(query), esc_space);
}

View File

@ -1,22 +1,22 @@
/* -*-mode: c; tab-width: 8; indent-tabs-mode: t; c-basic-offset: 8 -*-*/
/*
/*
** Copyright (C) 2008-2010 Dirk-Jan C. Binnema <djcb@djcbsoftware.nl>
**
** This program is free software; you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 3 of the License, or
** (at your option) any later version.
**
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software Foundation,
** Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
**
** Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
**
*/
#ifndef __MU_STR_H__
@ -43,9 +43,9 @@ G_BEGIN_DECLS
* buffer (ie, non-reentrant), while mu_str_display_contact
* returns a newly allocated string that you must free with g_free
* when done with it.
*
*
* @param str a 'contact str' (ie., what is in the To/Cc/Bcc/From fields), or NULL
*
*
* @return a newly allocated string with a display contact
*/
const char* mu_str_display_contact_s (const char *str);
@ -56,13 +56,13 @@ char *mu_str_display_contact (const char *str);
* get a display size for a given size_t; uses M for sizes >
* 1000*1000, k for smaller sizes. Note: this function use the
* 10-based SI units, _not_ the powers-of-2 based ones.
*
*
* mu_str_size_s returns a ptr to a static buffer,
* while mu_str_size returns dynamically allocated
* memory that must be freed after use.
*
* @param t the size as an size_t
*
*
* @return a string representation of the size; see above
* for what to do with it
*/
@ -70,17 +70,17 @@ const char* mu_str_size_s (size_t s) G_GNUC_CONST;
char* mu_str_size (size_t s) G_GNUC_WARN_UNUSED_RESULT;
/**
* get a display string for a given set of flags, OR'ed in
* get a display string for a given set of flags, OR'ed in
* @param flags; one character per flag:
* D=draft,F=flagged,N=new,P=passed,R=replied,S=seen,T=trashed
* D=draft,F=flagged,N=new,P=passed,R=replied,S=seen,T=trashed
* a=has-attachment,s=signed, x=encrypted
*
*
* mu_str_file_flags_s returns a ptr to a static buffer,
* while mu_str_file_flags returns dynamically allocated
* memory that must be freed after use.
*
* @param flags file flags
*
*
* @return a string representation of the flags; see above
* for what to do with it
*/
@ -91,10 +91,10 @@ char* mu_str_flags (MuFlags flags)
/**
* get a 'summary' of the string, ie. the first /n/ lines of the
* strings, with all newlines removed, replaced by single spaces
*
*
* @param str the source string
* @param max_lines the maximum number of lines to include in the summary
*
*
* @return a newly allocated string with the summary. use g_free to free it.
*/
char* mu_str_summarize (const char* str, size_t max_lines)
@ -106,8 +106,8 @@ char* mu_str_summarize (const char* str, size_t max_lines)
* 'Latin-1 Supplement' and 'Latin Extended-A'
*
* @param str a valid utf8 string or NULL
* @param downcase if TRUE, convert the string to lowercase
*
* @param downcase if TRUE, convert the string to lowercase
*
* @return the normalize string, or NULL in case of error or str was NULL
*/
char* mu_str_normalize (const char *str, gboolean downcase)
@ -120,10 +120,10 @@ char* mu_str_normalize (const char *str, gboolean downcase)
* optionally, downcase it. this happen by changing the string; if
* that is not desired, use mu_str_normalize. Works for accented chars
* in Unicode Blocks 'Latin-1 Supplement' and 'Latin Extended-A'
*
*
* @param str a valid utf8 string or NULL
* @param downcase if TRUE, convert the string to lowercase
*
*
* @return the normalized string, or NULL in case of error or str was
* NULL
*/
@ -139,12 +139,13 @@ char* mu_str_normalize_in_place (char *str, gboolean downcase);
* case, the string will be downcased.
*
* works for ascii strings, like e-mail addresses and message-id.
*
*
* @param query a query string
*
* @param esc_space escape space characters as well
*
* @return the escaped string or NULL in case of error
*/
char* mu_str_ascii_xapian_escape_in_place (char *query);
char* mu_str_ascii_xapian_escape_in_place (char *query, gboolean esc_space);
/**
* escape the string for use with xapian matching. in practice, if the
@ -153,12 +154,13 @@ char* mu_str_ascii_xapian_escape_in_place (char *query);
* as 'subject:', 't:' etc, as defined in mu-msg-fields.[ch]).
*
* works for ascii strings, like e-mail addresses and message-id.
*
*
* @param query a query string
*
* @param esc_space escape space characters as well
*
* @return the escaped string (free with g_free) or NULL in case of error
*/
char* mu_str_ascii_xapian_escape (const char *query)
char* mu_str_ascii_xapian_escape (const char *query, gboolean esc_space)
G_GNUC_WARN_UNUSED_RESULT;
@ -174,9 +176,9 @@ char* mu_str_ascii_xapian_escape (const char *query)
* practical sizes for email messages are in terms of Mb; even in
* extreme cases it should be under 100 Mb. Function return
* GUINT64_MAX if there a parsing error
*
*
* @param str a string with a size, such a "100", "100Kb", "1Mb"
*
*
* @return the corresponding size in bytes, or -1 in case of error
*/
gint64 mu_str_size_parse_bkm (const char* str);
@ -184,10 +186,10 @@ gint64 mu_str_size_parse_bkm (const char* str);
/**
* create a full path from a path + a filename. function is _not_
* reentrant.
*
*
* @param path a path (!= NULL)
* @param name a name (may be NULL)
*
*
* @return the path as a statically allocated buffer. don't free.
*/
const char* mu_str_fullpath_s (const char* path, const char* name);
@ -196,10 +198,10 @@ const char* mu_str_fullpath_s (const char* path, const char* name);
/**
* escape a string like a string literal in C; ie. replace \ with \\,
* and " with \"
*
*
* @param str a non-NULL str
* @param in_quotes whether the result should be enclosed in ""
*
*
* @return the escaped string, newly allocated (free with g_free)
*/
char* mu_str_escape_c_literal (const gchar* str, gboolean in_quotes)
@ -210,9 +212,9 @@ char* mu_str_escape_c_literal (const gchar* str, gboolean in_quotes)
/**
* turn a string into plain ascii by replacing each non-ascii
* character with a dot ('.'). replacement is done in-place.
*
*
* @param buf a buffer to asciify
*
*
* @return the buf ptr (as to allow for function composition)
*/
char* mu_str_asciify_in_place (char *buf);
@ -220,10 +222,10 @@ char* mu_str_asciify_in_place (char *buf);
/**
* convert a string in a certain charset into utf8
*
*
* @param buffer a buffer to convert
* @param charset source character set.
*
*
* @return a UTF8 string (which you need to g_free when done with it),
* or NULL in case of error
*/
@ -234,9 +236,9 @@ gchar* mu_str_convert_to_utf8 (const char* buffer, const char *charset);
/**
* macro to check whether the string is empty, ie. if it's NULL or
* it's length is 0
*
*
* @param S a string
*
*
* @return TRUE if the string is empty, FALSE otherwise
*/
#define mu_str_is_empty(S) ((!(S)||!(*S))?TRUE:FALSE)
@ -244,10 +246,10 @@ gchar* mu_str_convert_to_utf8 (const char* buffer, const char *charset);
/**
* convert a GSList of strings to a #sepa-separated list
*
*
* @param lst a GSList
* @param the separator character
*
*
* @return a newly allocated string
*/
char* mu_str_from_list (const GSList *lst, char sepa);
@ -255,7 +257,7 @@ char* mu_str_from_list (const GSList *lst, char sepa);
/**
* convert a #sepa-separated list of strings in to a GSList
*
*
* @param str a #sepa-separated list of strings
* @param the separator character
* @param remove leading/trailing whitespace from the string
@ -270,9 +272,9 @@ GSList* mu_str_to_list (const char *str, char sepa, gboolean strip);
* separated by one or more spaces. list items can be quoted (using
* '"'), and '"', ' ' and '\' use their special meaning when prefixed
* with \.
*
*
* @param str a string
*
*
* @return a list of elements or NULL in case of error
*/
GSList* mu_str_esc_to_list (const char *str);
@ -280,7 +282,7 @@ GSList* mu_str_esc_to_list (const char *str);
/**
* free a GSList consisting of allocated strings
*
*
* @param lst a GSList
*/
void mu_str_free_list (GSList *lst);
@ -289,9 +291,9 @@ void mu_str_free_list (GSList *lst);
/**
* strip the subject of Re:, Fwd: etc.
*
*
* @param str a subject string
*
*
* @return a new string -- this is pointing somewhere inside the @str;
* no copy is made, don't free
*/
@ -303,9 +305,9 @@ const gchar* mu_str_subject_normalize (const gchar* str);
* first name, last name, the nick will be first name + the first char
* of the last name. otherwise, it's just the first name. clearly,
* this is just a rough guess for setting an initial value for nicks.
*
*
* @param name a name
*
*
* @return the guessed nick, as a newly allocated string (free with g_free)
*/
gchar* mu_str_guess_nick (const char* name)
@ -315,9 +317,9 @@ gchar* mu_str_guess_nick (const char* name)
/**
* guess the first name for the given name; clearly,
* this is just a rough guess for setting an initial value.
*
*
* @param name a name
*
*
* @return the first name, as a newly allocated string (free with
* g_free)
*/
@ -329,9 +331,9 @@ gchar* mu_str_guess_first_name (const char* name)
/**
* guess the last name for the given name; clearly,
* this is just a rough guess for setting an initial value.
*
*
* @param name a name
*
*
* @return the last name, as a newly allocated string (free with
* g_free)
*/

View File

@ -284,7 +284,7 @@ mu_util_str_from_strv (const gchar **params)
for (i = 0; params[i]; ++i) {
if (i>0)
if (i > 0)
g_string_append_c (str, ' ');
g_string_append (str, params[i]);
@ -293,6 +293,11 @@ mu_util_str_from_strv (const gchar **params)
return g_string_free (str, FALSE);
}
int
mu_util_create_writeable_fd (const char* path, mode_t mode,
gboolean overwrite)
@ -482,5 +487,3 @@ mu_util_printerr_encoded (const char *frm, ...)
return rv;
}