|
a/website/BUGS.html |
|
b/website/BUGS.html |
|
... |
|
... |
| 33 |
|
33 |
|
| 34 |
<p><i>Bugs that are listed in an older version section are
|
34 |
<p><i>Bugs that are listed in an older version section are
|
| 35 |
supposedly fixed in later versions. Bugs listed in the
|
35 |
supposedly fixed in later versions. Bugs listed in the
|
| 36 |
topmost section may also exist in older versions.</i></p>
|
36 |
topmost section may also exist in older versions.</i></p>
|
| 37 |
|
37 |
|
| 38 |
<h2><a name="b_latest">recoll 1.13.04 + xapian 1.0.16</a></h2>
|
38 |
<h2><a name="b_latest">recoll 1.14.1 + xapian 1.2+ + qt 4.4.1+</a></h2>
|
| 39 |
<ul>
|
39 |
<ul>
|
| 40 |
<li>Does not compile with Xapian
|
|
|
| 41 |
1.2. Apply <a href="files/xapian12.patch">patch</a>.</li>
|
|
|
| 42 |
<li>Mac OS X + Qt 4.6.1 : the index configuration dialog
|
|
|
| 43 |
crashes.</li>
|
|
|
| 44 |
<li>Performing a full index with release 1.11 or later, over a
|
40 |
<li>Performing a full index with release 1.11 or later, over a
|
| 45 |
version created with a much older recoll release may
|
41 |
version created with a much older recoll release may
|
| 46 |
sometimes end with an error saying "backend doesn't
|
42 |
sometimes end with an error saying "backend doesn't
|
| 47 |
implement metadata". If this happens, you need to delete
|
43 |
implement metadata". If this happens, you need to delete
|
| 48 |
the index directory (typically <em>~/.recoll/xapiandb/</em>)
|
44 |
the index directory (typically <em>~/.recoll/xapiandb/</em>)
|
| 49 |
and restart indexing. For big indexes, to avoid losing
|
45 |
and restart indexing. For big indexes, to avoid losing
|
| 50 |
time, removing the directory preventively may be preferable .</li>
|
46 |
time, removing the directory preventively may be preferable .</li>
|
| 51 |
|
47 |
|
| 52 |
<li> When Recoll is built with qt 4.4.0, the icons in the
|
|
|
| 53 |
result list are all displayed at the top of the page and
|
|
|
| 54 |
garbled. This appears to be a qt bug, fixed in 4.4.1. Use
|
|
|
| 55 |
either qt 4.3.x or 4.4.1</li>
|
|
|
| 56 |
|
|
|
| 57 |
<li> If the user-chosen result list entry format results in
|
48 |
<li> If the user-chosen result list entry format results in
|
| 58 |
several paragraphs (in the qt textedit sense), right clicks
|
49 |
several paragraphs (in the qt textedit sense), right clicks
|
| 59 |
will only work inside the first one for each entry.</li>
|
50 |
will only work inside the first one for each entry.</li>
|
| 60 |
|
|
|
| 61 |
<li> When a mime type has an external viewer defined, but the
|
|
|
| 62 |
actual file is compressed (ie: xxx.txt.gz), recoll will try
|
|
|
| 63 |
to start the external viewer on the compressed file, which
|
|
|
| 64 |
will not work in most cases.</li>
|
|
|
| 65 |
|
|
|
| 66 |
<li>It seems that the recoll program sometimes segfaults when
|
|
|
| 67 |
exiting after the first execution ?</li>
|
|
|
| 68 |
|
|
|
| 69 |
<li> If you are seeing a delay of a few seconds before the
|
|
|
| 70 |
result list displays for the first query of a recoll
|
|
|
| 71 |
instance, try changing the result list font in the query
|
|
|
| 72 |
preferences. This is not a recoll problem, I don't know the
|
|
|
| 73 |
exact cause (I've seen it happen with "Sans Serif" and go
|
|
|
| 74 |
away with Helvetica or Arial).</li>
|
|
|
| 75 |
|
|
|
| 76 |
<li> Under some versions of KDE (ie: Fedora FC5 KDE
|
|
|
| 77 |
3.5.4-0.5.fc5), there is a problem with the window stacking
|
|
|
| 78 |
order. Opening the "browse" file selection dialog from the
|
|
|
| 79 |
advanced search dialog will stack the latter under the main
|
|
|
| 80 |
window, possibly making it invisible. This is quite probably
|
|
|
| 81 |
a Kwin bug, possibly related to
|
|
|
| 82 |
http://bugs.kde.org/show_bug.cgi?id=79183 or a correction
|
|
|
| 83 |
thereof.</li>
|
|
|
| 84 |
|
51 |
|
| 85 |
<li>Under Solaris, it is necessary to perform initial
|
52 |
<li>Under Solaris, it is necessary to perform initial
|
| 86 |
indexing with the recollindex program (the recoll index
|
53 |
indexing with the recollindex program (the recoll index
|
| 87 |
thread doesn't work for creating the database). Don't know
|
54 |
thread doesn't work for creating the database). Don't know
|
| 88 |
the reason. Only idea I have is problem with exception
|
55 |
the reason. Only idea I have is problem with exception
|
|
... |
|
... |
| 97 |
I'm afraid the text stream of this file is too small to
|
64 |
I'm afraid the text stream of this file is too small to
|
| 98 |
handle.
|
65 |
handle.
|
| 99 |
</blockquote>
|
66 |
</blockquote>
|
| 100 |
Only small files produced by Microsoft Word on a Mac, or by
|
67 |
Only small files produced by Microsoft Word on a Mac, or by
|
| 101 |
OpenOffice will trigger this message. As a workaround, install
|
68 |
OpenOffice will trigger this message. As a workaround, install
|
| 102 |
wvWare and modify mimeconf to use the rcldoc filter, which
|
69 |
wvWare and modify mimeconf to use the rcldoc filter (instead of
|
|
|
70 |
directly executing antiword). Rcldoc will try antiword, then
|
| 103 |
will use vwWare if it is available. This will result in
|
71 |
will use vwWare if it is available. This will result in
|
| 104 |
slower indexing for doc files.</li>
|
72 |
slightly slower indexing for doc files.</li>
|
| 105 |
</ul>
|
73 |
</ul>
|
|
|
74 |
|
|
|
75 |
|
|
|
76 |
<h2><a name="b_1_13_04">recoll 1.13.04</h2>
|
|
|
77 |
<li><b>Note:</b> some of the bugs listed here are not actually
|
|
|
78 |
"fixed", mostly they were problems caused by old versions of
|
|
|
79 |
external software (ie: kde, qt), and I stopped carrying them. Just
|
|
|
80 |
don't use these versions, or live with the problem.</li>
|
|
|
81 |
|
|
|
82 |
<li>In case a new style filter (persistent) crashed while indexing,
|
|
|
83 |
it was not restarted, and all further files of the same mime type
|
|
|
84 |
were not updated (ie: python zip crash on encrypted files).</li>
|
|
|
85 |
|
|
|
86 |
<li>Mac OS X + Qt 4.6.1 : the index configuration dialog
|
|
|
87 |
crashes.</li>
|
|
|
88 |
|
|
|
89 |
<li>If you are seeing a delay of a few seconds before the
|
|
|
90 |
result list displays for the first query of a recoll
|
|
|
91 |
instance, try changing the result list font in the query
|
|
|
92 |
preferences. This is not a recoll problem, I don't know the
|
|
|
93 |
exact cause (I've seen it happen with "Sans Serif" and go
|
|
|
94 |
away with Helvetica or Arial).</li>
|
|
|
95 |
|
|
|
96 |
<li> When a mime type has an external viewer defined, but the
|
|
|
97 |
actual file is compressed (ie: xxx.txt.gz), recoll will try
|
|
|
98 |
to start the external viewer on the compressed file, which
|
|
|
99 |
will not work in most cases.</li>
|
|
|
100 |
|
|
|
101 |
<li>It seems that the recoll program sometimes segfaults when
|
|
|
102 |
exiting after the first execution ?</li>
|
|
|
103 |
|
|
|
104 |
<li>Does not compile with Xapian
|
|
|
105 |
1.2. Apply <a href="files/xapian12.patch">patch</a>.</li>
|
|
|
106 |
|
|
|
107 |
<li>When Recoll is built with qt 4.4.0, the icons in the
|
|
|
108 |
result list are all displayed at the top of the page and
|
|
|
109 |
garbled. This appears to be a qt bug, fixed in 4.4.1. Use
|
|
|
110 |
either qt 4.3.x or 4.4.1 (stopped carrying this bug. Just don't use
|
|
|
111 |
4.4.0)</li>
|
|
|
112 |
|
|
|
113 |
<li> Under some versions of KDE (ie: Fedora FC5 KDE
|
|
|
114 |
3.5.4-0.5.fc5), there is a problem with the window stacking
|
|
|
115 |
order. Opening the "browse" file selection dialog from the
|
|
|
116 |
advanced search dialog will stack the latter under the main
|
|
|
117 |
window, possibly making it invisible. This is quite probably
|
|
|
118 |
a Kwin bug, possibly related to
|
|
|
119 |
http://bugs.kde.org/show_bug.cgi?id=79183 or a correction
|
|
|
120 |
thereof.</li>
|
| 106 |
|
121 |
|
| 107 |
|
122 |
|
| 108 |
<h2><a name="b_1_13_02">recoll 1.13.02</h2>
|
123 |
<h2><a name="b_1_13_02">recoll 1.13.02</h2>
|
| 109 |
<li>Stemming does not work in the 1.13 series. The stemming
|
124 |
<li>Stemming does not work in the 1.13 series. The stemming
|
| 110 |
database was not created at all. Things would sort of work as
|
125 |
database was not created at all. Things would sort of work as
|