recoll issueshttps://framagit.org/medoc92/recoll/-/issues2022-04-07T17:21:35Zhttps://framagit.org/medoc92/recoll/-/issues/1segfault error 4 in libaspell.so.15.2.02022-04-07T17:21:35ZNathaniel Beaversegfault error 4 in libaspell.so.15.2.0I think it happened while typing, not after pressing enter, but I'm not sure. Got this in `dmesg`:
recoll[7116]: segfault at 7f6966aae7a5 ip 00007f69245a2434 sp 00007ffd499f9b48 error 4 in libaspell.so.15.2.0[7f6924559000+a1000]
Be...I think it happened while typing, not after pressing enter, but I'm not sure. Got this in `dmesg`:
recoll[7116]: segfault at 7f6966aae7a5 ip 00007f69245a2434 sp 00007ffd499f9b48 error 4 in libaspell.so.15.2.0[7f6924559000+a1000]
Best I can do is that it's somewhere near `Rcl::Db::getSpellingSuggestions` and `Aspell::suggest`.
#21 0x00007f69245e5a6c in acommon::aspell_speller_suggest(acommon::Speller*, char const*, int) (ths=0x55648fa089e0, word=<optimized out>, word_size=<optimized out>) at lib/speller-c.cpp:148
s0 = 7
ret = {<acommon::PosibErrBase> = {err_ = 0x7ffd499fa66f}, data = 0x112166d34d6eee00}
#22 0x00007f6998914049 in Aspell::suggest(Rcl::Db&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::list<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >&) () at /usr/lib/recoll/librecoll-1.23.7.so
#23 0x00007f69989ef52f in Rcl::Db::getSpellingSuggestions(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >&) () at /usr/lib/recoll/librecoll-1.23.7.so
#24 0x000055648e0e4df5 in ()
#25 0x00007f69989c3a73 in ResListPager::displayPage(RclConfig*) () at /usr/lib/recoll/librecoll-1.23.7.so
#26 0x000055648e0e1838 in ()
#27 0x000055648e0e235b in ()
#28 0x000055648e0e2fa9 in ()
#29 0x00007f6993e0b555 in QMetaObject::activate(QObject*, int, int, void**) (sender=0x7ffd499fc9e0, signalOffset=<optimized out>, local_signal_index=<optimized out>, argv=<optimized out>) at kernel/qobject.cpp:3767
I've attached the full stack traces in case they are helpful.
Can't get the librecoll function calls because there is no recoll-dbgsym package.
https://bugs.launchpad.net/ubuntu/+source/recoll/+bug/1814973
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=928157
I could also provide a link to the `aspdict.en.rws` file, but it's a bit big.
$ du -sh ~/.recoll/aspdict.en.rws
45M /home/nathaniel/.recoll/aspdict.en.rws
Anecdotally, this is the most frequent reason for recoll to crash on Ubuntu 18.04.
Might be a duplicate of this bug?
> The aspell command used to generate the orthographic correction dictionary is
> broken on Debian Jessie, because of an aspell packaging mistake which will
> not be fixed for the release. Try the following command, replacing 'en' by
> your language code:
>
> /usr/bin/aspell --lang=en --encoding=utf-8 create master /tmp/dict.rws
>
> If it complains about a missing /usr/share/aspell/en.dat, the workaround is
> to link the .dat files from /usr/lib/aspell to /usr/share/aspell:
>
> cd /usr/share/aspell
> sudo ln -s /usr/lib/aspell/*.dat .
https://www.lesbonscomptes.com/recoll/BUGS.html
or maybe this bug?
> Observed behaviour: segfault at fc65e0ad ip 00000000f6097188 sp
> 00000000ffd47348 error 4 in libaspell.so.15.1.4[f604f000+af000]
>
> [ . . . ]
>
> I’m also afraid that the crash may be linked to the actual contents of the
> aspell dict, aspell is a quite sensitive beast.
>
> [ . . . ]
>
> This probably an aspell issue, and I can’t work on it for lack of feedback
> anyway
https://www.lesbonscomptes.com/recoll/bitbucket-issues-recoll/issue-47.html
$ apt-cache policy recoll
recoll:
Installed: 1.23.7-1
Candidate: 1.23.7-1
Version table:
*** 1.23.7-1 500
500 http://us.archive.ubuntu.com/ubuntu bionic/universe amd64 Packages
100 /var/lib/dpkg/status
$ apt-cache policy libaspell15
libaspell15:
Installed: 0.60.7~20110707-4ubuntu0.1
Candidate: 0.60.7~20110707-4ubuntu0.1
Version table:
*** 0.60.7~20110707-4ubuntu0.1 500
500 http://us.archive.ubuntu.com/ubuntu bionic-updates/main amd64 Packages
500 http://security.ubuntu.com/ubuntu bionic-security/main amd64 Packages
100 /var/lib/dpkg/status
0.60.7~20110707-4 500
500 http://us.archive.ubuntu.com/ubuntu bionic/main amd64 Packages
recoll 1.23.7-1
python 2.7.15~rc1-1
python3 3.6.7-1~18.04
libc6 2.27-3ubuntu1
libgcc1 1:8.3.0-26ubuntu1~18.04
libqt5core5a 5.9.5+dfsg-0ubuntu2.5
libqt5gui5 5.9.5+dfsg-0ubuntu2.5
libqt5printsupport5 5.9.5+dfsg-0ubuntu2.5
libqt5webkit5 5.212.0~alpha2-7ubuntu1
libqt5widgets5 5.9.5+dfsg-0ubuntu2.5
libqt5xml5 5.9.5+dfsg-0ubuntu2.5
libstdc++6 8.3.0-26ubuntu1~18.04
libx11-6 2:1.6.4-3ubuntu0.2
libxapian30 1.4.5-1ubuntu0.1
zlib1g 1:1.2.11.dfsg-0ubuntu2
aspell 0.60.7~20110707-4ubuntu0.1
xdg-utils 1.1.2-1ubuntu2.3
xsltproc 1.1.29-5ubuntu0.2
antiword 0.37-11build1
ghostscript 9.26~dfsg+0-0ubuntu0.18.04.12
libimage-exiftool-perl 10.80-1
poppler-utils 0.62.0-2ubuntu2.10
python-libxml2 2.9.4+dfsg1-6.1ubuntu1.3
python-lzma 0.5.3-3
unrtf 0.21.9-clean-3
untex 1:1.2-6
[Stacktrace](/uploads/c7f2214052e4bdff177df0e6d4c861cd/Stacktrace)
[ThreadStacktrace](/uploads/5b5c3685d7ec9d9ca07a0aae3f98cd3c/ThreadStacktrace)https://framagit.org/medoc92/recoll/-/issues/2Compiling on raspberry pi2022-07-15T01:07:39ZdebjanCompiling on raspberry piHi,
following the docs I cloned the repo, installed requirements and compiled with:
```
./configure --disable-webkit --disable-qtgui --disable-x11mon --prefix=/home/pi/.local
make
make install
```
However as a result I just got `recol...Hi,
following the docs I cloned the repo, installed requirements and compiled with:
```
./configure --disable-webkit --disable-qtgui --disable-x11mon --prefix=/home/pi/.local
make
make install
```
However as a result I just got `recollindex` executable, which seem to work fine after indexing my PDF directory, but there is no `recoll` or `recollq` executable.
I'm not sure why this can be, and how to query generated database.
Please advise.
Thanks
*PS: I posted to recoll-user, where I found the link to this repo*https://framagit.org/medoc92/recoll/-/issues/3Upgrade to 1.26.6 doesn't migrate .conf to .ini2020-04-03T13:51:20Zhameau1Upgrade to 1.26.6 doesn't migrate .conf to .iniHello,
- Previous version : 1.26.3-1~ppa1~buster
- Upgrade to : 1.26.6-1~ppa1~buster
- Debian 10 (buster) xfce
Upgrading from 1.26.3 to 1.26.6 (both from the Recoll ppa) ignores the existing recoll.conf and writes a new recoll.ini with...Hello,
- Previous version : 1.26.3-1~ppa1~buster
- Upgrade to : 1.26.6-1~ppa1~buster
- Debian 10 (buster) xfce
Upgrading from 1.26.3 to 1.26.6 (both from the Recoll ppa) ignores the existing recoll.conf and writes a new recoll.ini with default settings.
Renaming `~/.config/Recoll.org/recoll.conf` to `recoll.ini` restores the user settings, but it isn't immediately obvious. The Release notes for 1.26.6 mentions that the Windows settings are moved from the registry to a .ini file, but there is nothing about this affecting Linux installations.https://framagit.org/medoc92/recoll/-/issues/4Ctrl-Backspace does not delete word (version 1.26.5)2020-05-05T15:01:19ZNathaniel BeaverCtrl-Backspace does not delete word (version 1.26.5)Steps to reproduce:
1. Open recoll Qt GUI.
2. Click into the single entry field of simple search.
3. Type "word1 word2"
4. Type Ctrl-Backspace
Expected behavior: "word2" is deleted, leaving just "word1 "
Actual behavior: only "2" i...Steps to reproduce:
1. Open recoll Qt GUI.
2. Click into the single entry field of simple search.
3. Type "word1 word2"
4. Type Ctrl-Backspace
Expected behavior: "word2" is deleted, leaving just "word1 "
Actual behavior: only "2" is deleted, leaving "word1 word"
Disclaimer: I haven't tried to reproduce this on another machine, so it could be a quirk of my configuration.
I just noticed this in version 1.26.5, since the behavior is different from 1.23.7.
Just a guess, but maybe this commit is related?
> GUI simple search: always disable the popup after backspace and delete
https://framagit.org/medoc92/recoll/-/commit/931ed5dd7e63c66572aa240cb15ce75bc1927908
It seemed the most likely from commits since version 1.23.7.
https://framagit.org/medoc92/recoll/-/commits/master/src/qtgui/ssearch_w.cpp
Version information:
$ apt-cache policy recoll
recoll:
Installed: 1.26.5-1~ppa4~bionic1
Candidate: 1.26.5-1~ppa4~bionic1
Version table:
*** 1.26.5-1~ppa4~bionic1 500
500 http://ppa.launchpad.net/recoll-backports/exp/ubuntu bionic/main amd64 Packages
500 http://ppa.launchpad.net/recoll-backports/exp/ubuntu bionic/main i386 Packages
100 /var/lib/dpkg/status
1.23.7-1 500
500 http://us.archive.ubuntu.com/ubuntu bionic/universe amd64 Packageshttps://framagit.org/medoc92/recoll/-/issues/5Degraded indexing and seach performance on Xapian 1.4/glass compared to /chert2021-11-14T09:18:56ZFrank de LangeDegraded indexing and seach performance on Xapian 1.4/glass compared to /chertI recently rebuilt the databases on my document server and with that migrated from the *chert* to the *glass* format. The document server is based on a 6-core, 16GB container running on a 12-core DL380G7/128GB, the collection is stored o...I recently rebuilt the databases on my document server and with that migrated from the *chert* to the *glass* format. The document server is based on a 6-core, 16GB container running on a 12-core DL380G7/128GB, the collection is stored on a 7-drive array hosted on a DS4243 JBOD enclosure, in other words relatively up-to-date and by no means sluggish hardware. The original database was created using Recoll 1.19 and had been updated without being reset ever since.
Indexing the collection (420.000 files (mixture of PDF and EPUB) spanning 650GB) took almost 70 hours, far longer than expected. During indexing the container slowed down to a crawl but that was expected given the number of reports about `recollindex` hogging system resources.
A far bigger - and for me critical - problem is that search performance on the new database is extremely poor compared to the 'chert' format. The database size for all collections together (divided over 3 sets) is about 90GB using the *glass* format, ~75GB using *chert*. Searching for a single word in the largest of the sets (48GB database) using the *glass* format takes about 60 seconds, the same search using the *chert* format takes less than 2 seconds, in both cases starting with a cold cache. I tested using the python API through the (python3 version of the) web interface and through the `recollq` command line program (the document server is headless). The index was generated using mostly default options:
```
idxrundir = tmp
textfilemaxmbs = 200
nocjk = 1
idxflushmb = 200
```
I tested both with a freshly indexed database and after compressing it using `xapian-compact`, the results were similar.
I re-indexed the database after adding `indexStoreDocText = 0` to `recoll.conf` and moving `xapiandb` out of the way to get back to a *chert* database, this resolved the performance issues for now. Since *chert* support will be dropped in Xapian 1.5 this is not a permanent solution so it would be good to find a way to get back the lost performance using the *glass* format.
Tests performed on a DL380G7 with 2xX5675@3.07GHz, 128GB, system on the internal 8-drive array, data on external 7-drive (15K spinning rust) array in a DS4243 JBOD.
* Debian `10`
* Recoll `1.26.7-1~ppa1~buster`
* Xapian `1.4.11-1`https://framagit.org/medoc92/recoll/-/issues/6Complie on macos 10.15 due to Bison version & other problems2021-10-07T12:33:13ZGhost UserComplie on macos 10.15 due to Bison version & other problemsI am already successfully compile Recoll with the source pulled directly from the git with newest XCode and install all the libraries with Homebrew.
"Tweaks" used are listed below:
1. `./configure --disable-userdoc --disable-python-chm...I am already successfully compile Recoll with the source pulled directly from the git with newest XCode and install all the libraries with Homebrew.
"Tweaks" used are listed below:
1. `./configure --disable-userdoc --disable-python-chm --disable-webkit`
2. The Bison included within XCode is 2.3.X, and brewed version is [Key-only] which means it does not create any link into the path. I add the brewed bison 3.X.X to PATH by hand.
A Syntax error will arise at `/src/query/wasaparse.yp` -> `%destructor {delete $$;} <str>`
<https://framagit.org/medoc92/recoll/-/blob/master/src/query/wasaparse.ypp#L52>
I am not sure it is a version problem or a real syntax problem, but it works on bison 3.X.X. I never use bison before anyway,
3. OS X use a different linker and an error will rise `ld: unknown option: --no-undefined` (same as this [stackoverflow question](<https://stackoverflow.com/questions/26971333/what-is-clangs-equivalent-to-no-undefined-gcc-flag>)). I delete that option directly, I don't have enough time to investigate now.
4. `#include <malloc.h>` somewhere in the codebase not working, it have to be `#include <malloc/malloc.h>` on mac or `-I/usr/include/malloc/` instead.
======================
I know there is a Homebrew formula, will try later. Sorry, this is not a bug-report.
I want to push Recoll into the de facto Homebrew. I also notice there was a failed attempt to do this many years ago. [New formula for recoll](https://github.com/Homebrew/legacy-homebrew/pull/23125)
![Screen_Shot_2020-05-19_at_4.49.06_AM](/uploads/ab38a168a02508be3c8f8afa01a50feb/Screen_Shot_2020-05-19_at_4.49.06_AM.png)https://framagit.org/medoc92/recoll/-/issues/8Show warning when opening temporary file flag not working2020-05-28T08:00:38ZArthur HoffmannShow warning when opening temporary file flag not workingHi,
Every time I open a result from the list a warning pops up telling me that "Opening a temporary copy. Edits will be lost if you don't save them to a permanent location." Below that is an option to disable the warning next time.
Tic...Hi,
Every time I open a result from the list a warning pops up telling me that "Opening a temporary copy. Edits will be lost if you don't save them to a permanent location." Below that is an option to disable the warning next time.
Ticking this box does not disable the warning next time.
Also going to the GUI configuration in Preferences and unticking the 'Show warning when opening temporary file.' does not prevent the warning.
Recoll is a fantastic program and I enjoy using it. This issue is not a big deal, but I thought you may want to know about it anyway.
Thanks for this great tool.https://framagit.org/medoc92/recoll/-/issues/10recollindex -r does not handle relative paths2020-07-28T13:37:46ZNathaniel Beaverrecollindex -r does not handle relative pathsIf I invoke `recollindex` like this:
find out/ -print | recollindex -i
to index the `out/` directory, it works fine.
But if I invoke `recollindex` like this:
recollindex -r out/
or like this:
recollindex -r ./out/
it d...If I invoke `recollindex` like this:
find out/ -print | recollindex -i
to index the `out/` directory, it works fine.
But if I invoke `recollindex` like this:
recollindex -r out/
or like this:
recollindex -r ./out/
it does not update the index. However, if I instead run
recollindex -r $(realpath out/)
then the index is updated as expected. It appears that the `-r` flag requires an absolute path.
I'm not sure if this is intended behavior or not; from the `strace` output, it appears to look in `/tmp/out` rather than the current directory. If it is intended behavior, perhaps an update to the documentation or a warning message would be helpful, as I found it somewhat surprising.
Minimal test scripts and version information are attached.
[policy.txt](/uploads/0a7c0bcd7b5ca249daeafb6bd3fad32e/policy.txt)
[recollindex_-i.sh](/uploads/3c00ecf1bc70c581b8c09401eba559fe/recollindex_-i.sh)
[recollindex_-r.sh](/uploads/08057fc1055e47bb4d96c775b7e20a34/recollindex_-r.sh)
[recollindex_-r_realpath.sh](/uploads/1f88e4a20f0f62fef5d8cff080cb4194/recollindex_-r_realpath.sh)https://framagit.org/medoc92/recoll/-/issues/11feature request for recollindex: take a list of files and print the ones that...2021-04-01T16:56:58ZNathaniel Beaverfeature request for recollindex: take a list of files and print the ones that aren't already indexedSometimes I want to see what files in a given directory are in recoll's index and which ones aren't. Since recollindex does not have this built-in (as far as I know), I generally do something like this as a workaround:
xapian-delve...Sometimes I want to see what files in a given directory are in recoll's index and which ones aren't. Since recollindex does not have this built-in (as far as I know), I generally do something like this as a workaround:
xapian-delve -1 -A Q"$(realpath /path/to/directory/)" ~/.recoll/xapiandb/
However, there are some issues with this approach:
- There is no way to list files that *aren't* indexed, which is often more interesting since I may notice filetypes I want to be indexed but haven't added yet.
- The output is a little idiosyncratic and for long filenames is different from the true filename. This means it cannot be used as input to other programs that take a list of paths.
- This shows what files are currently in the index, but not which would be added if recollindex were to run.
This is a tentative feature request for `recollindex` to take a list of paths from stdin and output only those that meet particular criteria, such as:
- `indexed`: those that are already in the index
- `not-indexed`: those that are not already in the index
- `index-candidates`: those that `recollindex` would index if it were run
- `not-index-candidates`: those that `recollindex` would ignore if it were run
The usage I have in mind might look something like this:
find . -type f | recollindex -o not-indexed | less
The `-o` is a mnemonic for "output" or "only" but of course could be something else.https://framagit.org/medoc92/recoll/-/issues/12Web extension sources2020-07-24T17:31:10ZrhnWeb extension sourcesWhere can the web extension sources and issue tracker be found? I've only found a link to https://opensourceprojects.eu/, but that one doesn't work.Where can the web extension sources and issue tracker be found? I've only found a link to https://opensourceprojects.eu/, but that one doesn't work.https://framagit.org/medoc92/recoll/-/issues/13Issue with aspell dictionary generation at end of indexing2022-01-05T07:48:24ZAndréIssue with aspell dictionary generation at end of indexingHi,
While Recoll worked perfect in ubuntu 18.04 it did not in 20.04, I get this error message:
"Non-fatal indexing message:
aspell : aspell dictionary creation command failed: /usr/bin/aspell --lang=english --encoding=utf-8 create m...Hi,
While Recoll worked perfect in ubuntu 18.04 it did not in 20.04, I get this error message:
"Non-fatal indexing message:
aspell : aspell dictionary creation command failed: /usr/bin/aspell --lang=english --encoding=utf-8 create master /home/andre/.recoll/aspdict.english.rws One possible reason might be missing language data files for lang = english. Maybe try to execute the command by hand for a better diag."
I should add that I have tried to install aspell english: "sudo apt-get install aspell-en" but I already had that: "aspell-en is already the newest version (2018.04.16-0-1).
aspell-en är satt till manuellt installerad.
0 att uppgradera, 0 att nyinstallera, 0 att ta bort och 2 att inte uppgradera.
"
Any help is appreciated.https://framagit.org/medoc92/recoll/-/issues/14Recollindex activated by permission changes2023-05-02T09:25:42ZGhost UserRecollindex activated by permission changesHi, first of all, merci d'avoir écrit ce logiciel, je m'en sers très souvent !
I did a global permission change in my $HOME (chmod -R go-rwx $HOME) and then recollindex started to reindex everything again, with lots of python3, pdftotex...Hi, first of all, merci d'avoir écrit ce logiciel, je m'en sers très souvent !
I did a global permission change in my $HOME (chmod -R go-rwx $HOME) and then recollindex started to reindex everything again, with lots of python3, pdftotext, and other processes started, for hours.
The files' contents did not change, only their permissions. Thus recollindex did not need to reindex them. Surely theses changes were reported by the kernel monitoring (inotify, I believe), but recollindex could check whether the change was in files' contents or attributes before reindexing them.
I use Recoll 1.27.3-1~ppa2~bionic1 with realtime monitoring (-m option) in a Ubuntu 18 64bit box.
Best regards!https://framagit.org/medoc92/recoll/-/issues/15Indexing process stopped/crashed2020-08-18T06:51:00ZdarkshveinIndexing process stopped/crashedHello.
please, look my error log.
If problem solved at next version, please tell me, I will tried to compile under ubuntu.
I tried to setup maximum memory settings, for avoid limits errors.
thank you.
recoll 1.26.3, xapian 1.4.14, with...Hello.
please, look my error log.
If problem solved at next version, please tell me, I will tried to compile under ubuntu.
I tried to setup maximum memory settings, for avoid limits errors.
thank you.
recoll 1.26.3, xapian 1.4.14, with installed recommendations.
![common_params](/uploads/f913378a8d6ea0356d7a14bd378ad7ef/common_params.png)
![params2](/uploads/36b86c9ec158a012bb0dc28c4f58c374/params2.png)
[log_recoll_indexer_6](/uploads/abb4df78f7b357be82570862deb11727/log_recoll_indexer_6)https://framagit.org/medoc92/recoll/-/issues/16nonumbers issue2020-08-22T15:22:41ZPhilippe Hnonumbers issueHello bonjour,
I am a new recoll user (1.27.2) on windows 10 and it really looks good.
I cannot get nonumbers to work (tons of numbers appear in keywords completion, and they are found instantaneously in files).
I have a lot of scient...Hello bonjour,
I am a new recoll user (1.27.2) on windows 10 and it really looks good.
I cannot get nonumbers to work (tons of numbers appear in keywords completion, and they are found instantaneously in files).
I have a lot of scientific data in text files and a partial index of my drive is already 20gb (not finished).
I have reduced my conf file to the minimum just to check:
topdirs = C:/Desktop/singledir
nonumbers = 1
Maybe I am doing something obviously wrong. Any advice or things to try?
This is happening on both my desktop and laptop. I have not installed custom index filters, just added tesseract for scanned pdf (but not in the config above). Dictionnary does not seem to be working. No other error message.
Thanks!
Philippehttps://framagit.org/medoc92/recoll/-/issues/17Command line search does not return page number2020-09-01T07:43:02ZOli SturmCommand line search does not return page numberFor automation purposes, I'd like to run `recoll -t` or `recollq` from the command line, using `-F "filename abstract"` to generate a parseable result list. My indexed content consists exclusively of PDF files, but as far as I can see th...For automation purposes, I'd like to run `recoll -t` or `recollq` from the command line, using `-F "filename abstract"` to generate a parseable result list. My indexed content consists exclusively of PDF files, but as far as I can see there is no way of including the page number of results in the output. Am I missing something?
Many thanks for this great software!https://framagit.org/medoc92/recoll/-/issues/18Idea / Feature Request - Index content of PDF Markups (sticky notes, highligh...2023-04-24T21:36:36ZDavid RomeroIdea / Feature Request - Index content of PDF Markups (sticky notes, highlights and text boxes)This capability would make Recoll extremely helpful for research workflows. The ability to quickly search and find not only text snippets inside the PDF, but also my own comments and markups, would make organizing a research library a br...This capability would make Recoll extremely helpful for research workflows. The ability to quickly search and find not only text snippets inside the PDF, but also my own comments and markups, would make organizing a research library a breeze.
Anyone else thinks this would be cool?https://framagit.org/medoc92/recoll/-/issues/19Build error2020-09-01T17:30:51ZrezsoBuild errorWhen I try to compile 1.27.5 release, I get this error:
`reslist.cpp: In member function ‘virtual std::pair<int, int> ResList::parnumfromdocnum(int)’:
reslist.cpp:575:59: error: no matching function for call to ‘begin(std::_Rb_tree_iter...When I try to compile 1.27.5 release, I get this error:
`reslist.cpp: In member function ‘virtual std::pair<int, int> ResList::parnumfromdocnum(int)’:
reslist.cpp:575:59: error: no matching function for call to ‘begin(std::_Rb_tree_iterator<std::pair<const int, int> >&)’
575 | for (const auto& entry : m_pageParaToReldocnums.begin()) {
| ^
In file included from /usr/include/c++/10.2.0/bits/stl_map.h:62,
from /usr/include/c++/10.2.0/map:61,
from reslist.cpp:23:
/usr/include/c++/10.2.0/initializer_list:90:5: note: candidate: ‘template<class _Tp> constexpr const _Tp* std::begin(std::initializer_list<_Tp>)’
90 | begin(initializer_list<_Tp> __ils) noexcept
| ^~~~~
/usr/include/c++/10.2.0/initializer_list:90:5: note: template argument deduction/substitution failed:
reslist.cpp:575:59: note: ‘std::_Rb_tree_iterator<std::pair<const int, int> >’ is not derived from ‘std::initializer_list<_Tp>’
575 | for (const auto& entry : m_pageParaToReldocnums.begin()) {
| ^
In file included from /usr/include/c++/10.2.0/array:41,
from /usr/include/c++/10.2.0/tuple:39,
from /usr/include/c++/10.2.0/bits/stl_map.h:63,
from /usr/include/c++/10.2.0/map:61,
from reslist.cpp:23:
/usr/include/c++/10.2.0/bits/range_access.h:51:5: note: candidate: ‘template<class _Container> decltype (__cont.begin()) std::begin(_Container&)’
51 | begin(_Container& __cont) -> decltype(__cont.begin())
| ^~~~~
/usr/include/c++/10.2.0/bits/range_access.h:51:5: note: template argument deduction/substitution failed:
/usr/include/c++/10.2.0/bits/range_access.h: In substitution of ‘template<class _Container> decltype (__cont.begin()) std::begin(_Container&) [with _Container = std::_Rb_tree_iterator<std::pair<const int, int> >]’:
reslist.cpp:575:59: required from here
/usr/include/c++/10.2.0/bits/range_access.h:51:50: error: ‘struct std::_Rb_tree_iterator<std::pair<const int, int> >’ has no member named ‘begin’
51 | begin(_Container& __cont) -> decltype(__cont.begin())
| ~~~~~~~^~~~~
/usr/include/c++/10.2.0/bits/range_access.h:61:5: note: candidate: ‘template<class _Container> decltype (__cont.begin()) std::begin(const _Container&)’
61 | begin(const _Container& __cont) -> decltype(__cont.begin())
| ^~~~~
/usr/include/c++/10.2.0/bits/range_access.h:61:5: note: template argument deduction/substitution failed:
/usr/include/c++/10.2.0/bits/range_access.h: In substitution of ‘template<class _Container> decltype (__cont.begin()) std::begin(const _Container&) [with _Container = std::_Rb_tree_iterator<std::pair<const int, int> >]’:
reslist.cpp:575:59: required from here
/usr/include/c++/10.2.0/bits/range_access.h:61:56: error: ‘const struct std::_Rb_tree_iterator<std::pair<const int, int> >’ has no member named ‘begin’
61 | begin(const _Container& __cont) -> decltype(__cont.begin())
| ~~~~~~~^~~~~
/usr/include/c++/10.2.0/bits/range_access.h:90:5: note: candidate: ‘template<class _Tp, long unsigned int _Nm> _Tp* std::begin(_Tp (&)[_Nm])’
90 | begin(_Tp (&__arr)[_Nm])
| ^~~~~
/usr/include/c++/10.2.0/bits/range_access.h:90:5: note: template argument deduction/substitution failed:
reslist.cpp:575:59: note: mismatched types ‘_Tp [_Nm]’ and ‘std::_Rb_tree_iterator<std::pair<const int, int> >’
575 | for (const auto& entry : m_pageParaToReldocnums.begin()) {
| ^
reslist.cpp:575:59: error: no matching function for call to ‘end(std::_Rb_tree_iterator<std::pair<const int, int> >&)’
In file included from /usr/include/c++/10.2.0/bits/stl_map.h:62,
from /usr/include/c++/10.2.0/map:61,
from reslist.cpp:23:
/usr/include/c++/10.2.0/initializer_list:101:5: note: candidate: ‘template<class _Tp> constexpr const _Tp* std::end(std::initializer_list<_Tp>)’
101 | end(initializer_list<_Tp> __ils) noexcept
| ^~~
/usr/include/c++/10.2.0/initializer_list:101:5: note: template argument deduction/substitution failed:
reslist.cpp:575:59: note: ‘std::_Rb_tree_iterator<std::pair<const int, int> >’ is not derived from ‘std::initializer_list<_Tp>’
575 | for (const auto& entry : m_pageParaToReldocnums.begin()) {
| ^
In file included from /usr/include/c++/10.2.0/array:41,
from /usr/include/c++/10.2.0/tuple:39,
from /usr/include/c++/10.2.0/bits/stl_map.h:63,
from /usr/include/c++/10.2.0/map:61,
from reslist.cpp:23:
/usr/include/c++/10.2.0/bits/range_access.h:71:5: note: candidate: ‘template<class _Container> decltype (__cont.end()) std::end(_Container&)’
71 | end(_Container& __cont) -> decltype(__cont.end())
| ^~~
/usr/include/c++/10.2.0/bits/range_access.h:71:5: note: template argument deduction/substitution failed:
/usr/include/c++/10.2.0/bits/range_access.h: In substitution of ‘template<class _Container> decltype (__cont.end()) std::end(_Container&) [with _Container = std::_Rb_tree_iterator<std::pair<const int, int> >]’:
reslist.cpp:575:59: required from here
/usr/include/c++/10.2.0/bits/range_access.h:71:48: error: ‘struct std::_Rb_tree_iterator<std::pair<const int, int> >’ has no member named ‘end’
71 | end(_Container& __cont) -> decltype(__cont.end())
| ~~~~~~~^~~
/usr/include/c++/10.2.0/bits/range_access.h:81:5: note: candidate: ‘template<class _Container> decltype (__cont.end()) std::end(const _Container&)’
81 | end(const _Container& __cont) -> decltype(__cont.end())
| ^~~
/usr/include/c++/10.2.0/bits/range_access.h:81:5: note: template argument deduction/substitution failed:
/usr/include/c++/10.2.0/bits/range_access.h: In substitution of ‘template<class _Container> decltype (__cont.end()) std::end(const _Container&) [with _Container = std::_Rb_tree_iterator<std::pair<const int, int> >]’:
reslist.cpp:575:59: required from here
/usr/include/c++/10.2.0/bits/range_access.h:81:54: error: ‘const struct std::_Rb_tree_iterator<std::pair<const int, int> >’ has no member named ‘end’
81 | end(const _Container& __cont) -> decltype(__cont.end())
| ~~~~~~~^~~
/usr/include/c++/10.2.0/bits/range_access.h:100:5: note: candidate: ‘template<class _Tp, long unsigned int _Nm> _Tp* std::end(_Tp (&)[_Nm])’
100 | end(_Tp (&__arr)[_Nm])
| ^~~
/usr/include/c++/10.2.0/bits/range_access.h:100:5: note: template argument deduction/substitution failed:
reslist.cpp:575:59: note: mismatched types ‘_Tp [_Nm]’ and ‘std::_Rb_tree_iterator<std::pair<const int, int> >’
575 | for (const auto& entry : m_pageParaToReldocnums.begin()) {
| ^
make[3]: *** [Makefile:7505: .obj/reslist.o] Error 1`
How to fix this?https://framagit.org/medoc92/recoll/-/issues/20Where do I find these 3 'missing helpers'?2020-09-13T12:36:15ZHenrik RosenøWhere do I find these 3 'missing helpers'?Hi!
I like Recoll very much. I have used it under Ubuntu 18.04 for over 3 years. Now I have created a new partition with Ubuntu 20.04. I stopped all indexing in 18.04, copied the 3 configuration folders from the 18.04 partition to the 2...Hi!
I like Recoll very much. I have used it under Ubuntu 18.04 for over 3 years. Now I have created a new partition with Ubuntu 20.04. I stopped all indexing in 18.04, copied the 3 configuration folders from the 18.04 partition to the 20.04 partition. Then started Recoll (in 20.04) and chose "Rebuild index" (from scratch). The indexing finished quite fast, and everything seems to work, but there are 3 missing helpers:
python3:pylzma (application/x-7z-compressed)
untex (text/x-tex)
wpd2html (application/vnd.wordperfect)
How do I install them (in Ubuntu 20.04.1 LTS)?
Synaptic cannot find them. And I can't find any information on the recoll website either using the following Google search: 'site:lesbonscomptes.com python3 pylzma'.
Any help would be appreciated. Thank you.https://framagit.org/medoc92/recoll/-/issues/21Thanks!2020-09-14T14:25:54ZEgils ConsultingThanks!I really like the simplicity of recoll. I can set up the indexing completely according to my taste :)I really like the simplicity of recoll. I can set up the indexing completely according to my taste :)https://framagit.org/medoc92/recoll/-/issues/22Questions regarding MD5, underscore and dictionary2020-09-17T07:34:42ZPhilippe HQuestions regarding MD5, underscore and dictionaryHello,
I have three questions which are not bugs but I could not find the info in the documentation (perhaps missed it).
* MD5 seem to be computed (to detect duplicates etc). Is there a way to get the list of duplicates or export the w...Hello,
I have three questions which are not bugs but I could not find the info in the documentation (perhaps missed it).
* MD5 seem to be computed (to detect duplicates etc). Is there a way to get the list of duplicates or export the whole list of MD5? (that could help if using recoll to manage archives and not need to go again through the drive to compute again MD5)
* I often use "_" to name things. ex: test_01. But the "_" is dropped (and the 01 not indexed with nonumber). Is there a way to indicate in the settings that "_" should not be used to cut words?
* I have dictionnary errors at the end of indexing. Any page describing how to add dictionaries on windows?
Thanks again to Jean-François for this software!
Philippe
PS: fixing the nonumber really made a difference for me. It works great now with an index <2Gb on an index I previously had to kill it at over 15Gb. Searching in over 300,000 documents (in two indexes) is fast. Indexing was a bit slow due to over 2500 pdf to be OCRed (used tesseract, worked fine but took several days. It is great the cache is kept if reindexing by the way).https://framagit.org/medoc92/recoll/-/issues/23Recoll crashes when closing using "-c" in windows (not a show stopper)2020-09-17T07:32:31ZPhilippe HRecoll crashes when closing using "-c" in windows (not a show stopper)Hello,
When running recoll in windows 10 on a specific index using the -c option, a "recoll stopped working" (recoll a cessé de fonctionner in french) window pops up when closing the software.
This does not affect at all the behavior w...Hello,
When running recoll in windows 10 on a specific index using the -c option, a "recoll stopped working" (recoll a cessé de fonctionner in french) window pops up when closing the software.
This does not affect at all the behavior when the program is running (so the issue is really really minor).
It does that whether using a shortcut, typing it in a command line, using relative or full path to the index.
Any idea? Again no problem for me as I use the shortcut only for updates of the index.
Thanks again.
Philippehttps://framagit.org/medoc92/recoll/-/issues/24Recoll cannot finish indexing2020-10-30T15:25:31ZNick222Recoll cannot finish indexingRecoll cannot finish indexing
After update old index cannot read.
Index config - old - OK.
Try to indexing - and get errors
Recoll 1.27.7
Xubuntu 20.04
Log-file (recoll_log.zip) and problem file (2005-11.zip) attached
***
And some ...Recoll cannot finish indexing
After update old index cannot read.
Index config - old - OK.
Try to indexing - and get errors
Recoll 1.27.7
Xubuntu 20.04
Log-file (recoll_log.zip) and problem file (2005-11.zip) attached
***
And some files which get "recoll UTF-8 errors" (recoll_errors.zip)
***
[2015-11.zip](/uploads/0de8448a03674247317b9f69facbdd6d/2015-11.zip)
[recoll_errors.zip](/uploads/4e70a9220afe2075940a92d7697237fc/recoll_errors.zip)
[recoll_log.zip](/uploads/863cda4be493eb3f5bc1f2f325b4ac0b/recoll_log.zip)https://framagit.org/medoc92/recoll/-/issues/25Error in Python handler with Python 3.82020-11-02T13:55:11ZGerard SimonsError in Python handler with Python 3.8Hi there,
Thanks for the amazing work on this project! I am trying to use Recoll to index my Python scripts but it's not showing up. I tracked down the issue to the `rclpython` filter. If I run it manually on a Python script it outputs ...Hi there,
Thanks for the amazing work on this project! I am trying to use Recoll to index my Python scripts but it's not showing up. I tracked down the issue to the `rclpython` filter. If I run it manually on a Python script it outputs this error:
`ERROR: module 'cgi' has no attribute 'escape'`
It seems that Python 3.8 removed the escape in the cgi module in favor of `html.escape` so the fix is quite easy:
add `import html`
and replace `cgi.escape` with `html.escape`
Of course if we want to support both 3.8+ and older version we should do some version checking with a smart import. I could write something up if desiredhttps://framagit.org/medoc92/recoll/-/issues/26recollindex filling up the page cache, makes indexing in background inconvenient2021-05-05T19:49:38ZEgon Schielerecollindex filling up the page cache, makes indexing in background inconvenientThe recollindex process (and its sub processes) appears to fill up the kernel page cache when I manually start an indexing job.
This is inconvenient, as it means that older cached files from applications I'm running on my desktop (that ...The recollindex process (and its sub processes) appears to fill up the kernel page cache when I manually start an indexing job.
This is inconvenient, as it means that older cached files from applications I'm running on my desktop (that I'm currently not using but intending to use in near future) get removed and have to be loaded again when I reinvoke the application.
This makes the whole desktop less responsive.
As far as I understand the indexed files wouldn't need to be cached at all, because they're only read once. So caching them doesn't make sense.
So it'd be useful if the indexing processes made use of posix_fadvise() with POSIX_FADV_DONTNEED to prevent the read files from being kept in the cache.
Currently I'm testing if using https://github.com/Feh/nocache works and helps.
Maybe recollindex could be modified so that it works well and reliably with this, then there'd be no need to implement the fadvise thing at all.
Of course not caching the indexed files only makes sense for manual indexing, not for the background indexing from monitoring filesystem writes / modified files.https://framagit.org/medoc92/recoll/-/issues/27Recoll is failing to return matches found in the *contents* of Python scripts2020-11-03T14:12:18ZYekutielRecoll is failing to return matches found in the *contents* of Python scriptsRecoll seems great to me but I am having a problem using it.
I am using Recoll 1.26.3 + Xapian 1.4.14 on Lubuntu 20.04 LTS.
Recoll is only returning matches found in the **file names** of documents; however, Recoll is **failing** to r...Recoll seems great to me but I am having a problem using it.
I am using Recoll 1.26.3 + Xapian 1.4.14 on Lubuntu 20.04 LTS.
Recoll is only returning matches found in the **file names** of documents; however, Recoll is **failing** to return matches found in the **contents** of documents.
For example, if I were to search for "**submarine**" in Recoll, then Recoll would display in its results, a document with the file name "**How to build a submarine.txt**".
However, Recoll would **fail** to display a document with the file name, "**Navy vessels.txt**" which had the word **submarine** in its _contents_.
Whereas, if I were to search using [Catfish](https://launchpad.net/catfish-search) (instead of Recoll) Catfish would display in its results, a document with the file name "How to build a submarine.txt" and Catfish would also display a document with the file name "Navy vessels.txt which had the word submarine in in its **contents.* *
Put simply, Catfish is doing what I want but Recoll is failing to do what I want.
I doubt the following is consequential but..
1. A few days ago I installed Recoll. I have barely customized it at all.
1. The files being searched are .txt, .json, and .py files.
1. Only about 1000 files are being searched.
1. Finally, if you would like I can send you screenshot or even a screencast.https://framagit.org/medoc92/recoll/-/issues/28Enabling users to toggle more GUI elements2020-11-27T10:22:36ZYekutielEnabling users to toggle more GUI elementsRecoll should allow users to simply, quickly, and easily toggle more GUI elements. This does not require a GUI designer, nor, frankly, even much thought whatsoever. It's "dead simple."
Many open source software projects suffer from poor...Recoll should allow users to simply, quickly, and easily toggle more GUI elements. This does not require a GUI designer, nor, frankly, even much thought whatsoever. It's "dead simple."
Many open source software projects suffer from poor GUIs because the engineers who volunteer their time to work on these projects are not very interested in GUIs. Nonetheless, enabling users to simply, quickly, and easily toggle GUI elements is neither a subtle nor a sophisticated concept. Rather it is a remarkably simple concept as is demonstrated in the images (screenshots) below which I took of [Audacious](https://audacious-media-player.org/) (an open source audio player).
In case the menu bar were hidden, then, perhaps "right clicking" in the Recoll search box would show an option to make the menu bar visible.
![1](/uploads/d87b2417e51dbd1fdb68cf7be3633998/1.png)![2](/uploads/62ab9289f418360898b11e458278581f/2.png)![3](/uploads/4f1ae6c571f22ebbadda9bff0349e5a2/3.png)![4](/uploads/70af2ba45e1d93b955c89ad2b7303a08/4.png)https://framagit.org/medoc92/recoll/-/issues/29Recoll Indexing Stops with Error -> indexer: Indexing was interrupted.2021-02-09T09:30:26ZAlexander FraserRecoll Indexing Stops with Error -> indexer: Indexing was interrupted.Indexing stops with error, update index begins and stops with error, repeats for each update index. Will index a few hundred files before it stops again.
Windows Application Error from Event Viewer
Faulting application name: recollinde...Indexing stops with error, update index begins and stops with error, repeats for each update index. Will index a few hundred files before it stops again.
Windows Application Error from Event Viewer
Faulting application name: recollindex.exe, version: 0.0.0.0, time stamp: 0x5f7a0ff1
Faulting module name: ucrtbase.dll, version: 10.0.19041.546, time stamp: 0x73123758
Exception code: 0xc0000409
Fault offset: 0x0009edbb
Faulting process id: 0xc720
Faulting application start time: 0x01d6b9e42a0c1dbf
Faulting application path: C:\Program Files (x86)\Recoll\recollindex.exe
Faulting module path: C:\WINDOWS\System32\ucrtbase.dll
Report Id: 55f177ca-65a4-4649-8e5e-6a267e70e295
Faulting package full name:
Faulting package-relative application ID:
System
- Provider
[ Name] Application Error
- EventID 1000
[ Qualifiers] 0
Version 0
Level 2
Task 100
Opcode 0
Keywords 0x80000000000000
- TimeCreated
[ SystemTime] 2020-11-13T17:41:47.8310039Z
EventRecordID 8274
Correlation
- Execution
[ ProcessID] 0
[ ThreadID] 0
Channel Application
Computer LAPTOP-3DHR9H6F
Security
- EventData
recollindex.exe
0.0.0.0
5f7a0ff1
ucrtbase.dll
10.0.19041.546
73123758
c0000409
0009edbb
c720
01d6b9e42a0c1dbf
C:\Program Files (x86)\Recoll\recollindex.exe
C:\WINDOWS\System32\ucrtbase.dll
55f177ca-65a4-4649-8e5e-6a267e70e295https://framagit.org/medoc92/recoll/-/issues/30add info about how to find log to error message2020-11-19T10:17:34ZItsExtraadd info about how to find log to error messageHello,
I got an error while indexing:
> Indexing failed with additional message:
>
> indexer : Index creation failed. See log.
but I don't know how to find the log. It would be helpful to include this information.
I'll also note ...Hello,
I got an error while indexing:
> Indexing failed with additional message:
>
> indexer : Index creation failed. See log.
but I don't know how to find the log. It would be helpful to include this information.
I'll also note that I looked in the [user manual](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html), [FAQ](https://www.lesbonscomptes.com/recoll/faqsandhowtos/index.html#_indexing) and [man page](https://www.lesbonscomptes.com/recoll/manpages/recoll.1.html); none of them mention this; maybe they should.
if you decide to take my suggestion and add this, it also occurs to me that while you are at it, instructions about how to access support/documentation would be convenient. If you want to be really clear, state explicitly that the log file should be sent along with description of problem etc, because not everyone knows that.
am running Recoll 1.25.22 + Xapian 1.4.15 installed via homebrew on mac OShttps://framagit.org/medoc92/recoll/-/issues/31Recoll does not index files with "..." at the beginning of the name2020-11-26T10:35:23ZYekutielRecoll does not index files with "..." at the beginning of the nameThis is a minor quibble for what I suppose is an extreme, or perhaps even unique, [edge case](https://en.wikipedia.org/wiki/Edge_case). If it's difficult for you to fix then please don't bother because I can change my naming convention i...This is a minor quibble for what I suppose is an extreme, or perhaps even unique, [edge case](https://en.wikipedia.org/wiki/Edge_case). If it's difficult for you to fix then please don't bother because I can change my naming convention indicated below.
Recoll does not index files with "..." at the beginning of the name. It took me a while to figure out the problem. But eventually I realized why many of the files I was searching for were not appearing in Recoll's search results.
I use Recoll in conjunction with [AutoKey](https://github.com/autokey/autokey) (not to be confused with the presumably much more popular Auto**Hot**Key which runs on Microsoft Windows). AutoKey has many bugs, glitches, and is not actively maintained but once I figured out how to navigate it's myriad potholes (flaws) I have found that it works remarkably well for me. If I type, "**.tflmk**" then AutoKey replaces that text with "**Thanks for letting me know. I appreciate that.**"
**"tflmk"** is an acronym I created from the first letter of **T**hanks **f**or **l**etting **m**e **k**now. I prepend my shortcuts with "." to avoid creating shortcuts that are parts of the spelling of ordinary English words.
I like to prepend the names of some of the files I create in AutoKey with "..." as a way to remind myself that the title of that document comes from text I copied from within the document but is different than the first words in the document. For example, I might name a document, "**...this is an interesting approach.py**" if the document actually contains the text, **"I have found that this is an interesting approach."** It's as if I'm putting a tiny little bit of metadata into the file name.
Yes. I know. That is a strange way to name a file: a **very** strange way to name a file.https://framagit.org/medoc92/recoll/-/issues/32Indexing org-mode files2020-12-26T15:35:23ZjarifuriIndexing org-mode filesI keep most of my personal notes in the [org-mode](https://orgmode.org) files (mark-up language/mode heavily integrated with emacs). The [common strategy](https://orgmode.org/worg/org-faq.html#how-to-organize-org-mode-files) is to keep m...I keep most of my personal notes in the [org-mode](https://orgmode.org) files (mark-up language/mode heavily integrated with emacs). The [common strategy](https://orgmode.org/worg/org-faq.html#how-to-organize-org-mode-files) is to keep multiple notes in one file, i.e **each note represents the stand-alone document**.
My goal is to have the **org-mode files indexed according this logic** (find org-mode files -> split it by notes -> index individual notes). Currently I split the org-mode files outside of recoll and then they are indexed. This has several shortcomings - duplication of the original files, the search results point me to the duplicate file instead to the original one.
I thought I could write a filter solving this problem. I went through the doc and the filters code and tried to find some filter solving similar problem (split original and index splitted parts) but without success. Do you some idea or could you point me somewhere (or telling me this is just stupid idea is also perfectly fine)?
Example of my typical org-mode file with 3 notes
```org
* plotting and charting libs for bash :2009cli:
:PROPERTIES:
:dated: [2020-11-28 Sat 12:10]
:type: summary
:subject: programming sysadmin visualization
:ID: 20201128T121122.046072
:END:
- doporuceni https://stackoverflow.com/questions/123378/command-line-unix-ascii-based-charting-plotting-tool co tam je zmineno ve vetsinou dole
https://github.com/kroitor/asciichart - tahle bohuzel nema cli interface, ma plno portu
- .....................
* wireless sensor networks :2006hardware:
:PROPERTIES:
:dated: [2020-11-27 Fri 22:58]
:type: system
:subject: electronics hardware
:ID: 20201127T230007.537416
:END:
- concepts
- https://en.wikipedia.org/wiki/Wireless_sensor_network a https://en.wikipedia.org/wiki/Template:Wireless_Sensor_Network
- ......................
* law and legal code versioned on github
:PROPERTIES:
:dated: [2020-11-27 Fri 22:40]
:type: synergy cross-application
:subject: law programming business
:ID: 20201127T224147.109961
:END:
- origin https://qz.com/373326/the-entire-french-legal-code-dating-back-to-napoleon-is-now-on-github/
- french https://github.com/steeve/france.code-civil
- ........
```https://framagit.org/medoc92/recoll/-/issues/33I would like to be able to navigate down from "bbb" to "El Proyecto.py" by pr...2021-03-07T07:29:41ZYekutielI would like to be able to navigate down from "bbb" to "El Proyecto.py" by pressing the down arrow keyI would like to be able to navigate down from "bbb" to "El Proyecto.py" by pressing the down arrow key. Yes, I know I can press the tab key.
![Screenshot_at_12-32_on_Friday_18_Dec_2020](/uploads/f81170e1ce5bc1083ff604f5f00d2ab0/Screensh...I would like to be able to navigate down from "bbb" to "El Proyecto.py" by pressing the down arrow key. Yes, I know I can press the tab key.
![Screenshot_at_12-32_on_Friday_18_Dec_2020](/uploads/f81170e1ce5bc1083ff604f5f00d2ab0/Screenshot_at_12-32_on_Friday_18_Dec_2020.png)https://framagit.org/medoc92/recoll/-/issues/34Problem with recoll on M1 Mac running Big Sur - Couldn't stat '/Users/dylan/....2021-01-13T08:57:10Zroborobo2084Problem with recoll on M1 Mac running Big Sur - Couldn't stat '/Users/dylan/.recoll/xapiandbI attempted to install recoll on M1 Mac using Mac Ports. After various tinkering, it installed correctly. But I am getting this error which prevents recoll from indexing. It appears that it might be a folder/file permission issue. How...I attempted to install recoll on M1 Mac using Mac Ports. After various tinkering, it installed correctly. But I am getting this error which prevents recoll from indexing. It appears that it might be a folder/file permission issue. However, I have no idea how to fix it.
(I had installed recoll on another Big Sur Mac on got the libpng warnings, which were not an issue. But the "Couldn't stat '/Users/dylan/.recoll/xapiandb" seenms to be a fatal error
:3:common/rclinit.cpp:353::Recoll 1.25.22 + Xapian 1.4.17 [/Users/dylan/.recoll]
libpng warning: iCCP: known incorrect sRGB profile
libpng warning: iCCP: known incorrect sRGB profile
libpng warning: iCCP: known incorrect sRGB profile
:2:rcldb/rcldb.cpp:991::Db::open: exception while opening [/Users/dylan/.recoll/xapiandb]: Couldn't stat '/Users/dylan/.recoll/xapiandb'
:2:rcldb/rcldb.cpp:991::Db::open: exception while opening [/Users/dylan/.recoll/xapiandb]: Couldn't stat '/Users/dylan/.recoll/xapiandb'
:2:rcldb/rcldb.cpp:991::Db::open: exception while opening [/Users/dylan/.recoll/xapiandb]: Couldn't stat '/Users/dylan/.recoll/xapiandb'
qt.qpa.fonts: Populating font family aliases took 227 ms. Replace uses of missing font family "Sans Serif" with one that exists to avoid this cost.
:2:utils/execmd.cpp:557::ExecCmd::startExec: recollindex not found
:2:utils/execmd.cpp:557::ExecCmd::startExec: recollindex not found
2020-12-21 12:30:41.077 recoll[41905:774346] modalSession has been exited prematurely - check for a reentrant call to endModalSession:
:2:rcldb/rcldb.cpp:991::Db::open: exception while opening [/Users/dylan/.recoll/xapiandb]: Couldn't stat '/Users/dylan/.recoll/xapiandb'https://framagit.org/medoc92/recoll/-/issues/35cron can't find recollindex (Big Sur, MacPorts 2.6.4)2021-01-14T09:57:46Zedubyacron can't find recollindex (Big Sur, MacPorts 2.6.4)On my machine recollindex was installed for some reason by MacPorts 2.6.4 on macOS 11.1 to /opt/local/bin but cron looks for recollindex only in /bin and /usr/bin. I needed to write a small script to source my .bash_profile in the cronta...On my machine recollindex was installed for some reason by MacPorts 2.6.4 on macOS 11.1 to /opt/local/bin but cron looks for recollindex only in /bin and /usr/bin. I needed to write a small script to source my .bash_profile in the crontab ([something like this](https://stackoverflow.com/a/26044812/809189)).https://framagit.org/medoc92/recoll/-/issues/36Mime type in results2021-01-25T10:55:16ZMassimiliano MirraMime type in resultsI'm trying to read the output from `recoll -t` in another program, and I noticed that the returned mime type does not match the mime type of the file, e.g. it's `text/html` for both GNU info and ePub files. Is this intended?I'm trying to read the output from `recoll -t` in another program, and I noticed that the returned mime type does not match the mime type of the file, e.g. it's `text/html` for both GNU info and ePub files. Is this intended?https://framagit.org/medoc92/recoll/-/issues/37incomplete german translation2021-02-06T12:25:07Zfred fireincomplete german translationAt the moment the german translation is not complete.
I like to complete it, but i do not know how to fork this project.At the moment the german translation is not complete.
I like to complete it, but i do not know how to fork this project.https://framagit.org/medoc92/recoll/-/issues/38Recoll does not index the abstract of text/x-orgmode files | Recoll 1.28.5, X...2022-04-06T12:00:40ZHanshenWangRecoll does not index the abstract of text/x-orgmode files | Recoll 1.28.5, Xapian 1.4.14, Ubuntu 20.04, Org-Mode 9.4.4Good Morning,
Thank you for Recoll! It's been such a blessing to use and has improved my productivity :). Recoll is working fine except for the indexing of text/x-orgmode files. The abstract of .org files is missing. I have tried both t...Good Morning,
Thank you for Recoll! It's been such a blessing to use and has improved my productivity :). Recoll is working fine except for the indexing of text/x-orgmode files. The abstract of .org files is missing. I have tried both the GUI and the cli with -A. Not sure how I can generate an error message for you as I have made sure text/x-orgmode is included in the indexedMIMEs and the program tells me I am missing no helpers. I wish I had more information, and I will gladly run more tests.
The filenames of .org mode files show up fine. For anybody else who may have this issue, I found a workaround after much googling. Linking it here in case it might help anybody else. https://lists.gnu.org/archive/html/emacs-orgmode/2015-07/msg00425.html
The hack is to config your mimeconf to first convert .org to .html. I'm a happy camper, and this is a minor bug. Just thought you guys should know.
Cheers,https://framagit.org/medoc92/recoll/-/issues/39Building from tar ball2021-02-09T17:12:27ZaldobrantiBuilding from tar ball./configure throws an error with
```
./configure: line 13844: syntax error near unexpected token `XSLT,'
./configure: line 13844: `PKG_CHECK_MODULES(XSLT, libxslt, , as_fn_error $? "libxslt" "$LINENO" 5)
```
what is missing ?
```
$..../configure throws an error with
```
./configure: line 13844: syntax error near unexpected token `XSLT,'
./configure: line 13844: `PKG_CHECK_MODULES(XSLT, libxslt, , as_fn_error $? "libxslt" "$LINENO" 5)
```
what is missing ?
```
$ apt list --installed | grep -i xslt
libxml-libxslt-perl/stable,now 1.96-1+b1 amd64 [installed,automatic]
libxslt1.1/stable,now 1.1.32-2.2~deb10u1 amd64 [installed]
python-libxslt1/stable,now 1.1.32-2.2~deb10u1 amd64 [installed,automatic]
xsltproc/stable,now 1.1.32-2.2~deb10u1 amd64 [installed]
```https://framagit.org/medoc92/recoll/-/issues/40ctrl+1 lead to crash2021-10-06T17:58:58Zitkindctrl+1 lead to crashHi, I have a hotkey ctrl+1 (till 5) for changing between languages on windows. When I am in the search bar an press the hotkey, Recoll freezes and needs a restart. This happens without exception. The hotkey alt+shift works without freezi...Hi, I have a hotkey ctrl+1 (till 5) for changing between languages on windows. When I am in the search bar an press the hotkey, Recoll freezes and needs a restart. This happens without exception. The hotkey alt+shift works without freezing but this is not an ideal solution.https://framagit.org/medoc92/recoll/-/issues/41state seems wedged at Flushing2021-02-24T13:09:50Zaldobrantistate seems wedged at FlushingLooks like it finished` recollindex -k`
but then won't return to command line
```
:3:rcldb/rcldb.cpp:763::Db::add: docid 287176 updated [/ImageWillsMac/Public/Drop Box/GOPR0096.MP4|]
:3:rcldb/rcldb.cpp:763::Db::add: docid 287212 updated...Looks like it finished` recollindex -k`
but then won't return to command line
```
:3:rcldb/rcldb.cpp:763::Db::add: docid 287176 updated [/ImageWillsMac/Public/Drop Box/GOPR0096.MP4|]
:3:rcldb/rcldb.cpp:763::Db::add: docid 287212 updated [/ImageWillsMac/Public/Drop Box/GOPR0097.MP4|]
:3:index/fsindexer.cpp:246::FsIndexer::index missing helper program(s):
catdvi (application/x-dvi)
lyx (application/x-lyx)
:3:index/fsindexer.cpp:251::fsindexer index time: 2946325 mS
:3:rcldb/rcldb.cpp:2011::Db::add/delete: txt size >= 25 Mb, flushing
:3:rcldb/rcldb.cpp:2011::Db::add/delete: txt size >= 25 Mb, flushing
^C
^C
^Z
:3:rcldb/rcldb.cpp:2011::Db::add/delete: txt size >= 25 Mb, flushing
:3:rcldb/rcldb.cpp:2011::Db::add/delete: txt size >= 25 Mb, flushing
:3:rcldb/rcldb.cpp:2011::Db::add/delete: txt size >= 25 Mb, flushing
```
how long can it take to flush out 25Mb data ?https://framagit.org/medoc92/recoll/-/issues/42systematic corruption of `postlist`2021-02-25T14:31:35ZWalex Bsystematic corruption of `postlist`I am using Ubuntu LTS 20 with `recoll` 1.28.5-1 from the `recoll-backports` PPA and `libxapian30` 1.4.14-2 from the standard archive
After many years and since not long ago I am getting extreme database corruption on indexing from scrat...I am using Ubuntu LTS 20 with `recoll` 1.28.5-1 from the `recoll-backports` PPA and `libxapian30` 1.4.14-2 from the standard archive
After many years and since not long ago I am getting extreme database corruption on indexing from scratch or indexing from a known good pre-existing database:
```
termfreq 28 != # of entries 25
collfreq 35 != sum wdf 32
termfreq 20 != # of entries 19
collfreq 20 != sum wdf 19
termfreq 6 != # of entries 4
collfreq 6 != sum wdf 4
termfreq 736 != # of entries 729
collfreq 769 != sum wdf 762
termfreq 16 != # of entries 15
collfreq 18 != sum wdf 17
termfreq 462 != # of entries 248
collfreq 643 != sum wdf 350
Value stats frequency for slot 1 is 971742 but recounting gives 965518
Value stats frequency for slot 2 is 1306466 but recounting gives 1298824
Value stats frequency for slot 10 is 1306466 but recounting gives 1298824
postlist table errors found: 712930
position:
Counted 477861343 entries in the Btree
Non-sequential
B-tree checked okay
position table structure checked OK
spelling:
Lazily created, and not yet used.
synonym:
Counted 1071707 entries in the Btree
Non-sequential
B-tree checked okay
synonym table: Don't know how to check structure
Total errors found: 712931
```https://framagit.org/medoc92/recoll/-/issues/43music files -- metadata2021-02-26T11:39:42Zaldobrantimusic files -- metadata
I have a large hoard of muisc files accumulated over the years . Recollindex complains that their metadata is wrong
CMDTALK: rclaudio: b'/ImageWillsMac/Music/Iggy Pop/Lust For Life/08 Neighborhood Threat.mp3' : time data '1977-78-19' ...
I have a large hoard of muisc files accumulated over the years . Recollindex complains that their metadata is wrong
CMDTALK: rclaudio: b'/ImageWillsMac/Music/Iggy Pop/Lust For Life/08 Neighborhood Threat.mp3' : time data '1977-78-19' does not match format '%Y-%m-%d'
Can I turn off that check ?https://framagit.org/medoc92/recoll/-/issues/44Adding a bit of space to separate each search result2021-03-02T19:04:30ZR8s6Adding a bit of space to separate each search resultThe current "result list paragraph" html is:
```
<table class="respar">
<tr>
<td><a href='%U'><img src='%I' width='64'></a></td>
<td>%L &nbsp;<i>%S</i> &nbsp;&nbsp;<b>%T</b><br>
<span style='white-space:nowrap'><i>%M</i>&nbsp;%D</span>&...The current "result list paragraph" html is:
```
<table class="respar">
<tr>
<td><a href='%U'><img src='%I' width='64'></a></td>
<td>%L <i>%S</i> <b>%T</b><br>
<span style='white-space:nowrap'><i>%M</i> %D</span> <i>%U</i> %i<br>
%A %K</td>
</tr></table>
```
May I suggest to add a `<br>` before `</tr></table>` by default, to make each result stand out a little bit more from the other results:
```
<table class="respar">
<tr>
<td><a href='%U'><img src='%I' width='64'></a></td>
<td>%L <i>%S</i> <b>%T</b><br>
<span style='white-space:nowrap'><i>%M</i> %D</span> <i>%U</i> %i<br>
%A %K</td><br>
</tr></table>
```https://framagit.org/medoc92/recoll/-/issues/45Feature Request - Respect .gitignore2021-03-03T09:37:16ZalvitawaFeature Request - Respect .gitignoreHi, I've been testing recoll out on my home folder, but I have a problem:
Some of the files that get indexed are source code i've worked on, but that source code also has downloaded dependencies. I would not like to search through those...Hi, I've been testing recoll out on my home folder, but I have a problem:
Some of the files that get indexed are source code i've worked on, but that source code also has downloaded dependencies. I would not like to search through those dependencies when using recoll, since there is a lot of code in them and I don't care about it. A search for a simple word now tends to bring up a myriad of results from these dependencies which I am not interested in. Based on propper git practices however, these dependencies are .gitignored. So, is there a chance you could make recoll respect .gitignores? It also seems like a nice way in general to exclude certain files/folders from the search, regardless of whether you intend to use version control or not (or maybe it should be called .recollignore). I know there are tools specifically for searching through source code that do this (like the silver searcher), but those can't read pdf's and all the other nice things going for recoll. I also don't want to exclude my source code directory. I would just like to remove clutter from recoll's search results really.https://framagit.org/medoc92/recoll/-/issues/46Recollindex fails on certain mp3 files.2021-03-07T17:27:58ZLockywolfRecollindex fails on certain mp3 files.As mentioned on the mailing list, recollindex fails at certain mp3 files, reporting
```
lockywolf@delllaptop:~$ recollindex -i test-recoll-fails.mp3
CMDTALK: rclaudio: b'/home/lockywolf/test-recoll-fails.mp3' : Open failed: can't sync...As mentioned on the mailing list, recollindex fails at certain mp3 files, reporting
```
lockywolf@delllaptop:~$ recollindex -i test-recoll-fails.mp3
CMDTALK: rclaudio: b'/home/lockywolf/test-recoll-fails.mp3' : Open failed: can't sync to MPEG frame
```
Indeed, vlc and mpv report that the file is somehow imperfect, but they play it, and exiftool extracts tags no problem.
Attached:
![test-recoll-fails](/uploads/5721f81c3c3a800444050774d568f87b/test-recoll-fails.mp3)https://framagit.org/medoc92/recoll/-/issues/47Ctrl + number works properly but Ctrl + letter fails to work properly (Recoll...2021-03-09T12:07:07ZYekutielCtrl + number works properly but Ctrl + letter fails to work properly (Recoll 1.29.0 – Linux)Version: **Recoll 1.29.0** + Xapian 1.4.14
Operating System: **Linux**
I recall Ctrl + letter working properly in an alpha (testing) version of Recoll. But when I update Linux the PPA (Personal Package Archive) automatically updates Re...Version: **Recoll 1.29.0** + Xapian 1.4.14
Operating System: **Linux**
I recall Ctrl + letter working properly in an alpha (testing) version of Recoll. But when I update Linux the PPA (Personal Package Archive) automatically updates Recoll to the current standard release.
When I press `Ctrl + 0`, `Ctrl + 1`, `Ctrl + 2` and so on I properly "jump" to result 0, 1, 2, and so on. This is correct.
When I press `Ctrl + a`, `Ctrl + b`, `Ctrl + c` and so on nothing appears to happen. This is incorrect.
![Screenshot_at_13-39_on_Monday__8_Mar_2021](/uploads/0fc1fef0f84d8810a525dfbcd899f5d8/Screenshot_at_13-39_on_Monday__8_Mar_2021.png)https://framagit.org/medoc92/recoll/-/issues/48Housekeeping- "...disable Ctrl+ Shift + (a-z)..." instead of "...disable Ctrl...2021-03-12T07:58:41ZYekutielHousekeeping- "...disable Ctrl+ Shift + (a-z)..." instead of "...disable Ctrl+ (a-z)..."The following should say something like, "Disable Ctrl + (0-9) and disable Ctrl+ Shift + (a-z) shortcuts for jumping to a table row." As you know, Ctrl + (a-z) does not work.
![Screenshot_at_14-27_on_Thursday_11_Mar_2021](/uploads/f0d58...The following should say something like, "Disable Ctrl + (0-9) and disable Ctrl+ Shift + (a-z) shortcuts for jumping to a table row." As you know, Ctrl + (a-z) does not work.
![Screenshot_at_14-27_on_Thursday_11_Mar_2021](/uploads/f0d5809f31e96165f8ada218401f035c/Screenshot_at_14-27_on_Thursday_11_Mar_2021.png)https://framagit.org/medoc92/recoll/-/issues/49Feature request: ability to preview messages in Recoll similar to Gmail2021-03-27T06:41:16ZYekutielFeature request: ability to preview messages in Recoll similar to GmailI don't know if the following is too technically daunting (I don't know if the following is too difficult) but ideally I would like to be able to preview messages in Recoll similar to the way I can preview messages in Gmail.
Therefore, ...I don't know if the following is too technically daunting (I don't know if the following is too difficult) but ideally I would like to be able to preview messages in Recoll similar to the way I can preview messages in Gmail.
Therefore, when you have a chance, for about ten to twenty seconds please watch any of the three following YouTube videos...
1. [Gmail Tips: How to Customize Your Gmail Layout + How to Organize Your Gmail Inbox](https://youtu.be/QlnGEMF3d0o?t=376)
1. [Gmail View Settings You Need to Try Right Now (Quick Tips)](https://youtu.be/hrhOZrP1ZN0?t=379)
1. [How to get images preview in ranger](https://youtu.be/EKKKFdAXQdY?t=19)
In case you are curious and aren't familiar with [ranger](https://en.wikipedia.org/wiki/Ranger_(file_manager)) here's a link to [ranger (on Github)](https://github.com/ranger/ranger).https://framagit.org/medoc92/recoll/-/issues/50Feature request: Positive confirmation when "Copy current text result to clip...2021-03-17T17:29:51ZYekutielFeature request: Positive confirmation when "Copy current text result to clipboard" occursC'est un détail très mineur. N'hésitez pas à ignorer cette demande à moins qu'elle ne soit très simple à mettre en œuvre.
Quand je choisis…
`Copy current text result to clipboard`
Je voudrais voir une sorte de confirmation positiv...C'est un détail très mineur. N'hésitez pas à ignorer cette demande à moins qu'elle ne soit très simple à mettre en œuvre.
Quand je choisis…
`Copy current text result to clipboard`
Je voudrais voir une sorte de confirmation positive - comme une boîte de dialogue qui apparaîtrait sur mon écran pendant 1 ou 2 secondes (puis disparaîtrait) - disant quelque chose comme,
`The text was successfully copy to your clipboard.`
![Screenshot_at_11-57_on_Sunday_14_Mar_2021](/uploads/e38b43880880024c1d87d116a6a7e28c/Screenshot_at_11-57_on_Sunday_14_Mar_2021.png)https://framagit.org/medoc92/recoll/-/issues/51Quibble about how to toggle between the "weird" view and the "normal" view in...2021-03-17T15:59:42ZYekutielQuibble about how to toggle between the "weird" view and the "normal" view in the preview paneI suggest using two radio buttons instead of using a checkbox because I suppose that would a little clearer (easier) for ordinary users to understand.
![Screenshot_at_12-02_on_Tuesday_16_Mar_2021](/uploads/bb53a10837b7050acf661a350e64d2...I suggest using two radio buttons instead of using a checkbox because I suppose that would a little clearer (easier) for ordinary users to understand.
![Screenshot_at_12-02_on_Tuesday_16_Mar_2021](/uploads/bb53a10837b7050acf661a350e64d2e5/Screenshot_at_12-02_on_Tuesday_16_Mar_2021.png)https://framagit.org/medoc92/recoll/-/issues/52The advanced search dialog opens *behind* the main Recoll window (on my comp...2021-03-17T15:44:34ZYekutielThe advanced search dialog opens *behind* the main Recoll window (on my computer running Linux).I use Linux. Perhaps this is not a problem on Microsof Windows.
I don't normally use this feature nor do I intend to use this feature. But, of course, other users might use this feature. I just checked **"Start with advanced search dial...I use Linux. Perhaps this is not a problem on Microsof Windows.
I don't normally use this feature nor do I intend to use this feature. But, of course, other users might use this feature. I just checked **"Start with advanced search dialog open"**
![Screenshot_at_12-15_on_Tuesday_16_Mar_2021](/uploads/5fa73b52586e86ad11afbd86027bc7a8/Screenshot_at_12-15_on_Tuesday_16_Mar_2021.png)
and then restarted Recoll. When I did that, the advanced search dialog opened **behind** the main Recoll window. I suppose that the advanced search dialog should open **in front of** the main Recoll window. Right?https://framagit.org/medoc92/recoll/-/issues/53Relatively unimportant feature request: ten unique font colors and/or uniqu...2021-03-27T06:43:38ZYekutielRelatively unimportant feature request: ten unique font colors and/or unique background colors for search resultsPlease don't bother trying to implement this relatively unimportant feature request unless it is very easy for you.
See, I suppose adding this feature to Microsoft Excel by using Microsoft Excel VBA or adding this feature to Google Shee...Please don't bother trying to implement this relatively unimportant feature request unless it is very easy for you.
See, I suppose adding this feature to Microsoft Excel by using Microsoft Excel VBA or adding this feature to Google Sheets using Google Apps Script would be a trivial programming task. But I am unfamiliar with the programming language you are using to write the GUI for Recoll. Perhaps in that language this would be a Herculean task. I realize that a task that is trivial in one programming language can be extremely difficult in another programming language. Engineers have "worn out my ears" explaining this to me.
Be that as it may, here's what I see now...
![Screenshot_at_12-31_on_Tuesday_16_Mar_2021](/uploads/8eac58fcf09eb6f25f4cc951a55bbf63/Screenshot_at_12-31_on_Tuesday_16_Mar_2021.png)
I would like to be able to choose a unique font color and/or unique background color for, say, the first ten rows of search results in Recoll. In other words, I imagine I might choose for...
1. Row 1 to have a dark blue font on a light grey background and,
1. Row 2 to have a dark green font on a light yellow background and,
1. Row 3 to have a light brown font on a dark purple background and so on.
In other words, I would choose a unique font color and/or unique background color for rows 1 through 10. Then I imagine that...
1. Row 11 would have the same font color and/or unique background color as row 1 and,
1. Row 12 would have the same font color and/or unique background color as row 2 and,
1. Row 13 would have the same font color and/or unique background color as row 3 and so on.
and, furthermore I imagine that...
1. Row 21 would have the same font color and/or unique background color as row 1 (and row 11) and,
1. Row 22 would have the same font color and/or unique background color as row 2 (and row 12) and,
1. Row 23 would have the same font color and/or unique background color as row 3 (and row 13) and so on.
Traffic lights could use 1 to indicate go, 2 to indicate caution, and 3 to indicate stop. But generally traffic lights use green to indicate go, yellow or orange to indicate caution, and red to indicate stop. This is obviously not an accident. Normal people generally can identify significantly different colors very quickly and very easily.
Maybe for engineers it would make more sense if traffic lights were to use 1 to indicate go, 2 to indicate caution, and 3 to indicate stop. But that is not the case for normal people.https://framagit.org/medoc92/recoll/-/issues/54Minor glitch: Ctrl + Shift + letter shows "weird" view even when I have Recol...2021-03-17T16:54:43ZYekutielMinor glitch: Ctrl + Shift + letter shows "weird" view even when I have Recoll set to show "normal" viewI suppose this will be extremely easy for you to fix.
I checked the following box...
![Screenshot_at_13-00_on_Tuesday_16_Mar_2021](/uploads/1e2f1b89a6e22a2c717c0953529fbddd/Screenshot_at_13-00_on_Tuesday_16_Mar_2021.png)
so that I see...I suppose this will be extremely easy for you to fix.
I checked the following box...
![Screenshot_at_13-00_on_Tuesday_16_Mar_2021](/uploads/1e2f1b89a6e22a2c717c0953529fbddd/Screenshot_at_13-00_on_Tuesday_16_Mar_2021.png)
so that I see "normal" text in the preview pane instead of "weird" text in the preview pane.
Yet when I press `Ctrl + Shift + c` I see "weird" text in the preview pane. This is **incorrect**. I should see "normal" text in the preview pane. The following is what I see after I press `Ctrl + Shift + c`...
![Screenshot_at_12-57_on_Tuesday_16_Mar_2021](/uploads/35b026b9cd540dadaba01d5e500c1fe5/Screenshot_at_12-57_on_Tuesday_16_Mar_2021.png)
------------------
whereas when I press `Ctrl + 2` I see "normal" text in the preview pane. This is **correct**. The following is what I see after I press `Ctrl + 2`...
------------------
![Screenshot_at_12-58_on_Tuesday_16_Mar_2021](/uploads/ad460595ccc8b2bdf0bf52b29ad80d0f/Screenshot_at_12-58_on_Tuesday_16_Mar_2021.png)https://framagit.org/medoc92/recoll/-/issues/55Have you considered creating a configuration wizard?2021-03-25T16:10:47ZYekutielHave you considered creating a configuration wizard?Have you considered creating a [configuration wizard](https://en.wikipedia.org/wiki/Wizard_(software))? To ensure you understand my question, here is an example of a configuration wizard...
https://help.timecockpit.com/doc/getting-start...Have you considered creating a [configuration wizard](https://en.wikipedia.org/wiki/Wizard_(software))? To ensure you understand my question, here is an example of a configuration wizard...
https://help.timecockpit.com/doc/getting-started/configuration-wizard.html
Obviously, that particular configuration wizard would be completely inappropriate for Recoll. It is merely an example.
If you were to create a configuration wizard, one question I would recommend including (with two radio buttons) would be...
"Are you an engineer or a user who requires detailed technical information in your search results"
or
"Are you a non-engineer (an ordinary user) who wants to see text only in your search results (does not require detailed technical information in your search results)."
Please note, you will not be committing irrevocably to one of the two approaches indicated above. Rather, these options can be changed in Recoll's preferences.
-----------------
-----------------
-----------------
I suppose that most non-engineers (ordinary users) would like to see search results in the following format...
![Screenshot_at_13-32_on_Tuesday_16_Mar_2021](/uploads/e5ecea20f210dcd0debe35196f00db87/Screenshot_at_13-32_on_Tuesday_16_Mar_2021.png)
Of course, I realize that most engineers probably like to see more technical information.https://framagit.org/medoc92/recoll/-/issues/56unresolved dependencies when installing 1.29.x for Linux Mint 19.32021-04-11T13:26:53ZWolfgang Sprickunresolved dependencies when installing 1.29.x for Linux Mint 19.3Following description in https://www.lesbonscomptes.com/recoll/pages/download.html#debian
command
sudo apt-get install recoll python-recoll python3-recoll
leads to
Die folgenden Pakete haben unerfüllte Abhängigkeiten:
python-recoll : H...Following description in https://www.lesbonscomptes.com/recoll/pages/download.html#debian
command
sudo apt-get install recoll python-recoll python3-recoll
leads to
Die folgenden Pakete haben unerfüllte Abhängigkeiten:
python-recoll : Hängt ab von: python2 ist aber nicht installierbar
Hängt ab von: recollcmd (= 1.29.2-1~ppa1~buster) soll aber nicht installiert werden
python-recoll-dbgsym:i386 : Hängt ab von: python-recoll:i386 (= 1.29.2-1~ppa1~buster) soll aber nicht installiert werden
python3-recoll : Hängt ab von: recollcmd (= 1.29.2-1~ppa1~buster) soll aber nicht installiert werden
Hängt ab von: python3 (>= 3.7~) aber 3.6.7-1~18.04 soll installiert werden
recoll : Hängt ab von: recollcmd soll aber nicht installiert werden
Hängt ab von: recollgui soll aber nicht installiert werden
Sorry, German version, "hängt ab" means "depends on", "nicht installierbar" means "cannot be installed", "soll aber nicht installiert werden" means "but shall not be installed".https://framagit.org/medoc92/recoll/-/issues/57"copy filename" actually copies the filepath + not the full path inside a zip...2021-03-27T08:05:35Zthiswillbeyourgithub"copy filename" actually copies the filepath + not the full path inside a zip fileHi,
When searching something then right clicking on a document in the result list, the option "copy filename" doesn't behave as expected, it copies the filepath. So pretty similar to "copy url".
Additionaly, when selecting "copy filena...Hi,
When searching something then right clicking on a document in the result list, the option "copy filename" doesn't behave as expected, it copies the filepath. So pretty similar to "copy url".
Additionaly, when selecting "copy filename" (which supposedly copies the path) on file that has been found inside a zip, the path stops to the tip and not to which file in the content of the zip.
Tested on ubuntu 18, recoll 1.29.2
My preferred solution would be to have 3 options in a "copy" submenu :
1. copy > filename
2. copy > full URL
3. copy > full path
and have the full path end not at the zip level but at the file location inside the zip.
In the same vein, the file path is indexed as part of each file but for a file B.txt that is inside A.zip (so A.zip/B.txt) is the filename indexed as "A.zip" or "B.txt" or both or ? That is not clear to me.https://framagit.org/medoc92/recoll/-/issues/58in preview shift+scrolling changes the text size but moves the text around2021-03-27T08:09:04Zthiswillbeyourgithubin preview shift+scrolling changes the text size but moves the text aroundHi,
This is annoying as I have to then click on "next/previous" to go again where the match was. And it makes it really annoying when you were reading somewhere where no match is.
Thought this could be fixed somehow :)Hi,
This is annoying as I have to then click on "next/previous" to go again where the match was. And it makes it really annoying when you were reading somewhere where no match is.
Thought this could be fixed somehow :)https://framagit.org/medoc92/recoll/-/issues/59Is there a way to index my anki database or RSS feed into recoll? YES!2021-03-19T12:55:13ZthiswillbeyourgithubIs there a way to index my anki database or RSS feed into recoll? YES!Hi,
This is a shameless plug to my own little dumb script that I use to have recoll being able to search through everything that matters to me. So my anki database and my newsboat RSS reader.
When I was searching for the right desktop ...Hi,
This is a shameless plug to my own little dumb script that I use to have recoll being able to search through everything that matters to me. So my anki database and my newsboat RSS reader.
When I was searching for the right desktop search engine for myself I searched in the issues of each project to see if somehow someone had added a compatibility with anki or sql or my rss and found nothing. So I created my own dumb script and figured that it would help people to have a mention of it here.
here it is to index anki :
https://github.com/thiswillbeyourgithub/IndexableAnki
and to index my newsboat rss reader db :
https://github.com/thiswillbeyourgithub/IndexableNewsboat
I hope the owner won't mind, those scripts are really only functionnal and I don't expect any gain from "advertising" this here. It's really out of altruism for me :)
I'm closing this right away so as to not interfere with real issues.https://framagit.org/medoc92/recoll/-/issues/60"trigger incremental pass" is greyed in light mode but of normal color in dar...2021-03-19T16:49:57Zthiswillbeyourgithub"trigger incremental pass" is greyed in light mode but of normal color in dark modehttps://framagit.org/medoc92/recoll/-/issues/61"trigger incremental pass" is barely mentionned in the docs2021-03-19T16:32:41Zthiswillbeyourgithub"trigger incremental pass" is barely mentionned in the docsHi,
the only mentions of this button are :
> Real time monitoring/indexing is performed by starting the recollindex -m command. With this option, recollindex will detach from the terminal and become a daemon, permanently monitoring fil...Hi,
the only mentions of this button are :
> Real time monitoring/indexing is performed by starting the recollindex -m command. With this option, recollindex will detach from the terminal and become a daemon, permanently monitoring file changes and updating the index.
> In this situation, the recoll GUI File menu makes two operations available: 'Stop' and 'Trigger incremental pass'.
> While it is convenient that data is indexed in real time, repeated indexing can generate a significant load on the system when files such as email folders change. Also, monitoring large file trees by itself significantly taxes system resources. You probably do not want to enable it if your system is short on resources. Periodic indexing is adequate in most cases.
> As of Recoll 1.24, you can set the monitordirs configuration variable to specify that only a subset of your indexed files will be monitored for instant indexing. In this situation, an incremental pass on the full tree can be triggered by either restarting the indexer, or just running recollindex, which will notify the running process. The recoll GUI also has a menu entry for this.
Unfortunately I still don't understand what this button is for. Does it re-indexes all files? If so what would be the difference with "rebuild index"? I also don't understand *when* it is not greyed out.https://framagit.org/medoc92/recoll/-/issues/62questions regarding the button "rebuild index"2021-03-19T16:22:59Zthiswillbeyourgithubquestions regarding the button "rebuild index"Hi,
Here's what the doc says :
> The GUI File menu has entries to start or stop the current indexing operation. When indexing is not currently running, you have a choice of updating the index or rebuilding it (the first choice only pr...Hi,
Here's what the doc says :
> The GUI File menu has entries to start or stop the current indexing operation. When indexing is not currently running, you have a choice of updating the index or rebuilding it (the first choice only processes changed files, the second one zeroes the index before starting so that all files are processed)
What do you mean by "zero"? Do you simply mean that the whole index is trashed and restarted from scratch? If not what's the difference between "zeroing" and "restarting from scratch"? Also, what happens with the webcache? I recall you saying that it is kept but that should be mentionned IMO. As well as mentionning that it is kept in a separate folder and not in the main index IIRC, this is interesting for people that would be interested in backing up this in particular or editing it by hand.
also, I would prefer to word this sentence as containing the name of the buttons as it makes it faster to find ("...you have a choice between "update index" and "rebuild index"...) but that's just me.
Cheershttps://framagit.org/medoc92/recoll/-/issues/63Indexing failed2021-05-12T08:12:53ZGeorge HenryIndexing failed![Recoll_UI_error_message](/uploads/ee667aeb43fb83cbc8300b148dbcd9ab/Recoll_UI_error_message.png)[2021.03.14_13.42_recoll-indexer-log.txt](/uploads/5d18166bcd9f3eaa33d4d62d6390a6a9/2021.03.14_13.42_recoll-indexer-log.txt)
I changed the ...![Recoll_UI_error_message](/uploads/ee667aeb43fb83cbc8300b148dbcd9ab/Recoll_UI_error_message.png)[2021.03.14_13.42_recoll-indexer-log.txt](/uploads/5d18166bcd9f3eaa33d4d62d6390a6a9/2021.03.14_13.42_recoll-indexer-log.txt)
I changed the index configuration to limit indexing to three directories - documents, downloads and something else relatively modest - and still go the same result.https://framagit.org/medoc92/recoll/-/issues/64question regarding the option "find similar documents"2022-12-26T05:59:32Zthiswillbeyourgithubquestion regarding the option "find similar documents"Hi,
the docs says :
> Finding related documents. Selecting the Find similar documents entry in the result list paragraph right-click menu will select a set of "interesting" terms from the current result, and insert them into the simple...Hi,
the docs says :
> Finding related documents. Selecting the Find similar documents entry in the result list paragraph right-click menu will select a set of "interesting" terms from the current result, and insert them into the simple search entry field. You can then possibly edit the list and start a search to find documents which may be apparented to the current result.
I was curious as to the meaning of "interesting" here. Would you mind sharing details? As you said elsewhere, this is a peculiar function that is rarely used for most users. But I can envision use cases for people and companies but they would maybe be turned away by the (seemingly) lack of documentation as to how it works.
I have an interest in AI and know that this kind of functionnality is important to some but absolutely non trivial. Recent advancement were done by doc2vec for example (https://www.tutorialspoint.com/gensim/gensim_doc2vec_model.htm & https://github.com/bnosac/doc2vec)
So where I'm getting at is that I would be interested in reading more about how your algorithm implemented "similar document finding" and adding/linking it to the docs.
Cheers!https://framagit.org/medoc92/recoll/-/issues/65MINOR on switching between light and dark mode2021-03-22T13:20:18ZthiswillbeyourgithubMINOR on switching between light and dark modeWhen switching the dark/light mode, there's a popup that asks to do another query. But isn't it better to just rerun the current query automatically? Closing the popup will probably take as much time as rerunning a query anyway. That's j...When switching the dark/light mode, there's a popup that asks to do another query. But isn't it better to just rerun the current query automatically? Closing the popup will probably take as much time as rerunning a query anyway. That's just a minor annoyance.https://framagit.org/medoc92/recoll/-/issues/66MEDIUM add a way to change fonts in the "View" button2023-02-17T13:09:09ZthiswillbeyourgithubMEDIUM add a way to change fonts in the "View" buttonTo add in the View system menu, maybe a way to change the font size? Recoll is a text heavy application that is sometimes rendered a bit small on some setup. I know I often use a stand for my laptop and would like to change the font size...To add in the View system menu, maybe a way to change the font size? Recoll is a text heavy application that is sometimes rendered a bit small on some setup. I know I often use a stand for my laptop and would like to change the font size often.https://framagit.org/medoc92/recoll/-/issues/67MINOR add a "added date" and "file size" column to the webcache editor2021-03-22T15:09:02ZthiswillbeyourgithubMINOR add a "added date" and "file size" column to the webcache editorThe webcache editor should have a "added date" columns., this would help to find things. It should also contain a "file size" as some sites can be way larger than others in html.The webcache editor should have a "added date" columns., this would help to find things. It should also contain a "file size" as some sites can be way larger than others in html.https://framagit.org/medoc92/recoll/-/issues/68MAJOR add a way to see what's currently being index etc2021-04-02T10:02:54ZthiswillbeyourgithubMAJOR add a way to see what's currently being index etcThat's a bigger change but I dislike the fact that I can't easily see what recoll is currently indexing WHILE it is indexing. By that I mean that there is this line at the bottom of the main window that indicates the current file that is...That's a bigger change but I dislike the fact that I can't easily see what recoll is currently indexing WHILE it is indexing. By that I mean that there is this line at the bottom of the main window that indicates the current file that is being indexed but I don't see a way to see the whole list. I think it would be good to have an arrow at the right side of this line that when clicked on opens a window that shows the list of files being indexed in real time, with automatic scrolling but ability to scroll up if needed. I would go as far as to suggest to color in red the files that were not indexed because too slow, in orange the files that were slow to index, and in grey the files that were ignored because matching a rule like "extension is X" or "contains Y in the name" etc. Because as far as I saw there is no way to see if a file has been indexed apart from querying it in recoll. It makes me a bit anxious as I can't see a full list of indexed files and see if one of my matching rules was wrong. It would also show what's being added from the webcache, or removed if a page is being updated (ie was already present in the websote and was replaced by a hole).https://framagit.org/medoc92/recoll/-/issues/69MEDIUM keep all versions of the webcache editor2021-03-25T16:43:11ZthiswillbeyourgithubMEDIUM keep all versions of the webcache editorI think it would be good to have a way to specify if one wants to keep all versions or only the first of a webpage. Recently one of my favorite blog shutdown so all the pages were replaced with a sort of warning. I'd hate to have this re...I think it would be good to have a way to specify if one wants to keep all versions or only the first of a webpage. Recently one of my favorite blog shutdown so all the pages were replaced with a sort of warning. I'd hate to have this replace my past copies. And the "Web history" panel is quite empty anyway :p.
Removing those "holes" you mentionned by mail to keep cache size small if the user doesn't want to keep all versions would be good also anway.
Actually what would make more sense to me than archiving each time we visit the page would be to incorporate the new cached version ONLY if the date of browsing is different than the last time I was on the site. Effectively caching website only once per day at maximum. That should be a reasonable compromise between archiving a page 10 times when you just reload the page to see if your internet connection is working or whatever and actually caching something of value. This would also make use of the new date column that I suggested.https://framagit.org/medoc92/recoll/-/issues/70MINOR indicate current store size in the web history panel2021-03-23T09:19:47ZthiswillbeyourgithubMINOR indicate current store size in the web history panelThe web history panel of the index configuration should tell the user the current used size of the webcache, for example like so : "Max. size for the web store (in MB, currently used space is XXMB)"The web history panel of the index configuration should tell the user the current used size of the webcache, for example like so : "Max. size for the web store (in MB, currently used space is XXMB)"https://framagit.org/medoc92/recoll/-/issues/71BUG on the columns "date", "date and time" etc2021-03-23T08:34:34ZthiswillbeyourgithubBUG on the columns "date", "date and time" etcIn the latest ubuntu version, in the main window, after a query, I don't understand why there are 2 columns : "Date" and "Date and time", the latter being in unredered html (ie, contains "&nbsp" etc).
Similarly, I don't understand the d...In the latest ubuntu version, in the main window, after a query, I don't understand why there are 2 columns : "Date" and "Date and time", the latter being in unredered html (ie, contains " " etc).
Similarly, I don't understand the difference between the column "File Size" and "Document size". I think the names are not very intuitive. Is it refering to "archived size" and "original file size" ? If so I think those are better names. Same for "Document date" and "file date" columns, what is the unit? Is the the modification date or the creating date?https://framagit.org/medoc92/recoll/-/issues/72Use aspell with several languages?2021-03-25T16:10:07ZthiswillbeyourgithubUse aspell with several languages?Hi,
I'm confused about aspell. Am I correct in my understanding that, if checked in the settings, aspell will be used to generate alternate spellings from the queries and if UNchecked recoll with generate them itself?
I don't understan...Hi,
I'm confused about aspell. Am I correct in my understanding that, if checked in the settings, aspell will be used to generate alternate spellings from the queries and if UNchecked recoll with generate them itself?
I don't understand the relationship between aspell and stemmings. Especially, I find it confusing to be able to specify multiple stemming languages but only 1 language for aspell.
If you can spare some light on this that would be helpful to me :)https://framagit.org/medoc92/recoll/-/issues/73Exported .bas .frm .frx files from Visual Basic for Applications content not ...2021-03-25T17:04:25ZdonxrExported .bas .frm .frx files from Visual Basic for Applications content not indexing, or not showing in results correctlyExported .bas files from Visual Basic for Applications content not indexing, or not showing in results correctly. I've check the include/exclude filetypes.
The filenames are being indexed, but not their contents.
The files open eas...Exported .bas files from Visual Basic for Applications content not indexing, or not showing in results correctly. I've check the include/exclude filetypes.
The filenames are being indexed, but not their contents.
The files open easily in Notepad++, so I'm not sure what configuration options need to change.
The filetype is not listed in the MIMEs list to include, but i don't know what to add there.
VBA also exports .frm and .frx and .cls filetypes. but I think .frm or .frx may be partly binary.https://framagit.org/medoc92/recoll/-/issues/74How can I set my default file manager to PCManFM instead of Dolphin file mana...2021-04-06T04:34:22ZRaytleHow can I set my default file manager to PCManFM instead of Dolphin file manager?When, I right click on a result and choose Open Parent Folder then the Dolphin file manager opens. I would prefer PCManFM. How can I set my default file manager to PCManFM instead of Dolphin file manager?When, I right click on a result and choose Open Parent Folder then the Dolphin file manager opens. I would prefer PCManFM. How can I set my default file manager to PCManFM instead of Dolphin file manager?https://framagit.org/medoc92/recoll/-/issues/75Display only text contents in result table (not "Preview Open" and file size)2021-04-13T07:51:31ZRaytleDisplay only text contents in result table (not "Preview Open" and file size)Congratulations on new Recoll design. It is much more easy for me. Thanks to you sir! With your permission I make humble request.
I only want to see text contents for result table. The other Recoll format (Preview Open and file size) no...Congratulations on new Recoll design. It is much more easy for me. Thanks to you sir! With your permission I make humble request.
I only want to see text contents for result table. The other Recoll format (Preview Open and file size) not useful for me. I do not want to see Preview Open and file size.
If not too much trouble for you, I humbly ask if you can create an option to show text contents by default for result table. Also, when I float my mouse over search results of result table I want see text contents. I do not want to see Preview Open and file size. Also, when I press ctrl+0 I want see text contents. I mean, when I press ctrl+0 I do not want to need to click with my mouse on a result to toggle from Preview Open and file size to display text contents. In conclusion, I never want see Preview Open and file size.
Recoll is much more better now for me. Thanks again to you sir!https://framagit.org/medoc92/recoll/-/issues/76indexing fails without specific message2021-04-14T18:29:33ZWolfgang Sprickindexing fails without specific messageLinux Mint 19.3 Tricia
recoll 1.29.2 + xapian 1.4.5
Indexing starts normal but seems to stop at some file. Later ends with unspecific error message. See Screenshots encloses. Also find enclosed file for stderr und inxing-logfile (last ...Linux Mint 19.3 Tricia
recoll 1.29.2 + xapian 1.4.5
Indexing starts normal but seems to stop at some file. Later ends with unspecific error message. See Screenshots encloses. Also find enclosed file for stderr und inxing-logfile (last lines).
Greetings Wolfgang[recoll_indexer_last_lines.log](/uploads/c81da038d2dfed0261e43d8d2fc63293/recoll_indexer_last_lines.log)[stderr.log](/uploads/4bc7d175a3970fb53d3823aa7b9b9f6d/stderr.log)![recoll_fails](/uploads/d7888c7083b8d98b4c3ac860e55dfe2f/recoll_fails.png)
![recoll_stops](/uploads/7fa8b0e4ed37fbd1948df9aea3a9ca57/recoll_stops.png)https://framagit.org/medoc92/recoll/-/issues/77When press Ctrl + 0 I no see preview2021-04-13T07:51:19ZRaytleWhen press Ctrl + 0 I no see previewWhen press Ctrl + 0 I no see any preview in result table detail area. (The preview in result table detail area is blank).
But when press Ctrl + 1 yes, I see preview in result table detail area.
Also when press Ctrl + 2 yes, I see previe...When press Ctrl + 0 I no see any preview in result table detail area. (The preview in result table detail area is blank).
But when press Ctrl + 1 yes, I see preview in result table detail area.
Also when press Ctrl + 2 yes, I see preview in result table detail area.
Also when press Ctrl + 3 yes, I see preview in result table detail area.https://framagit.org/medoc92/recoll/-/issues/78I press Ctrl + g (Copy current result text to clipboard) but result is a litt...2021-04-14T13:47:55ZRaytleI press Ctrl + g (Copy current result text to clipboard) but result is a little wrongWhen a text document contain,
`&ref=seller_location%3AAL%2CAR%2CBA%2CBG%2CBO%2CBR%2CCL%2CCO%2CCR%2CDO%2CEC%2CGT%2CHN%2CME%2CMK%2CMX%2CPE%2CPR%2CPY%2CRO%2CRS%2CUY%2CVE%2CXK`
then I press Ctrl + g (Copy current result text to clipboard) ...When a text document contain,
`&ref=seller_location%3AAL%2CAR%2CBA%2CBG%2CBO%2CBR%2CCL%2CCO%2CCR%2CDO%2CEC%2CGT%2CHN%2CME%2CMK%2CMX%2CPE%2CPR%2CPY%2CRO%2CRS%2CUY%2CVE%2CXK`
then I press Ctrl + g (Copy current result text to clipboard) but result is a little wrong. Recoll prepends carriage returns to text I copy.https://framagit.org/medoc92/recoll/-/issues/79I would like new option named "Copy File Path/File Name"2021-04-16T07:18:21ZRaytleI would like new option named "Copy File Path/File Name"Please sir, another request.
When I right mouse click on file in search results, I want new option named "Copy File Path/File Name."
Now I use too many steps. Now I must "Copy File Name" then "Copy File Path". Then I pastes "Copy File ...Please sir, another request.
When I right mouse click on file in search results, I want new option named "Copy File Path/File Name."
Now I use too many steps. Now I must "Copy File Name" then "Copy File Path". Then I pastes "Copy File Path" then I pastes (appends) "Copy File Name".https://framagit.org/medoc92/recoll/-/issues/80Not remembering my custom shortcuts2021-04-28T01:28:09ZRaytleNot remembering my custom shortcutsHello sir! Please helps me with a small problem. I changed Clear search from ctrl+s (default) to ctrl+f (custom shortcut) Then ctrl+f works good. No problems. But when I exits Recoll and then restarts Recoll, ctrl+f no work. Instead Clea...Hello sir! Please helps me with a small problem. I changed Clear search from ctrl+s (default) to ctrl+f (custom shortcut) Then ctrl+f works good. No problems. But when I exits Recoll and then restarts Recoll, ctrl+f no work. Instead Clear search go back to ctrl+s (default).https://framagit.org/medoc92/recoll/-/issues/81No valid error when a user tries to open a zip file or unknown format eventho...2021-08-11T07:49:03Zjavadevelopment4No valid error when a user tries to open a zip file or unknown format eventhough preview is possible or provided as link.![image](/uploads/4a2d57224254d2d948c3e471c32264ef/image.png)![image](/uploads/4a2d57224254d2d948c3e471c32264ef/image.png)https://framagit.org/medoc92/recoll/-/issues/82Full file name is not getting displayed in the preview..it would be better if...2021-06-19T07:57:05Zjavadevelopment4Full file name is not getting displayed in the preview..it would be better if you could increase the size of the tab name or leave it to the file name![image](/uploads/ed3a933a9d52e475df9e43ef325ee3ce/image.png)![image](/uploads/ed3a933a9d52e475df9e43ef325ee3ce/image.png)https://framagit.org/medoc92/recoll/-/issues/83keyword is not highlighted when choosed text and preview2021-05-26T06:34:17Zjavadevelopment4keyword is not highlighted when choosed text and preview![image](/uploads/08c9bdc65d190734b227e0542d289b1f/image.png)![image](/uploads/08c9bdc65d190734b227e0542d289b1f/image.png)https://framagit.org/medoc92/recoll/-/issues/84On clicking open option, it throws the below error2021-08-11T07:49:23Zjavadevelopment4On clicking open option, it throws the below error![image](/uploads/bbc307e353c646851e4844a0c4a7ed3a/image.png)![image](/uploads/bbc307e353c646851e4844a0c4a7ed3a/image.png)https://framagit.org/medoc92/recoll/-/issues/85Small feature ... Copy current result text to clipboard + quit2021-08-30T11:56:48ZRaytleSmall feature ... Copy current result text to clipboard + quitCopy current result text to clipboard work good for me. Thanks to you sir!
My request is similar to current feature with name of, Open current result and quit.
I want to save one small step. After I copy current result text to clipboar...Copy current result text to clipboard work good for me. Thanks to you sir!
My request is similar to current feature with name of, Open current result and quit.
I want to save one small step. After I copy current result text to clipboard I do not want to manually quit Recoll.
Copy current result text to clipboard + quit feature will save me that small step.https://framagit.org/medoc92/recoll/-/issues/86recollindex "indexing failed" - "Data ran out unexpectedly when reading posti...2021-12-18T18:15:34ZEgon Schielerecollindex "indexing failed" - "Data ran out unexpectedly when reading posting list"I started indexing with `recollindex -d -z` to recreate the index.
Stopped a few times using Ctrl+c (and waited for the process to terminate cleanly each time).
Continued indexing with `recollindex -d`.
On the last run of recollindex th...I started indexing with `recollindex -d -z` to recreate the index.
Stopped a few times using Ctrl+c (and waited for the process to terminate cleanly each time).
Continued indexing with `recollindex -d`.
On the last run of recollindex that then failed, it displayed the following message hundreds/thousands times:
```
:2:rcldb/rcldb.cpp:2130::Db::needUpdate: get_document error: Data ran out unexpectedly when reading posting list
```
The last messages right before the process terminated were:
```:2:rcldb/rcldb.cpp:2130::Db::needUpdate: get_document error: Data ran out unexpectedly when reading posting list
:2:rcldb/rcldb.cpp:2027::Db::doFlush: flush() failed: EOF reading block 4294967295
:2:rcldb/rcldb.cpp:206::DbUpdWorker: xxWrite failed
:2:rcldb/rcldb.cpp:2130::Db::needUpdate: get_document error: Data ran out unexpectedly when reading posting list
:2:./utils/workqueue.h:116::WorkQueue::put:DbUpd: !ok
:2:rcldb/rcldb.cpp:1891::Db::addOrUpdate:Cant queue task
:2:index/fsindexer.cpp:541::FsIndexerDbUpdWorker: addOrUpdate failed
:2:./utils/workqueue.h:116::WorkQueue::put:Split: !ok
:2:index/fsindexer.cpp:637::processonefile: wqueue.put failed
:2:index/fsindexer.cpp:565::FsIndexerInternfileWorker: processone failed
:2:index/fsindexer.cpp:234::FsIndexer::index: error while indexing /home/user:
:2:./utils/workqueue.h:116::WorkQueue::put:Split: !ok
:2:index/fsindexer.cpp:637::processonefile: wqueue.put failed
:2:index/fsindexer.cpp:565::FsIndexerInternfileWorker: processone failed
Indexing failed
```https://framagit.org/medoc92/recoll/-/issues/87When I install Recoll on a new machine it not remember Top Directories2021-06-13T07:46:24ZRaytleWhen I install Recoll on a new machine it not remember Top DirectoriesWhen I install Recoll on a new machine it not remember Top Directories (under Global Parameters).When I install Recoll on a new machine it not remember Top Directories (under Global Parameters).https://framagit.org/medoc92/recoll/-/issues/88feature request: add real time indexing support for windows2022-03-12T07:57:28Zleoliufeature request: add real time indexing support for windowsthank you for the great Recoll, ever best desktop search so far, compared with dozon of similiar ones
Real time index is a killer feature, and expect to get it implemented on windows (at least win10 since its mainstream nowadays)
BTW, ...thank you for the great Recoll, ever best desktop search so far, compared with dozon of similiar ones
Real time index is a killer feature, and expect to get it implemented on windows (at least win10 since its mainstream nowadays)
BTW, while i get from doc: real time index is based on inotify, i found no clue on why its not implemented on windows, in case we can get inotify function on windowshttps://framagit.org/medoc92/recoll/-/issues/89Feature request: convert rclmon.sh to a systemd service file2021-07-02T01:33:28ZGeorger AraújoFeature request: convert rclmon.sh to a systemd service fileHi,
Please consider converting rclmon.sh to a systemd service file so that it can be used with e.g. Ubuntu Server.Hi,
Please consider converting rclmon.sh to a systemd service file so that it can be used with e.g. Ubuntu Server.https://framagit.org/medoc92/recoll/-/issues/90Go to the exact match option after the search results2021-06-19T07:18:43Zjavadevelopment4Go to the exact match option after the search resultsI indexed the folder and searched for a keyword. The published results are fine..
On top it says - Documents 1-8 out of at least 433
If i want to go 320....I need to click several times..It would be great..if we have a go to exact matc...I indexed the folder and searched for a keyword. The published results are fine..
On top it says - Documents 1-8 out of at least 433
If i want to go 320....I need to click several times..It would be great..if we have a go to exact match option or pagination page ..Paginate to 10th or 12...Along with the next, previous option.https://framagit.org/medoc92/recoll/-/issues/91Docs request: debug build2021-08-30T11:45:47ZrhnDocs request: debug buildI tried to debug recoll a bit, but I was unable to attach a debugger meaningfully. The default build seems to be stripped of debug symbols:
```
$ QMAKE=qmake-qt5 ./configure --disable-userdoc
$ make
$ gdb qtgui/recoll
GNU gdb (GDB) Fed...I tried to debug recoll a bit, but I was unable to attach a debugger meaningfully. The default build seems to be stripped of debug symbols:
```
$ QMAKE=qmake-qt5 ./configure --disable-userdoc
$ make
$ gdb qtgui/recoll
GNU gdb (GDB) Fedora 10.1-4.fc33
[...]
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from qtgui/recoll...
(No debugging symbols found in qtgui/recoll)
(gdb)
```
Could the documentation gain a section explaining how to get a debug build?https://framagit.org/medoc92/recoll/-/issues/92I would prefer to manually choose one of the previous searches which Recoll p...2021-06-13T07:09:42ZRaytleI would prefer to manually choose one of the previous searches which Recoll presentsIn, [1.2. Full text search](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html) I read this...
> In many cases, you are looking for all the forms of a word, including plurals, different tenses for a verb, or terms derived ...In, [1.2. Full text search](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html) I read this...
> In many cases, you are looking for all the forms of a word, including plurals, different tenses for a verb, or terms derived from the same root or stem (example: floor, floors, floored, flooring...). Queries are usually automatically expanded to all such related terms (words that reduce to the same stem). **This can be prevented for searching for a specific form.** [Emphasis added.]
Imagine my **first** search is,
1. I open Recoll.
1. I type `Ctrl+s`.
1. I input, "A little foo and some more foo.txt".
1. I press `Enter`.
1. Then Recoll finds, "A little foo and some more foo.txt".
1. This is good. That is what I expect and that is what I want.
1. I exit out of Recoll.
Now imagine my **second** search is,
1. I open Recoll.
1. I type `Ctrl+s`.
1. I input, "foo".
1. I press `Enter`.
1. Then Recoll automatically fills in, "A little foo and some more foo.txt" and searches for, "A little foo and some more foo.txt"
1. This is bad. I did not want this. My command was search for "foo". My command was not search for "A little foo and some more foo.txt"
My current **workaround** is,
1. I type `Ctrl+s`.
1. I input, "foo".
1. I enter, "!" after "foo" so that I my input is, "foo!"
1. I press `Backspace` to delete "!".
1. I press `Enter`.
1. Then Recoll finds all files that contain, "foo".
My **proposed solution** is,
1. I open Recoll.
1. I type `Ctrl+s`.
1. I input, "foo".
1. I press `Enter`.
1. Then Recoll searches for "foo".
In cases when I want to Recoll to reuse a previous search, I suggest that I must **manually** choose one of the previous searches which Recoll presents.https://framagit.org/medoc92/recoll/-/issues/93Windows: Recoll 1.31 crashing2024-01-12T14:32:37Zeu2021Windows: Recoll 1.31 crashingHi,
I just started using Recoll on Windows, but it crashes if I try to open Index Configuration or if I try to Stop Indexing.
Thanks,Hi,
I just started using Recoll on Windows, but it crashes if I try to open Index Configuration or if I try to Stop Indexing.
Thanks,https://framagit.org/medoc92/recoll/-/issues/94Anchored searches using distance parameter seem to exhibit off-by-one error2021-06-28T07:00:25ZFeRD (Frank Dana)Anchored searches using distance parameter seem to exhibit off-by-one errorUsing recoll 1.31 from the Fedora distribution repository, it seems as though the maximum-distance parameter of anchored searches is slightly more forgiving than one would expect.
Imagine a database that indexes documents with all of th...Using recoll 1.31 from the Fedora distribution repository, it seems as though the maximum-distance parameter of anchored searches is slightly more forgiving than one would expect.
Imagine a database that indexes documents with all of the following titles:
1. Where the Wild Things Are
1. Where the Wild Things Were
1. Where the Wild Things Should Be
1. Max : Where the Wild Things Are
1. Goat Thing : Where the Wild Things Are
1. Kind of Rhinoceros Guy : Where the Wild Things Are
A search for:
* `title:"^where the wild"` : Will match all of the titles, as expected
* `title:"^where the wild"o0` : Will match the first three (as expected), _and_ the fourth
* `title:"^where the wild"o1` : Will match _all_ but the last title
Effectively, a distance of "zero words" from the start of the text seems to be interpreted as "up to 1 word", "one word" comes out as "up to 2 words", and so on... As a result, there doesn't seem to be any way to _truly_ anchor the search to the start of the text, and match **only** the first three example titles.
The same is true for searches anchored to the end of text, i.e. `title:"wild things$"o0` will match all of the titles on the list except the third.https://framagit.org/medoc92/recoll/-/issues/95Colors by file extension (Feature request)2022-07-03T20:48:39ZRaytleColors by file extension (Feature request)[Colors by file extension](https://dystroy.org/broot/conf_file/#colors-by-file-extension)[Colors by file extension](https://dystroy.org/broot/conf_file/#colors-by-file-extension)https://framagit.org/medoc92/recoll/-/issues/96Themes (Feature request)2023-02-22T01:49:11ZRaytleThemes (Feature request)[Themes](https://dystroy.org/broot/conf_file/#syntax-theme). They look like this [Syntect screenshots](https://github.com/trishume/syntect#screenshots).[Themes](https://dystroy.org/broot/conf_file/#syntax-theme). They look like this [Syntect screenshots](https://github.com/trishume/syntect#screenshots).https://framagit.org/medoc92/recoll/-/issues/97Image preview using Kitty (Feature request)2021-07-04T15:43:21ZRaytleImage preview using Kitty (Feature request)[Preview files](https://github.com/Canop/broot#preview-files).
Please pay attention to this: **"If your terminal is kitty, then the image preview is high definition..."**[Preview files](https://github.com/Canop/broot#preview-files).
Please pay attention to this: **"If your terminal is kitty, then the image preview is high definition..."**https://framagit.org/medoc92/recoll/-/issues/98Open a file and quit Recoll (Feature request)2021-08-27T09:18:30ZRaytleOpen a file and quit Recoll (Feature request)This is similar to issue #85
I copied this idea from, "...[hit alt enter](https://github.com/Canop/broot#never-lose-track-of-file-hierarchy-while-you-search) to open it in your system's default program and close broot..."This is similar to issue #85
I copied this idea from, "...[hit alt enter](https://github.com/Canop/broot#never-lose-track-of-file-hierarchy-while-you-search) to open it in your system's default program and close broot..."https://framagit.org/medoc92/recoll/-/issues/99Doubt of markdown files indexing2021-07-02T00:04:55ZAdrien AndréDoubt of markdown files indexingHi, with Recoll 1.23.7 (Xapian 1.4.5) on Ubuntu 18.04.5 `.md` files might not be indexed.
Not sure if there is a way to force it.
Does someone confirm this?Hi, with Recoll 1.23.7 (Xapian 1.4.5) on Ubuntu 18.04.5 `.md` files might not be indexed.
Not sure if there is a way to force it.
Does someone confirm this?https://framagit.org/medoc92/recoll/-/issues/100Allow users to limit their searches to a particular directory (Feature Request)2022-03-13T05:03:34ZRaytleAllow users to limit their searches to a particular directory (Feature Request)I reviewed from [3.2.8. Complex/advanced search](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html#RCL.SEARCH.GUI.COMPLEX) until [3.2.10. Multiple indexes](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html#R...I reviewed from [3.2.8. Complex/advanced search](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html#RCL.SEARCH.GUI.COMPLEX) until [3.2.10. Multiple indexes](https://www.lesbonscomptes.com/recoll/usermanual/usermanual.html#RCL.SEARCH.GUI.MULTIDB), but I did not see the following feature.
I like Recoll very much. I use it throughout my workday. However, I am often overwhelmed with search results. To decrease the number of search results I receive, I propose the following method.
First, I imagine I would create an abbreviation for each directory which I had added to "Index Configuration -> Global Parameters." For example, I might create an abbreviation **"docs"** for a directory of mine with the path `/home/documents.` Also, for example, I might create an abbreviation "**pscripts**" for a directory of mine with the path `/home/python_scripts.`
Second, when I search on Recoll I would have the option to type **`docs:`** or **`pscripts:`** to limit my search to `/home/documents` or `/home/python_scripts.` Therefore, to limit my search for the term "tkinter" to only `/home/documents` I would type the following into the search box: **`docs:tkinter`**; whereas to limit my search for the term "tkinter" to only `/home/python_scripts` I would type the following into the search box: **`pscripts:tkinter`**. Of course, to limit a search to both **`docs:`** and **`pscripts:`** I imagine that I would enter something like, **`docs,pscripts:tkinter`**
What I indicated above is essentially what I want. The following adds some clarification and a refinement.
In other words, my current problem is this: when I search for a term such as `tkinter` I often see too many search results. As a result, I often think to myself something like, "I know which directory I want to search, but I don't know an easy way to command Recoll to only search that specific directory."
I think that it would be nice, but not necessary, for users to be able to limit searches to a particular sub-directory within each directory which a user had added to "Index Configuration -> Global Parameters." For example, if `/home/python_scripts/` were listed under "Index Configuration -> Global Parameters" then it would be nice to be able to limit searches to `/home/python_scripts/parsing_scripts` instead of only being able to limit searches to `/home/python_scripts/`.
Finally, imagine you had misplaced your wallet. You almost certainly wouldn't look through all of the pockets in all of your pants and jackets. Instead, you would very probably limit your search to pants and jackets you had worn within the last, say, approximately 12 to 24 hours. Of course, I imagine that you wouldn't bother searching in pants and jackets you had not worn for, say, two months. Similarly, if you had a closet which you had not opened for over a week (more than a week), you probably wouldn't bother opening that closet to search for your lost wallet.https://framagit.org/medoc92/recoll/-/issues/101Using my keyboard to view the drop-down list of items I recently searched for...2021-08-31T08:21:55ZRaytleUsing my keyboard to view the drop-down list of items I recently searched for (Feature request)I would like to be able to press, for example, the tab key followed by the enter key to view the drop-down list of items I recently searched for. Currently I need to use my mouse to accomplish this.
Please see:
![Image_1](/uploads/b268...I would like to be able to press, for example, the tab key followed by the enter key to view the drop-down list of items I recently searched for. Currently I need to use my mouse to accomplish this.
Please see:
![Image_1](/uploads/b26813a54ea15f8f922e9b91228e3d6a/Image_1.png)https://framagit.org/medoc92/recoll/-/issues/102Recoll on Windows 10, Microsoft Outlook PST-Files, Date2021-08-07T11:27:46ZEberhard OpitzRecoll on Windows 10, Microsoft Outlook PST-Files, DateI'm using Recoll on Windows 10 Enterprise version 1909 build 18363.1646. Indexing of **Outlook-PST-files** works great, but Mails in the result list are shown with the **date of indexing** and not with their **real date** (see attached f...I'm using Recoll on Windows 10 Enterprise version 1909 build 18363.1646. Indexing of **Outlook-PST-files** works great, but Mails in the result list are shown with the **date of indexing** and not with their **real date** (see attached file; real mail-date is "29.4.2020")![Recoll_screenshot](/uploads/10aff7ff3b0c68207fdd59281713b851/Recoll_screenshot.png). What's wrong?