1. Forum
  2. Usenet
  3. UK.COMP.OS.LINUX

  • Specifying an action to copy a file from within feh

    From Java Jive@2:250/1 to All on Friday, March 29, 2024 22:54:07
    As some here may recall, I'm in the midst of several years' work
    scanning family documents, post-processing them, and then releasing them either to the general public for generations sufficiently removed from
    living people, or else just back to the source branch of the family.
    Currently there are nearly 15,000 original documents with equal numbers
    of post-processed versions arranged as below, but around 7,000 of them completed in previous stages already have final documents, so there are
    around 8,000 newly added documents left to do. What I need to do is use
    a program, feh seems a likely choice, to work through those documents
    that don't yet have a final version, choose one from five possible
    versions described below, copy it to its corresponding final directory,
    and move on to the next.

    The documents are arranged as follows:

    Family History
    /_all/1o/subs Original scans
    /_all/2n/subs Normalised versions of the originals
    /_all/3t/subs Textcleaned versions ditto
    /_all/4nt/subs Normalised then textcleaned versions ditto
    /_all/5tn/subs Textcleaned then normalised versions ditto
    /subs Final docs chosen from the best of the above

    Example for one particular subdirectory:

    Family History/_all/1o/Alston
    Family History/_all/2n/Alston
    Family History/_all/3t/Alston
    Family History/_all/4nt/Alston
    Family History/_all/5tn/Alston
    Family History/Alston

    I've tried the following command line for feh (sorry can't help any line wrap):

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    .... which results in ...

    cp -a "Family History/_all/1o/Alston/Alston Descent - 2nd Earl Of
    Montrose, William - 1 (o).png" "Family History/Alston/Alston Descent -
    2nd Earl Of Montrose, William - 1.png"

    .... which looks as though it should do the job, however removing the
    echo ...

    user@computer:~# feh -Fdr -A 'cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    .... results in ...

    cp: target '1.png"' is not a directory

    Can anyone suggest a magical incantation which will yield the desired
    result?

    Another problem is that I'm going to have order the files so that each
    of the 5 possible versions are displayed successively so that I can use
    the arrow keys to go back and forth between them to choose the best. I
    think I might be able to use a file list to do this, but as yet I
    haven't given much thought as to how to create it, it would have to be
    of the form:

    Family History/_all/1o/Alston/Document 1 (o).png
    Family History/_all/2n/Alston/Document 1 (n).png
    Family History/_all/3t/Alston/Document 1 (t).png
    Family History/_all/4nt/Alston/Document 1 (nt).png
    Family History/_all/5tn/Alston/Document 1 (tn).png
    Family History/_all/1o/Alston/Document 2 (o).png
    Family History/_all/2n/Alston/Document 2 (n).png
    Family History/_all/3t/Alston/Document 2 (t).png
    Family History/_all/4nt/Alston/Document 2 (nt).png
    Family History/_all/5tn/Alston/Document 2 (tn).png
    etc.

    I could also use Windows if there's a better solution available there,
    but that seems unlikely to me.

    Any help gratefully received.

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk

    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, March 30, 2024 01:04:53
    First problem solved ...

    On 29/03/2024 22:54, Java Jive wrote:
    As some here may recall, I'm in the midst of several years' work
    scanning family documents, post-processing them, and then releasing them either to the general public for generations sufficiently removed from living people, or else just back to the source branch of the family. Currently there are nearly 15,000 original documents with equal numbers
    of post-processed versions arranged as below, but around 7,000 of them completed in previous stages already have final documents, so there are around 8,000 newly added documents left to do.  What I need to do is use
    a program, feh seems a likely choice, to work through those documents
    that don't yet have a final version, choose one from five possible
    versions described below, copy it to its corresponding final directory,
    and move on to the next.

    The documents are arranged as follows:

    Family History
        /_all/1o/subs    Original scans
        /_all/2n/subs    Normalised versions of the originals
        /_all/3t/subs    Textcleaned versions ditto
        /_all/4nt/subs    Normalised then textcleaned versions ditto
        /_all/5tn/subs    Textcleaned then normalised versions ditto
        /subs        Final docs chosen from the best of the above

    Example for one particular subdirectory:

    Family History/_all/1o/Alston
    Family History/_all/2n/Alston
    Family History/_all/3t/Alston
    Family History/_all/4nt/Alston
    Family History/_all/5tn/Alston
    Family History/Alston

    I've tried the following command line for feh (sorry can't help any line wrap):

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    .... which results in ...

    cp -a "Family History/_all/1o/Alston/Alston Descent - 2nd Earl Of
    Montrose, William - 1 (o).png" "Family History/Alston/Alston Descent -
    2nd Earl Of Montrose, William - 1.png"

    .... which looks as though it should do the job, however removing the
    echo ...

    user@computer:~# feh -Fdr -A 'cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    .... results in ...

    cp: target '1.png"' is not a directory

    Can anyone suggest a magical incantation which will yield the desired result?

    I solved this by doing the path conversion in a one-line shell script,
    so the feh command line now reads ...

    feh -Fdr -A '/home/Family\ History/_all/FHCopyIn.sh %F' 'Family History/_all/1o'

    .... while the one-line shell script reads ...

    cp -a "$1" "$(echo $1 | sed -E '{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }' )"

    Another problem is that I'm going to have order the files so that each
    of the 5 possible versions are displayed successively so that I can use
    the arrow keys to go back and forth between them to choose the best.  I think I might be able to use a file list to do this, but as yet I
    haven't given much thought as to how to create it, it would have to be
    of the form:

    Family History/_all/1o/Alston/Document 1 (o).png
    Family History/_all/2n/Alston/Document 1 (n).png
    Family History/_all/3t/Alston/Document 1 (t).png
    Family History/_all/4nt/Alston/Document 1 (nt).png
    Family History/_all/5tn/Alston/Document 1 (tn).png
    Family History/_all/1o/Alston/Document 2 (o).png
    Family History/_all/2n/Alston/Document 2 (n).png
    Family History/_all/3t/Alston/Document 2 (t).png
    Family History/_all/4nt/Alston/Document 2 (nt).png
    Family History/_all/5tn/Alston/Document 2 (tn).png
    etc.

    I could also use Windows if there's a better solution available there,
    but that seems unlikely to me.

    Still to do.

    Any help gratefully received.

    Stet

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk

    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, March 30, 2024 11:20:47
    On 30/03/2024 01:04, Java Jive wrote:

    First problem solved ...

    On 29/03/2024 22:54, Java Jive wrote:

    Another problem is that I'm going to have order the files so that each
    of the 5 possible versions are displayed successively so that I can
    use the arrow keys to go back and forth between them to choose the
    best.  I think I might be able to use a file list to do this, but as
    yet I haven't given much thought as to how to create it, it would have
    to be of the form:

    Family History/_all/1o/Alston/Document 1 (o).png
    Family History/_all/2n/Alston/Document 1 (n).png
    Family History/_all/3t/Alston/Document 1 (t).png
    Family History/_all/4nt/Alston/Document 1 (nt).png
    Family History/_all/5tn/Alston/Document 1 (tn).png
    Family History/_all/1o/Alston/Document 2 (o).png
    Family History/_all/2n/Alston/Document 2 (n).png
    Family History/_all/3t/Alston/Document 2 (t).png
    Family History/_all/4nt/Alston/Document 2 (nt).png
    Family History/_all/5tn/Alston/Document 2 (tn).png
    etc.

    I've solved this part now as well. The following command does it
    (originally all one line):

    find 'Family History/_all' -name "*.png" | sort | sed -E '{ h; s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ ; G ; s/\n/\t/ }' |
    sort | sed 's/^[^\t]\+\t//' > 'Family History/_all/FHBlinkList.txt'

    The most complex part is the first sed command, so I will explain it:

    h Copy input, the original full pathname, to the hold space
    s 2x Remove the '/_all/*' and ' (*)' parts of the pathname
    G Retrieve and append the original line from the hold space
    s Replace the intervening new line character with a tab

    The second sed command removes the first part of the line up to and
    including the tab which was used as a key in the final sort immediately previously.

    However, although the problems as originally stated are now both solved,
    I need to do this second stage better, as I need to exclude from the
    list those files that already have a chosen final copy in the main part
    of the archive.

    I could also use Windows if there's a better solution available there,
    but that seems unlikely to me.

    Still to do.

    Any help gratefully received.

    Stet

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk

    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Saturday, March 30, 2024 13:06:32
    Subject: Long lines [Was: Re: Specifying an action to copy a file from within
    feh]

    On 2024-03-29 23:54, Java Jive wrote:
    I've tried the following command line for feh (sorry can't help any line wrap):

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    There is an addon for Thunderbird that allows writing long lines.

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'


    Unfortunately, it applies to the entire post or email, not to paragraphs.


    Toggle Line Wrap
    Toggle line wrapping in the message composition window.

    Author Jan Kiszka


    Toggle wrapping of long lines in the message composition window when writing plaintext messages. Wrapping can be controlled via a toolbar button and a keyboard shortcut (CTRL-SHIFT-W by default). The current state is visualized in the toolbar button. The shortcut can be adjusted via the Add-on Manager (click on the gear).

    On order to use the add-on, turn off HTML composition (per account setting, Composition & Addressing → Composition → Compose messages in HTML format) and also make sure that "mailnews.send_plaintext_flowed" is disabled in the configuration (about:config). Furthermore, mailnews.wraplength needs to be non-zero, or the add-on will be deactivated (toolbar button grayed out).

    Note that this add-on requires an WebExtension Experiment API, thus will request "full access" on installation.




    --
    Cheers, Carlos.


    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, March 30, 2024 14:36:46
    Subject: Re: Long lines [Was: Re: Specifying an action to copy a file from
    within feh]

    On 30/03/2024 13:06, Carlos E.R. wrote:

    On 2024-03-29 23:54, Java Jive wrote:

    I've tried the following command line for feh (sorry can't help any
    line wrap):

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E
    "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family
    History/_all/1o'

    There is an addon for Thunderbird that allows writing long lines.

    Unfortunately, it applies to the entire post or email, not to paragraphs.

    [snip]

    Thanks, but as the long code lines only wrapped at the edge of the
    window anyway, I'll give that a miss.

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk

    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Paul@2:250/1 to All on Saturday, March 30, 2024 16:06:07
    On 3/29/2024 6:54 PM, Java Jive wrote:
    As some here may recall, I'm in the midst of several years' work scanning family documents, post-processing them, and then releasing them either to the general public for generations sufficiently removed from living people, or else just back to the source branch of the family. Currently there are nearly 15,000 original documents with equal numbers of post-processed versions arranged as below, but around 7,000 of them completed in previous stages already have final documents, so there are around 8,000 newly added documents left to do.  What I need to do is use a program, feh seems a likely choice, to work through those documents that don't yet have a final version, choose one from five possible versions described below, copy it to its corresponding final directory, and move on to the next.

    The documents are arranged as follows:

    Family History
        /_all/1o/subs    Original scans
        /_all/2n/subs    Normalised versions of the originals     /_all/3t/subs    Textcleaned versions ditto     /_all/4nt/subs    Normalised then textcleaned versions ditto     /_all/5tn/subs    Textcleaned then normalised versions ditto     /subs        Final docs chosen from the best of the above

    Example for one particular subdirectory:

    Family History/_all/1o/Alston
    Family History/_all/2n/Alston
    Family History/_all/3t/Alston
    Family History/_all/4nt/Alston
    Family History/_all/5tn/Alston
    Family History/Alston

    I've tried the following command line for feh (sorry can't help any line wrap):

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    ... which results in ...

    cp -a "Family History/_all/1o/Alston/Alston Descent - 2nd Earl Of Montrose, William - 1 (o).png" "Family History/Alston/Alston Descent - 2nd Earl Of Montrose, William - 1.png"

    ... which looks as though it should do the job, however removing the echo ...

    user@computer:~# feh -Fdr -A 'cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    ... results in ...

    cp: target '1.png"' is not a directory

    Can anyone suggest a magical incantation which will yield the desired result?

    Another problem is that I'm going to have order the files so that each of the 5 possible versions are displayed successively so that I can use the arrow keys to go back and forth between them to choose the best.  I think I might be able to use a file list to do this, but as yet I haven't given much thought as to how to create it, it would have to be of the form:

    Family History/_all/1o/Alston/Document 1 (o).png
    Family History/_all/2n/Alston/Document 1 (n).png
    Family History/_all/3t/Alston/Document 1 (t).png
    Family History/_all/4nt/Alston/Document 1 (nt).png
    Family History/_all/5tn/Alston/Document 1 (tn).png
    Family History/_all/1o/Alston/Document 2 (o).png
    Family History/_all/2n/Alston/Document 2 (n).png
    Family History/_all/3t/Alston/Document 2 (t).png
    Family History/_all/4nt/Alston/Document 2 (nt).png
    Family History/_all/5tn/Alston/Document 2 (tn).png
    etc.

    I could also use Windows if there's a better solution available there, but that seems unlikely to me.

    Any help gratefully received.


    If cp had a --dry-run, you could obtain what it sees that way.
    When it doesn't have such an option, you can debug another way.

    Make yourself a "dumper" to dump the parameters seen on the command line.
    You can put "dumper" in place of "cp" and get the parameters that way.

    gcc -o dumper dumper.c

    #include<stdio.h>
    int main(int argc, char *argv[])
    {
    int i;
    printf("Number of arguments = %d\n", argc-1);
    for(i=0; i<argc; i++)
    {
    if (i == 0) {
    printf("Program name = %s\n", argv[i]);
    } else {
    printf("%s\n",argv[i]);
    }
    }
    }

    $ ./dumper
    Number of arguments = 0
    Program name = dumper

    $ ./dumper a b c
    Number of arguments = 3
    Program name = dumper
    a
    b
    c

    Your problem will likely involve "escaping" and "how many
    levels of shells, subshells, and interpreting are going on".
    You know the drill by now :-) It is the "land of the
    random number of backslashes" :-)

    Echo is both a built-in for the bash shell, as well
    as a /usr/bin/echo existing (maybe some other shell
    might need it).

    By inspection, I don't understand why your command
    is failing, but shine a flashlight on it and
    see if you can figure it out. The usage of "dumper"
    should have the same "nested shell count" as using "cp",
    that's my hope.

    Paul

    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, March 30, 2024 16:31:50
    On 30/03/2024 16:06, Paul wrote:
    On 3/29/2024 6:54 PM, Java Jive wrote:
    As some here may recall, I'm in the midst of several years' work scanning family documents, post-processing them, and then releasing them either to the general public for generations sufficiently removed from living people, or else just back to the source branch of the family. Currently there are nearly 15,000 original documents with equal numbers of post-processed versions arranged as below, but around 7,000 of them completed in previous stages already have final documents, so there are around 8,000 newly added documents left to do.  What I need to do is use a program, feh seems a likely choice, to work through those documents that don't yet have a final version, choose one from five possible versions described below, copy it to its corresponding final directory, and move on to the next.

    The documents are arranged as follows:

    Family History
        /_all/1o/subs    Original scans
        /_all/2n/subs    Normalised versions of the originals
        /_all/3t/subs    Textcleaned versions ditto
        /_all/4nt/subs    Normalised then textcleaned versions ditto
        /_all/5tn/subs    Textcleaned then normalised versions ditto
        /subs        Final docs chosen from the best of the above

    Example for one particular subdirectory:

    Family History/_all/1o/Alston
    Family History/_all/2n/Alston
    Family History/_all/3t/Alston
    Family History/_all/4nt/Alston
    Family History/_all/5tn/Alston
    Family History/Alston

    I've tried the following command line for feh (sorry can't help any line wrap):

    user@computer:~# feh -Fdr -A 'echo cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    ... which results in ...

    cp -a "Family History/_all/1o/Alston/Alston Descent - 2nd Earl Of Montrose, William - 1 (o).png" "Family History/Alston/Alston Descent - 2nd Earl Of Montrose, William - 1.png"

    ... which looks as though it should do the job, however removing the echo ...

    user@computer:~# feh -Fdr -A 'cp -a \"%F\" \"$(echo %F | sed -E "{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }" )\"' 'Family History/_all/1o'

    ... results in ...

    cp: target '1.png"' is not a directory

    Can anyone suggest a magical incantation which will yield the desired result?

    Another problem is that I'm going to have order the files so that each of the 5 possible versions are displayed successively so that I can use the arrow keys to go back and forth between them to choose the best.  I think I might be able to use a file list to do this, but as yet I haven't given much thought as to how to create it, it would have to be of the form:

    Family History/_all/1o/Alston/Document 1 (o).png
    Family History/_all/2n/Alston/Document 1 (n).png
    Family History/_all/3t/Alston/Document 1 (t).png
    Family History/_all/4nt/Alston/Document 1 (nt).png
    Family History/_all/5tn/Alston/Document 1 (tn).png
    Family History/_all/1o/Alston/Document 2 (o).png
    Family History/_all/2n/Alston/Document 2 (n).png
    Family History/_all/3t/Alston/Document 2 (t).png
    Family History/_all/4nt/Alston/Document 2 (nt).png
    Family History/_all/5tn/Alston/Document 2 (tn).png
    etc.

    I could also use Windows if there's a better solution available there, but that seems unlikely to me.

    Any help gratefully received.


    If cp had a --dry-run, you could obtain what it sees that way.
    When it doesn't have such an option, you can debug another way.

    Make yourself a "dumper" to dump the parameters seen on the command line.
    You can put "dumper" in place of "cp" and get the parameters that way.

    gcc -o dumper dumper.c

    #include<stdio.h>
    int main(int argc, char *argv[])
    {
    int i;
    printf("Number of arguments = %d\n", argc-1);
    for(i=0; i<argc; i++)
    {
    if (i == 0) {
    printf("Program name = %s\n", argv[i]);
    } else {
    printf("%s\n",argv[i]);
    }
    }
    }

    $ ./dumper
    Number of arguments = 0
    Program name = dumper

    $ ./dumper a b c
    Number of arguments = 3
    Program name = dumper
    a
    b
    c

    Your problem will likely involve "escaping" and "how many
    levels of shells, subshells, and interpreting are going on".
    You know the drill by now :-) It is the "land of the
    random number of backslashes" :-)

    Echo is both a built-in for the bash shell, as well
    as a /usr/bin/echo existing (maybe some other shell
    might need it).

    By inspection, I don't understand why your command
    is failing, but shine a flashlight on it and
    see if you can figure it out. The usage of "dumper"
    should have the same "nested shell count" as using "cp",
    that's my hope.

    Thanks Paul, I've copied the above into a folder to keep it. It's a
    clever technique which I have used before with Windows BATch files - I
    still have PrintCMD.COM in my Utilities folder! However, as regards the current problems, as already described in a follow-up post, I solved it another way, by putting the cp command in a one-line shell script.

    In fact, I've now solved all the problems and have my list. It's 42,940
    lines along representing at least 9,000 documents, but actually more
    than that, because photos, maps, sketches, water-colours, etc will not
    have textcleaned versions, only 'original' and 'normalised' versions.

    Phew! Better get to it ...

    Here is the FHBlinkList.sh script that created the list:

    #!/bin/sh
    ME="${0##*/}"
    DIRY="${0%/*}"

    # The Family History sub-directory
    FH="~/Family History/_all"

    OLDIFS=${IFS}
    IFS=$'\n'
    for A in $(find "${FH}" -name "*.png" | sort)
    do
    TARGET="$(echo $A | sed -E '{ s~/_all/[1-5][ont]{1,2}~~ ; s~ +\([ont]{1,2}\)~~ }' )"
    if [ ! -f "${TARGET}" ] || [ "${A}" -nt "${TARGET}" ]
    then
    echo -e "${TARGET}\t${A}"
    fi
    done | sort | sed 's/^[^\t]\+\t//' > "${FH}/FHBlinkList.txt" IFS=${OLDIFS}

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk

    --- MBSE BBS v1.0.8.6 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)