accessing data from json array of json arrays
25 views
Skip to first unread message
Xander Cage
Jun 13, 2023, 7:56:43 AM6/13/23
to help-cfengine
Hi,
i am reading in a inventory file in csv format with readdcsv() and get what i believe is a array of arrays.
R: From aix inventory, got data [["xlsmp.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.en_US.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.EN_US.rte","5.1.0.0","default","lslpp"],["xlsmp.aix61.rte","5.1.0.0","default","lslpp"],["xlmass.lib","6.1.0.4","default","lslpp"],["xlmass.aix53.lib","6.1.0.4","default","lslpp"],["xlmass.adt.include","6.1.0.4","default","lslpp"],["xlmass.8.1.3","8.1.3.1","default","lslpp"],["xlmass.10.1.0","10.1.0.0","default","lslpp"],["xlfrte.aix61","15.1.3.5","default","lslpp"],["xlfrte","15.1.3.5","default","lslpp"]]
now i need to extract the version number from a package (by name). tried filter() and grep() but both returned nothing.
example:
["xlfrte.aix61","15.1.3.5","default","lslpp"]
| |
package name version
test promise:
bundle agent eval_numbers {
vars:
"my_data" data => readcsv("$(sys.workdir)/state/software_packages.csv");
"aix_inventory_str" string => format("%S", my_data);
"test_filtergrep" slist => grep("^ xlfrte .*$", my_data);
#"keys_unsorted" slist => getindices("my_data");
#"keys" slist => sort(keys_unsorted, "lex");
#methods:
#"call_to_get_stuff" usebundle => get_stuff ( @(my_data), $(keys) );
reports:
"From aix inventory, got data $(aix_inventory_str)";
"Filtered result $(test_filtergrep)";
#"Keys sorted $(keys)";
}
vars:
"my_data" data => readcsv("$(sys.workdir)/state/software_packages.csv");
"aix_inventory_str" string => format("%S", my_data);
"test_filtergrep" slist => grep("^ xlfrte .*$", my_data);
#"keys_unsorted" slist => getindices("my_data");
#"keys" slist => sort(keys_unsorted, "lex");
#methods:
#"call_to_get_stuff" usebundle => get_stuff ( @(my_data), $(keys) );
reports:
"From aix inventory, got data $(aix_inventory_str)";
"Filtered result $(test_filtergrep)";
#"Keys sorted $(keys)";
}
so how to do it?
wbr
chris
Nick Anderson
Jun 13, 2023, 10:43:25 AM6/13/23
to help-cfengine
Hi, Chris
Hi,
i am reading in a inventory file in csv format with readdcsv() and get what i believe is a array of arrays.
R: From aix inventory, got data [["xlsmp.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.en_US.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.EN_US.rte","5.1.0.0","default","lslpp"], ["xlsmp.aix61.rte","5.1.0.0","default","lslpp"],["xlmass.lib","6.1.0.4","default","lslpp"],["xlmass.aix53.lib","6.1.0.4","default","lslpp"], ["xlmass.adt.include","6.1.0.4","default","lslpp"],["xlmass.8.1.3","8.1.3.1","default","lslpp"],["xlmass.10.1.0","10.1.0.0","default","lslpp"], ["xlfrte.aix61","15.1.3.5","default","lslpp"],["xlfrte","15.1.3.5","default","lslpp"]]
Yep, that's what it looks like to me.
By the way, have you seen classfiltercsv()? One of it's nice features is named columns.
now i need to extract the version number from a package (by name). tried filter() and grep() but both returned nothing.
example:
["xlfrte.aix61","15.1.3.5","default","lslpp"]
package name version
so how to do it?
What is it you want to do with the version number?
I often like to use associative arrays for to re-construct data structures.
I might access it like this:
bundle agent __main__ { methods: "eval_numbers"; } bundle agent eval_numbers { vars: #"my_data" data => readcsv("$(sys.workdir)/state/software_packages.csv"); "my_data" data => '[["xlsmp.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.en_US.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.EN_US.rte","5.1.0.0","default","lslpp"],
["xlsmp.aix61.rte","5.1.0.0","default","lslpp"],["xlmass.lib","6.1.0.4","default","lslpp"],["xlmass.aix53.lib","6.1.0.4","default","lslpp"], ["xlmass.adt.include","6.1.0.4","default","lslpp"],["xlmass.8.1.3","8.1.3.1","default","lslpp"],["xlmass.10.1.0","10.1.0.0","default","lslpp"],
["xlfrte.aix61","15.1.3.5","default","lslpp"],["xlfrte","15.1.3.5","default","lslpp"]]'; # here we get the positional indices of each array which we use to iterate over the data "l_indices_my_data" slist => getindices( my_data ); # We might build an associative array which can make the data easier to # work with. You may or may not want to use all the fields. You can even # do filtering here if you only need to use part of the data, e.g. by # adding if => regcmp( "^xlfrte.*$", "$(my_data[$(l_indices_my_data)][0])" ) "arr_my_data[$(my_data[$(l_indices_my_data)][0])][version]" string =>"$(my_data[$(l_indices_my_data)][1])"; "arr_my_data[$(my_data[$(l_indices_my_data)][0])][field3]" # Perhaps useless string =>"$(my_data[$(l_indices_my_data)][2])"; "arr_my_data[$(my_data[$(l_indices_my_data)][0])][command]" # Perhaps useless string =>"$(my_data[$(l_indices_my_data)][3])"; # From here you can get a consolidated list of package names, which could be filtered "l_my_data_packages" slist => getindices( arr_my_data ); "l_my_data_filtered" slist => filter( "^xlfrte.*$", getindices( arr_my_data ), # Could have used l_my_data_packages instead of getindices again true, false, inf); reports: # Print out a list of the filtered package names "Filtered result: $(with)" with => join( ", ", l_my_data_filtered ); # Print out a list of the filtered package names and their versions "Package $(l_my_data_filtered): $(arr_my_data[$(l_my_data_filtered)][version])"; }
# cf-agent --no-lock --log-level info --show-evaluated-vars=eval_numbers --file /tmp/example.cf R: Filtered result: xlfrte, xlfrte.aix61 R: Package xlfrte: 15.1.3.5 R: Package xlfrte.aix61: 15.1.3.5 Variable name Variable value Meta tags Comment default:eval_numbers.arr_my_data[xlfrte.aix61][command] lslpp source=promise default:eval_numbers.arr_my_data[xlfrte.aix61][field3] default source=promise default:eval_numbers.arr_my_data[xlfrte.aix61][version] 15.1.3.5 source=promise default:eval_numbers.arr_my_data[xlfrte][command] lslpp source=promise default:eval_numbers.arr_my_data[xlfrte][field3] default source=promise default:eval_numbers.arr_my_data[xlfrte][version] 15.1.3.5 source=promise default:eval_numbers.arr_my_data[xlmass.10.1.0][command] lslpp source=promise default:eval_numbers.arr_my_data[xlmass.10.1.0][field3] default source=promise default:eval_numbers.arr_my_data[xlmass.10.1.0][version] 10.1.0.0 source=promise default:eval_numbers.arr_my_data[xlmass.8.1.3][command] lslpp source=promise default:eval_numbers.arr_my_data[xlmass.8.1.3][field3] default source=promise default:eval_numbers.arr_my_data[xlmass.8.1.3][version] 8.1.3.1 source=promise default:eval_numbers.arr_my_data[xlmass.adt.include][command] lslpp source=promise default:eval_numbers.arr_my_data[xlmass.adt.include][field3] default source=promise default:eval_numbers.arr_my_data[xlmass.adt.include][version] 6.1.0.4 source=promise default:eval_numbers.arr_my_data[xlmass.aix53.lib][command] lslpp source=promise default:eval_numbers.arr_my_data[xlmass.aix53.lib][field3] default source=promise default:eval_numbers.arr_my_data[xlmass.aix53.lib][version] 6.1.0.4 source=promise default:eval_numbers.arr_my_data[xlmass.lib][command] lslpp source=promise default:eval_numbers.arr_my_data[xlmass.lib][field3] default source=promise default:eval_numbers.arr_my_data[xlmass.lib][version] 6.1.0.4 source=promise default:eval_numbers.arr_my_data[xlsmp.aix61.rte][command] lslpp source=promise default:eval_numbers.arr_my_data[xlsmp.aix61.rte][field3] default source=promise default:eval_numbers.arr_my_data[xlsmp.aix61.rte][version] 5.1.0.0 source=promise default:eval_numbers.arr_my_data[xlsmp.msg.EN_US.rte][command] lslpp source=promise default:eval_numbers.arr_my_data[xlsmp.msg.EN_US.rte][field3] default source=promise default:eval_numbers.arr_my_data[xlsmp.msg.EN_US.rte][version] 5.1.0.0 source=promise default:eval_numbers.arr_my_data[xlsmp.msg.en_US.rte][command] lslpp source=promise default:eval_numbers.arr_my_data[xlsmp.msg.en_US.rte][field3] default source=promise default:eval_numbers.arr_my_data[xlsmp.msg.en_US.rte][version] 5.1.0.0 source=promise default:eval_numbers.arr_my_data[xlsmp.rte][command] lslpp source=promise default:eval_numbers.arr_my_data[xlsmp.rte][field3] default source=promise default:eval_numbers.arr_my_data[xlsmp.rte][version] 5.1.0.0 source=promise default:eval_numbers.l_indices_my_data {"0","1","2","3","4","5","6","7","8","9","10"} source=promise default:eval_numbers.l_my_data_filtered {"xlfrte","xlfrte.aix61"} source=promise default:eval_numbers.l_my_data_packages {"xlfrte","xlmass.adt.include","xlmass.8.1.3","xlfrte.aix61","xlmass.10.1.0","xlsmp.aix61.rte","xlmass.lib","xlmass.aix53.lib","xlsmp.rte","xlsmp.msg.EN_US.rte","xlsmp.msg.en_US.rte"} source=promise default:eval_numbers.my_data [["xlsmp.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.en_US.rte","5.1.0.0","default","lslpp"],["xlsmp.msg.EN_US.rte","5.1.0.0","default","lslpp"],["xlsmp.aix61.rte","5.1.0.0","default","lslpp"],["xlmass.lib","6.1.0.4","default","lslpp"],["xlmass.aix53.lib","6.1.0.4","default","lslpp"],["xlmass.adt.include","6.1.0.4","default","lslpp"],["xlmass.8.1.3","8.1.3.1","default","lslpp"],["xlmass.10.1.0","10.1.0.0","default","lslpp"],["xlfrte.aix61","15.1.3.5","default","lslpp"],["xlfrte","15.1.3.5","default","lslpp"]] source=promise
Xander Cage
Jun 13, 2023, 11:07:35 AM6/13/23
to help-cfengine
Hi Nick,
<<
What is it you want to do with the version number?>>
in the end, i need the version number to upgrade aix filesets if downlevel. until now, i did this by exuting plain old system commands, but i thought it would be cool to use the already existing inventory file.
anyway...your example helped a lot.... as always ;-)
thank you for that
chris
Nick Anderson
Jun 13, 2023, 11:17:00 AM6/13/23
to help-cfengine
in the end, i need the version number to upgrade aix filesets if downlevel. until now, i did this by exuting plain old system commands, but i thought it would be cool to use the already existing inventory file. anyway…your example helped a lot…. as always ;-) thank you for that
Cool, what specifically was enlightening in the example I posted?
Xander Cage
Jun 14, 2023, 3:02:34 AM6/14/23
to help-cfengine
My initial thougt was that i can do the data filtering directly in the json array and ommit the intermediate legacy array. my existing promise splits the output of the lslpp command into an array. as this command is somewhat cpu intensive i contemplated with the inventory file solution. so i learned data filtering is not possible in an json array.
{
vars:
"lslpp_info" string => execresult ("/usr/bin/lslpp -O u -cl $(fileset) | /usr/bin/grep -v '^#'", "useshell"),
comment => "get info from lslpp fileset";
"info_array" slist => splitstring("$(lslpp_info)", ":", 8),
comment => "get each field from lslpp output";
"lpp_version" string => nth("info_array", 2);
"lpp_status" string => nth("info_array", 4);
"lpp_long_desc" string => nth("info_array", 6);
reports:
"Time: $(sys.date) - Bundle: $(this.bundle) - Message: [$(lslpp_info)] reported by lslpp."
comment => "do not remove this report .. won't call lslpp otherwise ..";
"$(lslpp_info)" bundle_return_value_index => "0",
comment => "give output back as index[0] (makes grepping for 'not installed' possible";
"$(lpp_version)" bundle_return_value_index => "1";
"$(lpp_status)" bundle_return_value_index => "2";
"$(lpp_long_desc)" bundle_return_value_index => "3";
}
Reply all
Reply to author
Forward
0 new messages