added -i option to list to search case insensitive

Updated manual

README.md

@@ -27,6 +27,8 @@ reasons:
   - templates for custom output for maximum flexibility
   - includes a tiny web server, which serves a restful API
 
+And I wrote a very similar [tool](https://www.daemon.de/projects/dbtool/) 24 years ago and wanted to do it again wit go.
+
 **anydb** can do all the things you can do with skate:
 
 ```shell

TODO.md

@@ -1,2 +1,4 @@
 - repl
 - mime-type => exec app + value
+- add waitgroup to db.go funcs
+- RestList does not support any params?

anydb.1

@@ -133,7 +133,7 @@
 .\" ========================================================================
 .\"
 .IX Title "ANYDB 1"
-.TH ANYDB 1 "2024-12-22" "1" "User Commands"
+.TH ANYDB 1 "2024-12-23" "1" "User Commands"
 .\" For nroff, turn off justification.  Always turn off hyphenation; it makes
 .\" way too many mistakes in technical documents.
 .if n .ad l
@@ -151,14 +151,20 @@ anydb \- a personal key value store
 \&    Available Commands:
 \&      completion  Generate the autocompletion script for the specified shell
 \&      del         Delete key
+\&      edit        Edit a key
 \&      export      Export database to json
 \&      get         Retrieve value for a key
 \&      help        Help about any command
 \&      import      Import database dump
+\&      info        info
 \&      list        List database contents
+\&      man         show manual page
+\&      serve       run REST API listener
 \&      set         Insert key/value pair
 \&    
 \&    Flags:
+\&      \-b, \-\-bucket string   use other bucket (default: data) (default "data")
+\&      \-c, \-\-config string   toml config file
 \&      \-f, \-\-dbfile string   DB file to use (default "/home/scip/.config/anydb/default.db")
 \&      \-d, \-\-debug           Enable debugging
 \&      \-h, \-\-help            help for anydb
@@ -168,9 +174,559 @@ anydb \- a personal key value store
 .Ve
 .SH "DESCRIPTION"
 .IX Header "DESCRIPTION"
-Anydb is a simple to use commandline tool to store anything you'd
+Anydb is a  commandline personal key value store, it  is simple to use
-like, even binary files etc. It uses a key/value store (bbolt) in your
+and  can be  used  to store  anything you'd  like,  even binary  files
-home directory.
+etc. It uses a key/value store (bbolt) in your home directory.
+.PP
+The tool provides a number of subcommands to use it, there are global
+options and each subcommand has its own set of options.
+.SH "GLOBAL OPTIONS"
+.IX Header "GLOBAL OPTIONS"
+.ie n .IP """\-f, \-\-dbfile filename""" 4
+.el .IP "\f(CW\-f, \-\-dbfile filename\fR" 4
+.IX Item "-f, --dbfile filename"
+The default location of your databas is
+\&\f(CW\*(C`$HOME/.config/anydb/default.db\*(C'\fR. You can change this with the \f(CW\*(C`\-f\*(C'\fR
+option.
+.ie n .IP """\-b, \-\-bucket name""" 4
+.el .IP "\f(CW\-b, \-\-bucket name\fR" 4
+.IX Item "-b, --bucket name"
+Data in a bbolt key-value-store are managed in so called
+buckets. These are kind of namespaces, where each key must be
+unique. However, a database may contain more than one bucket.
+.Sp
+By default anydb uses a bucket named \*(L"data\*(R", but you can change this
+using the option \f(CW\*(C`\-b\*(C'\fR.
+.Sp
+Buckets can be configured to always encrypt values, see \s-1ENCRYTPTION\s0.
+.ie n .IP """\-c, \-\-config filename""" 4
+.el .IP "\f(CW\-c, \-\-config filename\fR" 4
+.IX Item "-c, --config filename"
+Under normal circumstances you don't need a configuration file. But if
+you want, you can provide one using the option \f(CW\*(C`\-c\*(C'\fR.
+.Sp
+Anydb looks for a couple of default locations for a config file. You
+only need this option if you want to supply a configuration on a
+non-standard location. See \s-1CONFIGURATION\s0 for more details.
+.ie n .IP """\-d, \-\-debug""" 4
+.el .IP "\f(CW\-d, \-\-debug\fR" 4
+.IX Item "-d, --debug"
+Enable debug output.
+.ie n .IP """\-h, \-\-help""" 4
+.el .IP "\f(CW\-h, \-\-help\fR" 4
+.IX Item "-h, --help"
+Show the usage of anydb.
+.ie n .IP """\-v, \-\-version""" 4
+.el .IP "\f(CW\-v, \-\-version\fR" 4
+.IX Item "-v, --version"
+Show the program version.
+.PP
+All of these options can be used with subcommands as well.
+.SH "SUBCOMMANDS"
+.IX Header "SUBCOMMANDS"
+.SS "completion"
+.IX Subsection "completion"
+The \fBcompletion\fR command can be used to setup completion for
+anydb. Just put something like this into your shell's configuration
+file:
+.PP
+.Vb 1
+\&    source <(anydb completion bash)
+.Ve
+.PP
+If you use another shell, specify it instead of bash, of course.
+.SS "set"
+.IX Subsection "set"
+The \fBset\fR command is being used to insert or update a key-value pair.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb set <key> [<value> | \-r <file>] [\-t <tag>] [flags]
+\&    
+\&    Aliases:
+\&      set, add, s, +
+\&    
+\&    Flags:
+\&      \-e, \-\-encrypt            encrypt value
+\&      \-r, \-\-file string        Filename or \- for STDIN
+\&      \-h, \-\-help               help for set
+\&      \-t, \-\-tags tag,tag,...   tags, multiple allowed
+.Ve
+.PP
+The standard way to insert a new entry is really simple:
+.PP
+.Vb 1
+\&    anydb set key value
+.Ve
+.PP
+If you don't specify a value, anydb expects you to feed it some data
+via \s-1STDIN.\s0 For example:
+.PP
+.Vb 1
+\&    anydb set key < file
+.Ve
+.PP
+You might as well specify a file directly using the \f(CW\*(C`\-f\*(C'\fR option:
+.PP
+.Vb 1
+\&    anydb set key \-f file
+.Ve
+.PP
+Values can be encrypted using \fBChaCha20Poly1305\fR when you specify the
+\&\f(CW\*(C`\-e\*(C'\fR option. Anydb will ask you interactively for a passphrase. You
+might as well provide the passphrase using the environment variable
+\&\f(CW\*(C`ANYDB_PASSWORD\*(C'\fR. To encrypt the value, a cryptographically secure
+key will be derived from the passphrase using the ArgonID2
+algorithm. Each value can be encrypted with another passphrase. So,
+the database itself is not encrypted, just the values.
+.PP
+You can supply tags by using the option \f(CW\*(C`\-t\*(C'\fR. Multiple tags can be
+provided either by separating them with a comma or by using multiple
+\&\f(CW\*(C`\-t\*(C'\fR parameters:
+.PP
+.Vb 2
+\&    anydb set key value \-t tag1,tag2
+\&    anydb set key value \-t tag1 \-t tag2
+.Ve
+.PP
+You can later filter entries by tag or by a combination of tags.
+.PP
+To edit or modify an entry, just use the \fBset\fR command with the same
+key, the value in the database will be overwritten with the new
+value. An alternative option is the \fBedit\fR command, see below.
+.SS "get"
+.IX Subsection "get"
+To retrieve the value of a key, use the \fBget\fR subcommand.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb get  <key> [\-o <file>] [\-m <mode>] [\-n \-N] [\-T <tpl>] [flags]
+\&    
+\&    Aliases:
+\&      get, show, g, .
+\&    
+\&    Flags:
+\&      \-h, \-\-help              help for get
+\&      \-m, \-\-mode string       output format (simple|wide|json|template) (default \*(Aqsimple\*(Aq)
+\&      \-n, \-\-no\-headers        omit headers in tables
+\&      \-N, \-\-no\-human          do not translate to human readable values
+\&      \-o, \-\-output string     output value to file (ignores \-m)
+\&      \-T, \-\-template string   go template for \*(Aq\-m template\*(Aq
+.Ve
+.PP
+In its simplest form you just call the \fBget\fR subcommand with the key
+you want to have the value for. The value is being printed to \s-1STDOUT\s0
+by default:
+.PP
+.Vb 1
+\&    anydb get key
+.Ve
+.PP
+If the  value is binary  content, it will  not just being  printed. In
+those cases you need to either redirect  output into a file or use the
+option \f(CW\*(C`\-o\*(C'\fR to write to a file:
+.PP
+.Vb 2
+\&    anydb get key > file
+\&    anydb get key \-o file
+.Ve
+.PP
+If the value is encrypted, you will be asked for the passphrase to
+decrypt it. If the environment variable \f(CW\*(C`ANYDB_PASSWORD\*(C'\fR is set, its
+value will be used instead.
+.PP
+There are different output modes you  can choos from: simple, wide and
+json. The \*(L"simple\*(R"  mode is the default one, it  just prints the value
+as is. The \*(L"wide\*(R" mode prints  a tabular output similar to the \fBlist\fR
+subcommand, see  there for more  details. The options \f(CW\*(C`\-n\*(C'\fR  and \f(CW\*(C`\-N\*(C'\fR
+have the same  meaning as in the list command.  The \*(L"json\*(R" mode prints
+the raw  \s-1JSON\s0 representation of  the whole database  entry. Decryption
+will only take place in \*(L"simple\*(R"  and \*(L"json\*(R" mode. The \*(L"template\*(R" mode
+provides  the   most  flexibily,  it   is  detailed  in   the  section
+\&\s-1TEMPLATES\s0.
+.SS "list"
+.IX Subsection "list"
+The \fBlist\fR subcommand displays a list of all database entries.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb list  [<filter\-regex>] [\-t <tag>] [\-m <mode>] [\-n \-N] [\-T <tpl>] [flags]
+\&    
+\&    Aliases:
+\&      list, /, ls
+\&    
+\&    Flags:
+\&      \-h, \-\-help               help for list
+\&      \-m, \-\-mode string        output format (table|wide|json|template), wide is a verbose table. (default \*(Aqtable\*(Aq)
+\&      \-n, \-\-no\-headers         omit headers in tables
+\&      \-N, \-\-no\-human           do not translate to human readable values
+\&      \-t, \-\-tags stringArray   tags, multiple allowed
+\&      \-T, \-\-template string    go template for \*(Aq\-m template\*(Aq
+\&      \-l, \-\-wide\-output        output mode: wide
+.Ve
+.PP
+In its simplest form \- without any options \- , the \fBlist\fR command
+just prints all keys with their values to \s-1STDOUT.\s0 Values are being
+truncated to maximum of 60 characters, that is, multiline values are
+not completely shown in order to keep the tabular view readable.
+.PP
+To get more informations about each entry, use the \f(CW\*(C`\-o wide\*(C'\fR or \f(CW\*(C`\-l\*(C'\fR
+option.  In addition  to  the  key and  value  also  the size,  update
+timestamp and tags will be printed. Time and size values are converted
+into a  human readable form, you  can suppress this behavior  with the
+\&\f(CW\*(C`\-N\*(C'\fR option. You may omit the headers using the option \f(CW\*(C`\-n\*(C'\fR
+.PP
+Sometimes you might want to filter the list of entries. Either because
+your database grew too large or because you're searching for
+something. In that case you have two options: You may supply one or
+more tags or provide a filter regexp. To filter by tag, do:
+.PP
+.Vb 3
+\&    anydb list \-t tag1
+\&    anydb list \-t tag1,tag2
+\&    anydb list \-t tag1 \-t tag2
+.Ve
+.PP
+To filter using a regular expression, do:
+.PP
+.Vb 1
+\&   anydb list "foo.*bar"
+.Ve
+.PP
+Regular expressions follow the golang \fBre2\fR syntax. For more details
+about the syntax, refer to
+<https://github.com/google/re2/wiki/Syntax>. Please note, that this
+regexp dialect is not \s-1PCRE\s0 compatible, but supports most of its
+features.
+.PP
+You can \- as with the \fBget\fR command \- use other output modes. The
+default mode is \*(L"table\*(R". The \*(L"wide\*(R" mode is, as already mentioned, a
+more detailed table. Also supported is \*(L"json\*(R" mode and \*(L"template\*(R"
+mode. For details about using templates see \s-1TEMPLATES\s0.
+.SS "del"
+.IX Subsection "del"
+Use the \fBdel\fR command to delete database entries.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb del <key> [flags]
+\&    
+\&    Aliases:
+\&      del, d, rm
+\&    
+\&    Flags:
+\&      \-h, \-\-help   help for del
+.Ve
+.PP
+The subcommand  \fBdel\fR does not  provide any further options,  it just
+deletes the entry referred to by the given key. No questions are being
+asked.
+.SS "edit"
+.IX Subsection "edit"
+The \fBedit\fR command makes it easier to modify larger entries.
+.PP
+Usage:
+.PP
+.Vb 2
+\&   Usage:
+\&      anydb edit <key> [flags]
+\&    
+\&    Aliases:
+\&      edit, modify, mod, ed, vi
+\&    
+\&    Flags:
+\&      \-h, \-\-help   help for edit
+.Ve
+.PP
+The subcommand  \fBedit\fR does not  provide any further options. It
+works like this:
+.IP "1. Write the value info a temporary file." 4
+.IX Item "1. Write the value info a temporary file."
+.PD 0
+.IP "2. Execute the editor (which one, see below!) with that file." 4
+.IX Item "2. Execute the editor (which one, see below!) with that file."
+.IP "3. Now you can edit the file and save+close it when done." 4
+.IX Item "3. Now you can edit the file and save+close it when done."
+.IP "4. Anydb picks up the file and if the content has changed, puts its value into the \s-1DB.\s0" 4
+.IX Item "4. Anydb picks up the file and if the content has changed, puts its value into the DB."
+.PD
+.PP
+By default anydb executes the \f(CW\*(C`vi\*(C'\fR command. You can modify this
+behavior by setting the environment variable \f(CW\*(C`EDITOR\*(C'\fR appropriately.
+.PP
+Please note, that this does not work with binary content!
+.SS "export"
+.IX Subsection "export"
+Since the bbold database file is not portable across platforms (it is
+bound to the endianess of the \s-1CPU\s0 it was being created on), you might
+want to create a backup file of your database. You can do this with
+the \fBexport\fR subcommand.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb export [\-o <json filename>] [flags]
+\&    
+\&    Aliases:
+\&      export, dump, backup
+\&    
+\&    Flags:
+\&      \-h, \-\-help            help for export
+\&      \-o, \-\-output string   output to file
+.Ve
+.PP
+The database dump is a \s-1JSON\s0 representation of the whole database and
+will be printed to \s-1STDOUT\s0 by default. Redirect it to a file or use the
+\&\f(CW\*(C`\-o\*(C'\fR option:
+.PP
+.Vb 2
+\&    anydb export > dump.json
+\&    anydb export \-o dump.json
+.Ve
+.PP
+Please note, that encrypted values will not be decrypted. This might
+change in a future version of anydb.
+.SS "import"
+.IX Subsection "import"
+The \fBimport\fR subcommand can be used to restore a database from a \s-1JSON\s0
+dump.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb import [<json file>] [flags]
+\&    
+\&    Aliases:
+\&      import, restore
+\&    
+\&    Flags:
+\&      \-r, \-\-file string        Filename or \- for STDIN
+\&      \-h, \-\-help               help for import
+\&      \-t, \-\-tags stringArray   tags, multiple allowed
+.Ve
+.PP
+By default the \f(CW\*(C`import\*(C'\fR subcommand reads the \s-1JSON\s0 contents from
+\&\s-1STDIN.\s0 You might pipe the dump into it or use the option \f(CW\*(C`\-r\*(C'\fR:
+.PP
+.Vb 3
+\&    anydb import < dump.json
+\&    anydb import \-r dump.json
+\&    cat dump.json | anydb import
+.Ve
+.PP
+If there is already a database, it will be saved by appending a
+timestamp and a new database with the contents of the dump will be
+created.
+.SS "serve"
+.IX Subsection "serve"
+Anydb provides a RESTful \s-1API,\s0 which you can use to manage the database
+from somewhere  else. The \s-1API\s0  does not provide any  authentication or
+any other security measures, so better only use it on localhost.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb serve [\-l host:port] [flags]
+\&    
+\&    Flags:
+\&      \-h, \-\-help            help for serve
+\&      \-l, \-\-listen string   host:port (default "localhost:8787")
+.Ve
+.PP
+To start the listener, just execute the \fBserve\fR subcommand. You can
+tweak the ip address and tcp port using the \f(CW\*(C`\-l\*(C'\fR option. The listener
+will not fork and run in the foreground. Logs are being printed to
+\&\s-1STDOUT\s0 as long as the listener runs.
+.PP
+For more details about the \s-1API,\s0 please see the \*(L"\s-1REST API\*(R"\s0 section.
+.SS "info"
+.IX Subsection "info"
+The \fBinfo\fR subcommand shows you some information about your current
+database.
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb info [flags]
+\&    
+\&    Flags:
+\&      \-h, \-\-help       help for info
+\&      \-N, \-\-no\-human   do not translate to human readable values
+.Ve
+.PP
+Data being shown are: filename and size, number of keys per bucket. If
+you supply the \f(CW\*(C`\-d\*(C'\fR option (debug), some bbolt internals are being
+displayed as well.
+.SS "man"
+.IX Subsection "man"
+The \fBman\fR subcommand shows an unformatted text variant of the manual
+page (which are currently reading).
+.PP
+Usage:
+.PP
+.Vb 2
+\&    Usage:
+\&      anydb man [flags]
+\&    
+\&    Flags:
+\&      \-h, \-\-help   help for man
+.Ve
+.PP
+The manual is being piped into the \f(CW\*(C`more\*(C'\fR command, which is being
+expected to exist according to the \s-1POSIX\s0 standard on all supported
+unix platforms. It might not work on Windows.
+.SH "TEMPLATES"
+.IX Header "TEMPLATES"
+The \fBget\fR and \fBlist\fR commands support a template feature, which is
+very handy to create you own kind of formatting. The template syntax
+being used is the \s-1GO\s0 template language, refer to
+<https://pkg.go.dev/text/template> for details.
+.PP
+Each template operates on one or more entries, no loop construct is
+required, the template provided applies to every matching entry
+separatley.
+.PP
+The following template variables can be used:
+.IP "\fBKey\fR \- string =item \fBValue\fR \- string =item \fBBin\fR \- []byte =item \fBCreated\fR \- time.Time =item \fBTags\fR \- []string =item \fBEncrypted\fR bool" 4
+.IX Item "Key - string =item Value - string =item Bin - []byte =item Created - time.Time =item Tags - []string =item Encrypted bool"
+.PP
+Prepend a single dot (\*(L".\*(R") before each variable name.
+.PP
+Here are some examples how to use the feature:
+.PP
+Only show the keys of all entries:
+.PP
+.Vb 1
+\&    anydb list \-m template \-T "{{ .Key }}"
+.Ve
+.PP
+Format the list in a way so that is possible to evaluate it in a
+shell:
+.PP
+.Vb 2
+\&    eval $(anydb get foo \-m template \-T "key=\*(Aq{{ .Key }}\*(Aq value=\*(Aq{{ .Value }}\*(Aq ts=\*(Aq{{ .Created}}\*(Aq")
+\&    echo "Key: $key, Value: $value"
+.Ve
+.PP
+Print the values in \s-1CSV\s0 format \s-1ONLY\s0 if they have some tag:
+.PP
+.Vb 1
+\&    anydb list \-m template \-T "{{ if .Tags }}{{ .Key }},{{ .Value }},{{ .Created}}{{ end }}"
+.Ve
+.SH "CONFIGURATION"
+.IX Header "CONFIGURATION"
+Anydb looks at the following location for a configuration file, in
+that order:
+.ie n .IP """$HOME/.config/anydb/anydb.toml""" 4
+.el .IP "\f(CW$HOME/.config/anydb/anydb.toml\fR" 4
+.IX Item "$HOME/.config/anydb/anydb.toml"
+.PD 0
+.ie n .IP """$HOME/.anydb.toml""" 4
+.el .IP "\f(CW$HOME/.anydb.toml\fR" 4
+.IX Item "$HOME/.anydb.toml"
+.ie n .IP """anydb.toml"" in the current directory" 4
+.el .IP "\f(CWanydb.toml\fR in the current directory" 4
+.IX Item "anydb.toml in the current directory"
+.ie n .IP "or specify one using ""\-c""" 4
+.el .IP "or specify one using \f(CW\-c\fR" 4
+.IX Item "or specify one using -c"
+.PD
+The configuration format uses the \s-1TOML\s0 language, refer to
+<https://toml.io/en/> for more details. The key names correspond to
+the commandline options in most cases.
+.Sp
+Configuration follows a certain precedence: the files are tried to be
+read in the given order, followed by commandline options. That is, the
+last configuration file wins, unless the user provides a commandline
+option, then this setting will be taken.
+.Sp
+A complete configuration file might look like this:
+.Sp
+.Vb 7
+\&    # defaults
+\&    dbfile     = "~/.config/anydb/default.db"
+\&    dbbucket   = "data"
+\&    noheaders  = false
+\&    nohumanize = false
+\&    encrypt    = false
+\&    listen     = "localhost:8787"
+\&    
+\&    # different setups for different buckets
+\&    [buckets.data]
+\&    encrypt = true
+\&    
+\&    [buckets.test]
+\&    encrypt = false
+.Ve
+.Sp
+Under normal circumstances you don't need a configuration
+file. However, if you want to use different buckets, then this might
+be a handy option. Buckets are being configured in ini-style with the
+term \*(L"bucket.\*(R" followed by the bucket name. In the example above we
+enable encryption for the default bucket \*(L"data\*(R" and disable it for a
+bucket \*(L"test\*(R". To use different buckets, use the \f(CW\*(C`\-b\*(C'\fR option.
+.SH "REST API"
+.IX Header "REST API"
+The subcommand \fBserve\fR starts a simple \s-1HTTP\s0 service, which responds
+to RESTful \s-1HTTP\s0 requests. The listener responds to all requests with a
+\&\s-1JSON\s0 encoded response. The response contains the status and the
+content \- if any \- of the requested resource.
+.PP
+The following requests are supported:
+.IP "\fB\s-1GET\s0 /anydb/v1/\fR" 4
+.IX Item "GET /anydb/v1/"
+Returns a \s-1JSON\s0 encoded list of all entries.
+.IP "\fB\s-1GET\s0 /anydb/v1/key\fR" 4
+.IX Item "GET /anydb/v1/key"
+Returns the \s-1JSON\s0 encoded entry, if found.
+.IP "\fB\s-1PUT\s0 /anydb/v1/\fR" 4
+.IX Item "PUT /anydb/v1/"
+Create an entry. Expects a \s-1JSON\s0 encoded request object in \s-1POST\s0 data.
+.IP "\fB\s-1DELETE\s0 /anydb/v1/key\fR" 4
+.IX Item "DELETE /anydb/v1/key"
+Delete an entry.
+.PP
+Some curl example calls to the \s-1API:\s0
+.PP
+Post a new key:
+    curl \-X \s-1PUT\s0 localhost:8787/anydb/v1/ \e
+      \-H 'Content\-Type: application/json' \e
+      \-d '{\*(L"key\*(R":\*(L"foo\*(R",\*(L"val\*(R":\*(L"bar\*(R"}'
+.PP
+Retrieve the value:
+.PP
+.Vb 1
+\&    curl localhost:8787/anydb/v1/foo
+.Ve
+.PP
+List all keys:
+.PP
+.Vb 1
+\&    curl localhost:8787/anydb/v1/
+.Ve
+.SH "BUGS"
+.IX Header "BUGS"
+In order to report a bug, unexpected behavior, feature requests
+or to submit a patch, please open an issue on github:
+<https://github.com/TLINDEN/anydb/issues>.
+.PP
+Please repeat the failing command with debugging enabled \f(CW\*(C`\-d\*(C'\fR and
+include the output in the issue.
+.SH "LIMITATIONS"
+.IX Header "LIMITATIONS"
+The \s-1REST API\s0 list request doesn't provide any filtering capabilities yet.
 .SH "LICENSE"
 .IX Header "LICENSE"
 This software is licensed under the \s-1GNU GENERAL PUBLIC LICENSE\s0 version 3.

anydb.pod

@@ -11,14 +11,20 @@ anydb - a personal key value store
     Available Commands:
       completion  Generate the autocompletion script for the specified shell
       del         Delete key
+      edit        Edit a key
       export      Export database to json
       get         Retrieve value for a key
       help        Help about any command
       import      Import database dump
+      info        info
       list        List database contents
+      man         show manual page
+      serve       run REST API listener
       set         Insert key/value pair
     
     Flags:
+      -b, --bucket string   use other bucket (default: data) (default "data")
+      -c, --config string   toml config file
       -f, --dbfile string   DB file to use (default "/home/scip/.config/anydb/default.db")
       -d, --debug           Enable debugging
       -h, --help            help for anydb
@@ -26,11 +32,551 @@ anydb - a personal key value store
     
     Use "anydb [command] --help" for more information about a command.
 
+
 =head1 DESCRIPTION
 
-Anydb is a simple to use commandline tool to store anything you'd
+Anydb is a  commandline personal key value store, it  is simple to use
-like, even binary files etc. It uses a key/value store (bbolt) in your
+and  can be  used  to store  anything you'd  like,  even binary  files
-home directory.
+etc. It uses a key/value store (bbolt) in your home directory.
+
+The tool provides a number of subcommands to use it, there are global
+options and each subcommand has its own set of options.
+
+=head1 GLOBAL OPTIONS
+
+=over
+
+=item C<-f, --dbfile filename>
+
+The default location of your databas is
+C<$HOME/.config/anydb/default.db>. You can change this with the C<-f>
+option.
+
+=item C<-b, --bucket name>
+
+Data in a bbolt key-value-store are managed in so called
+buckets. These are kind of namespaces, where each key must be
+unique. However, a database may contain more than one bucket.
+
+By default anydb uses a bucket named "data", but you can change this
+using the option C<-b>.
+
+Buckets can be configured to always encrypt values, see L<ENCRYTPTION>.
+
+=item C<-c, --config filename>
+
+Under normal circumstances you don't need a configuration file. But if
+you want, you can provide one using the option C<-c>.
+
+Anydb looks for a couple of default locations for a config file. You
+only need this option if you want to supply a configuration on a
+non-standard location. See L<CONFIGURATION> for more details.
+
+=item C<-d, --debug>
+
+Enable debug output.
+
+=item C<-h, --help>
+
+Show the usage of anydb.
+
+=item C<-v, --version>
+
+Show the program version.
+
+=back
+
+All of these options can be used with subcommands as well.
+
+=head1 SUBCOMMANDS
+
+=head2 completion
+
+The B<completion> command can be used to setup completion for
+anydb. Just put something like this into your shell's configuration
+file:
+
+    source <(anydb completion bash)
+
+If you use another shell, specify it instead of bash, of course.
+
+=head2 set
+
+The B<set> command is being used to insert or update a key-value pair.
+
+Usage:
+
+    Usage:
+      anydb set <key> [<value> | -r <file>] [-t <tag>] [flags]
+    
+    Aliases:
+      set, add, s, +
+    
+    Flags:
+      -e, --encrypt            encrypt value
+      -r, --file string        Filename or - for STDIN
+      -h, --help               help for set
+      -t, --tags tag,tag,...   tags, multiple allowed
+    
+The standard way to insert a new entry is really simple:
+
+    anydb set key value
+
+If you don't specify a value, anydb expects you to feed it some data
+via STDIN. For example:
+
+    anydb set key < file
+
+You might as well specify a file directly using the C<-f> option:
+
+    anydb set key -f file
+
+Values can be encrypted using B<ChaCha20Poly1305> when you specify the
+C<-e> option. Anydb will ask you interactively for a passphrase. You
+might as well provide the passphrase using the environment variable
+C<ANYDB_PASSWORD>. To encrypt the value, a cryptographically secure
+key will be derived from the passphrase using the ArgonID2
+algorithm. Each value can be encrypted with another passphrase. So,
+the database itself is not encrypted, just the values.
+
+You can supply tags by using the option C<-t>. Multiple tags can be
+provided either by separating them with a comma or by using multiple
+C<-t> parameters:
+
+    anydb set key value -t tag1,tag2
+    anydb set key value -t tag1 -t tag2
+
+You can later filter entries by tag or by a combination of tags.
+
+To edit or modify an entry, just use the B<set> command with the same
+key, the value in the database will be overwritten with the new
+value. An alternative option is the B<edit> command, see below.
+
+=head2 get
+
+To retrieve the value of a key, use the B<get> subcommand.
+
+Usage:
+
+    Usage:
+      anydb get  <key> [-o <file>] [-m <mode>] [-n -N] [-T <tpl>] [flags]
+    
+    Aliases:
+      get, show, g, .
+    
+    Flags:
+      -h, --help              help for get
+      -m, --mode string       output format (simple|wide|json|template) (default 'simple')
+      -n, --no-headers        omit headers in tables
+      -N, --no-human          do not translate to human readable values
+      -o, --output string     output value to file (ignores -m)
+      -T, --template string   go template for '-m template'
+
+In its simplest form you just call the B<get> subcommand with the key
+you want to have the value for. The value is being printed to STDOUT
+by default:
+
+    anydb get key
+
+If the  value is binary  content, it will  not just being  printed. In
+those cases you need to either redirect  output into a file or use the
+option C<-o> to write to a file:
+
+    anydb get key > file
+    anydb get key -o file
+
+If the value is encrypted, you will be asked for the passphrase to
+decrypt it. If the environment variable C<ANYDB_PASSWORD> is set, its
+value will be used instead.
+
+There are different output modes you  can choos from: simple, wide and
+json. The "simple"  mode is the default one, it  just prints the value
+as is. The "wide" mode prints  a tabular output similar to the B<list>
+subcommand, see  there for more  details. The options C<-n>  and C<-N>
+have the same  meaning as in the list command.  The "json" mode prints
+the raw  JSON representation of  the whole database  entry. Decryption
+will only take place in "simple"  and "json" mode. The "template" mode
+provides  the   most  flexibily,  it   is  detailed  in   the  section
+L<TEMPLATES>.
+
+=head2 list
+
+The B<list> subcommand displays a list of all database entries.
+
+Usage:
+
+    Usage:
+      anydb list  [<filter-regex>] [-t <tag>] [-m <mode>] [-n -N] [-T <tpl>] [-i] [flags]
+    
+    Aliases:
+      list, /, ls
+    
+    Flags:
+      -i, --case-insensitive   filter case insensitive
+      -h, --help               help for list
+      -m, --mode string        output format (table|wide|json|template),
+                               wide is a verbose table. (default 'table')
+      -n, --no-headers         omit headers in tables
+      -N, --no-human           do not translate to human readable values
+      -t, --tags stringArray   tags, multiple allowed
+      -T, --template string    go template for '-m template'
+      -l, --wide-output        output mode: wide
+
+In its simplest form - without any options - , the B<list> command
+just prints all keys with their values to STDOUT. Values are being
+truncated to maximum of 60 characters, that is, multiline values are
+not completely shown in order to keep the tabular view readable.
+
+To get more informations about each entry, use the C<-o wide> or C<-l>
+option.  In addition  to  the  key and  value  also  the size,  update
+timestamp and tags will be printed. Time and size values are converted
+into a  human readable form, you  can suppress this behavior  with the
+C<-N> option. You may omit the headers using the option C<-n>
+
+Sometimes you might want to filter the list of entries. Either because
+your database grew too large or because you're searching for
+something. In that case you have two options: You may supply one or
+more tags or provide a filter regexp. To filter by tag, do:
+
+    anydb list -t tag1
+    anydb list -t tag1,tag2
+    anydb list -t tag1 -t tag2
+
+To filter using a regular expression, do:
+
+   anydb list "foo.*bar"
+
+Regular expressions follow the golang B<re2> syntax. For more details
+about the syntax, refer to
+L<https://github.com/google/re2/wiki/Syntax>. Please note, that this
+regexp dialect is not PCRE compatible, but supports most of its
+features.
+
+If you want to search case insensitive, add the option C<-i>.
+
+You can - as with the B<get> command - use other output modes. The
+default mode is "table". The "wide" mode is, as already mentioned, a
+more detailed table. Also supported is "json" mode and "template"
+mode. For details about using templates see L<TEMPLATES>.
+
+=head2 del
+
+Use the B<del> command to delete database entries.
+
+Usage:
+
+    Usage:
+      anydb del <key> [flags]
+    
+    Aliases:
+      del, d, rm
+    
+    Flags:
+      -h, --help   help for del
+
+The subcommand  B<del> does not  provide any further options,  it just
+deletes the entry referred to by the given key. No questions are being
+asked.
+
+=head2 edit
+
+The B<edit> command makes it easier to modify larger entries.
+
+Usage:
+
+   Usage:
+      anydb edit <key> [flags]
+    
+    Aliases:
+      edit, modify, mod, ed, vi
+    
+    Flags:
+      -h, --help   help for edit
+
+The subcommand  B<edit> does not  provide any further options. It
+works like this:
+
+=over
+
+=item 1. Write the value info a temporary file.
+
+=item 2. Execute the editor (which one, see below!) with that file.
+
+=item 3. Now you can edit the file and save+close it when done.
+
+=item 4. Anydb picks up the file and if the content has changed, puts its value into the DB.
+
+=back
+
+By default anydb executes the C<vi> command. You can modify this
+behavior by setting the environment variable C<EDITOR> appropriately.
+
+Please note, that this does not work with binary content!
+
+=head2 export
+
+Since the bbold database file is not portable across platforms (it is
+bound to the endianess of the CPU it was being created on), you might
+want to create a backup file of your database. You can do this with
+the B<export> subcommand.
+
+Usage:
+
+    Usage:
+      anydb export [-o <json filename>] [flags]
+    
+    Aliases:
+      export, dump, backup
+    
+    Flags:
+      -h, --help            help for export
+      -o, --output string   output to file
+
+The database dump is a JSON representation of the whole database and
+will be printed to STDOUT by default. Redirect it to a file or use the
+C<-o> option:
+
+    anydb export > dump.json
+    anydb export -o dump.json
+
+Please note, that encrypted values will not be decrypted. This might
+change in a future version of anydb.
+
+=head2 import
+
+The B<import> subcommand can be used to restore a database from a JSON
+dump.
+
+Usage:
+
+    Usage:
+      anydb import [<json file>] [flags]
+    
+    Aliases:
+      import, restore
+    
+    Flags:
+      -r, --file string        Filename or - for STDIN
+      -h, --help               help for import
+      -t, --tags stringArray   tags, multiple allowed
+
+By default the C<import> subcommand reads the JSON contents from
+STDIN. You might pipe the dump into it or use the option C<-r>:
+
+    anydb import < dump.json
+    anydb import -r dump.json
+    cat dump.json | anydb import
+
+If there is already a database, it will be saved by appending a
+timestamp and a new database with the contents of the dump will be
+created.
+
+=head2 serve
+
+Anydb provides a RESTful API, which you can use to manage the database
+from somewhere  else. The API  does not provide any  authentication or
+any other security measures, so better only use it on localhost.
+
+Usage:
+
+    Usage:
+      anydb serve [-l host:port] [flags]
+    
+    Flags:
+      -h, --help            help for serve
+      -l, --listen string   host:port (default "localhost:8787")
+
+To start the listener, just execute the B<serve> subcommand. You can
+tweak the ip address and tcp port using the C<-l> option. The listener
+will not fork and run in the foreground. Logs are being printed to
+STDOUT as long as the listener runs.
+
+For more details about the API, please see the L<REST API> section.
+
+=head2 info
+
+The B<info> subcommand shows you some information about your current
+database.
+
+Usage:
+
+    Usage:
+      anydb info [flags]
+    
+    Flags:
+      -h, --help       help for info
+      -N, --no-human   do not translate to human readable values
+
+Data being shown are: filename and size, number of keys per bucket. If
+you supply the C<-d> option (debug), some bbolt internals are being
+displayed as well.
+
+=head2 man
+
+The B<man> subcommand shows an unformatted text variant of the manual
+page (which are currently reading).
+
+Usage:
+
+    Usage:
+      anydb man [flags]
+    
+    Flags:
+      -h, --help   help for man
+
+The manual is being piped into the C<more> command, which is being
+expected to exist according to the POSIX standard on all supported
+unix platforms. It might not work on Windows.
+
+=head1 TEMPLATES
+
+The B<get> and B<list> commands support a template feature, which is
+very handy to create you own kind of formatting. The template syntax
+being used is the GO template language, refer to
+L<https://pkg.go.dev/text/template> for details.
+
+Each template operates on one or more entries, no loop construct is
+required, the template provided applies to every matching entry
+separatley.
+
+The following template variables can be used:
+
+=over
+
+=item B<Key> - string
+=item B<Value> - string
+=item B<Bin> - []byte
+=item B<Created> - time.Time
+=item B<Tags> - []string
+=item B<Encrypted> bool
+
+=back
+
+Prepend a single dot (".") before each variable name.
+
+Here are some examples how to use the feature:
+
+Only show the keys of all entries:
+
+    anydb list -m template -T "{{ .Key }}"
+
+Format the list in a way so that is possible to evaluate it in a
+shell:
+
+    eval $(anydb get foo -m template -T "key='{{ .Key }}' value='{{ .Value }}' ts='{{ .Created}}'")
+    echo "Key: $key, Value: $value"
+
+Print the values in CSV format ONLY if they have some tag:
+
+    anydb list -m template -T "{{ if .Tags }}{{ .Key }},{{ .Value }},{{ .Created}}{{ end }}"
+
+
+=head1 CONFIGURATION
+
+Anydb looks at the following location for a configuration file, in
+that order:
+
+=over
+
+=item C<$HOME/.config/anydb/anydb.toml>
+
+=item C<$HOME/.anydb.toml>
+
+=item C<anydb.toml> in the current directory
+
+=item or specify one using C<-c>
+
+The configuration format uses the TOML language, refer to
+L<https://toml.io/en/> for more details. The key names correspond to
+the commandline options in most cases.
+
+Configuration follows a certain precedence: the files are tried to be
+read in the given order, followed by commandline options. That is, the
+last configuration file wins, unless the user provides a commandline
+option, then this setting will be taken.
+
+A complete configuration file might look like this:
+
+    # defaults
+    dbfile     = "~/.config/anydb/default.db"
+    dbbucket   = "data"
+    noheaders  = false
+    nohumanize = false
+    encrypt    = false
+    listen     = "localhost:8787"
+    
+    # different setups for different buckets
+    [buckets.data]
+    encrypt = true
+    
+    [buckets.test]
+    encrypt = false
+
+Under normal circumstances you don't need a configuration
+file. However, if you want to use different buckets, then this might
+be a handy option. Buckets are being configured in ini-style with the
+term "bucket." followed by the bucket name. In the example above we
+enable encryption for the default bucket "data" and disable it for a
+bucket "test". To use different buckets, use the C<-b> option.
+
+=back
+
+=head1 REST API
+
+The subcommand B<serve> starts a simple HTTP service, which responds
+to RESTful HTTP requests. The listener responds to all requests with a
+JSON encoded response. The response contains the status and the
+content - if any - of the requested resource.
+
+The following requests are supported:
+
+=over
+
+=item B<GET /anydb/v1/>
+
+Returns a JSON encoded list of all entries.
+
+=item B<GET /anydb/v1/key>
+
+Returns the JSON encoded entry, if found.
+
+=item B<PUT /anydb/v1/>
+
+Create an entry. Expects a JSON encoded request object in POST data.
+
+=item B<DELETE /anydb/v1/key>
+
+Delete an entry.
+
+=back
+
+Some curl example calls to the API: 
+
+Post a new key:
+    curl -X PUT localhost:8787/anydb/v1/ \
+      -H 'Content-Type: application/json' \
+      -d '{"key":"foo","val":"bar"}'
+
+Retrieve the value:
+
+    curl localhost:8787/anydb/v1/foo
+
+List all keys: 
+
+    curl localhost:8787/anydb/v1/
+
+=head1 BUGS
+
+In order to report a bug, unexpected behavior, feature requests
+or to submit a patch, please open an issue on github:
+L<https://github.com/TLINDEN/anydb/issues>.
+
+Please repeat the failing command with debugging enabled C<-d> and
+include the output in the issue.
+
+=head1 LIMITATIONS
+
+The REST API list request doesn't provide any filtering capabilities yet.
 
 =head1 LICENSE
 

cfg/config.go

@@ -26,23 +26,24 @@ import (
 	"github.com/tlinden/anydb/common"
 )
 
-var Version string = "v0.0.6"
+var Version string = "v0.0.7"
 
 type BucketConfig struct {
 	Encrypt bool
 }
 
 type Config struct {
-	Debug      bool
+	Debug           bool
-	Dbfile     string
+	Dbfile          string
-	Dbbucket   string
+	Dbbucket        string
-	Template   string
+	Template        string
-	Mode       string // wide, table, yaml, json
+	Mode            string // wide, table, yaml, json
-	NoHeaders  bool
+	NoHeaders       bool
-	NoHumanize bool
+	NoHumanize      bool
-	Encrypt    bool // one entry
+	Encrypt         bool // one entry
-	Listen     string
+	CaseInsensitive bool
-	Buckets    map[string]BucketConfig // config file only
+	Listen          string
+	Buckets         map[string]BucketConfig // config file only
 
 	Tags []string // internal
 	DB   *app.DB  // internal

cmd/anydb.go

@@ -1,19 +1,3 @@
-/*
-Copyright © 2024 Thomas von Dein
-
-This program is free software: you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation, either version 3 of the License, or
-(at your option) any later version.
-
-This program is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with this program. If not, see <http://www.gnu.org/licenses/>.
-*/
 package cmd
 
 var manpage = `
@@ -28,14 +12,20 @@ SYNOPSIS
         Available Commands:
           completion  Generate the autocompletion script for the specified shell
           del         Delete key
+          edit        Edit a key
           export      Export database to json
           get         Retrieve value for a key
           help        Help about any command
           import      Import database dump
+          info        info
           list        List database contents
+          man         show manual page
+          serve       run REST API listener
           set         Insert key/value pair
     
         Flags:
+          -b, --bucket string   use other bucket (default: data) (default "data")
+          -c, --config string   toml config file
           -f, --dbfile string   DB file to use (default "/home/scip/.config/anydb/default.db")
           -d, --debug           Enable debugging
           -h, --help            help for anydb
@@ -44,9 +34,479 @@ SYNOPSIS
         Use "anydb [command] --help" for more information about a command.
 
 DESCRIPTION
-    Anydb is a simple to use commandline tool to store anything you'd like,
+    Anydb is a commandline personal key value store, it is simple to use and
-    even binary files etc. It uses a key/value store (bbolt) in your home
+    can be used to store anything you'd like, even binary files etc. It uses
-    directory.
+    a key/value store (bbolt) in your home directory.
+
+    The tool provides a number of subcommands to use it, there are global
+    options and each subcommand has its own set of options.
+
+GLOBAL OPTIONS
+    "-f, --dbfile filename"
+        The default location of your databas is
+        "$HOME/.config/anydb/default.db". You can change this with the "-f"
+        option.
+
+    "-b, --bucket name"
+        Data in a bbolt key-value-store are managed in so called buckets.
+        These are kind of namespaces, where each key must be unique.
+        However, a database may contain more than one bucket.
+
+        By default anydb uses a bucket named "data", but you can change this
+        using the option "-b".
+
+        Buckets can be configured to always encrypt values, see ENCRYTPTION.
+
+    "-c, --config filename"
+        Under normal circumstances you don't need a configuration file. But
+        if you want, you can provide one using the option "-c".
+
+        Anydb looks for a couple of default locations for a config file. You
+        only need this option if you want to supply a configuration on a
+        non-standard location. See CONFIGURATION for more details.
+
+    "-d, --debug"
+        Enable debug output.
+
+    "-h, --help"
+        Show the usage of anydb.
+
+    "-v, --version"
+        Show the program version.
+
+    All of these options can be used with subcommands as well.
+
+SUBCOMMANDS
+  completion
+    The completion command can be used to setup completion for anydb. Just
+    put something like this into your shell's configuration file:
+
+        source <(anydb completion bash)
+
+    If you use another shell, specify it instead of bash, of course.
+
+  set
+    The set command is being used to insert or update a key-value pair.
+
+    Usage:
+
+        Usage:
+          anydb set <key> [<value> | -r <file>] [-t <tag>] [flags]
+    
+        Aliases:
+          set, add, s, +
+    
+        Flags:
+          -e, --encrypt            encrypt value
+          -r, --file string        Filename or - for STDIN
+          -h, --help               help for set
+          -t, --tags tag,tag,...   tags, multiple allowed
+
+    The standard way to insert a new entry is really simple:
+
+        anydb set key value
+
+    If you don't specify a value, anydb expects you to feed it some data via
+    STDIN. For example:
+
+        anydb set key < file
+
+    You might as well specify a file directly using the "-f" option:
+
+        anydb set key -f file
+
+    Values can be encrypted using ChaCha20Poly1305 when you specify the "-e"
+    option. Anydb will ask you interactively for a passphrase. You might as
+    well provide the passphrase using the environment variable
+    "ANYDB_PASSWORD". To encrypt the value, a cryptographically secure key
+    will be derived from the passphrase using the ArgonID2 algorithm. Each
+    value can be encrypted with another passphrase. So, the database itself
+    is not encrypted, just the values.
+
+    You can supply tags by using the option "-t". Multiple tags can be
+    provided either by separating them with a comma or by using multiple
+    "-t" parameters:
+
+        anydb set key value -t tag1,tag2
+        anydb set key value -t tag1 -t tag2
+
+    You can later filter entries by tag or by a combination of tags.
+
+    To edit or modify an entry, just use the set command with the same key,
+    the value in the database will be overwritten with the new value. An
+    alternative option is the edit command, see below.
+
+  get
+    To retrieve the value of a key, use the get subcommand.
+
+    Usage:
+
+        Usage:
+          anydb get  <key> [-o <file>] [-m <mode>] [-n -N] [-T <tpl>] [flags]
+    
+        Aliases:
+          get, show, g, .
+    
+        Flags:
+          -h, --help              help for get
+          -m, --mode string       output format (simple|wide|json|template) (default 'simple')
+          -n, --no-headers        omit headers in tables
+          -N, --no-human          do not translate to human readable values
+          -o, --output string     output value to file (ignores -m)
+          -T, --template string   go template for '-m template'
+
+    In its simplest form you just call the get subcommand with the key you
+    want to have the value for. The value is being printed to STDOUT by
+    default:
+
+        anydb get key
+
+    If the value is binary content, it will not just being printed. In those
+    cases you need to either redirect output into a file or use the option
+    "-o" to write to a file:
+
+        anydb get key > file
+        anydb get key -o file
+
+    If the value is encrypted, you will be asked for the passphrase to
+    decrypt it. If the environment variable "ANYDB_PASSWORD" is set, its
+    value will be used instead.
+
+    There are different output modes you can choos from: simple, wide and
+    json. The "simple" mode is the default one, it just prints the value as
+    is. The "wide" mode prints a tabular output similar to the list
+    subcommand, see there for more details. The options "-n" and "-N" have
+    the same meaning as in the list command. The "json" mode prints the raw
+    JSON representation of the whole database entry. Decryption will only
+    take place in "simple" and "json" mode. The "template" mode provides the
+    most flexibily, it is detailed in the section TEMPLATES.
+
+  list
+    The list subcommand displays a list of all database entries.
+
+    Usage:
+
+        Usage:
+          anydb list  [<filter-regex>] [-t <tag>] [-m <mode>] [-n -N] [-T <tpl>] [flags]
+    
+        Aliases:
+          list, /, ls
+    
+        Flags:
+          -h, --help               help for list
+          -m, --mode string        output format (table|wide|json|template), wide is a verbose table. (default 'table')
+          -n, --no-headers         omit headers in tables
+          -N, --no-human           do not translate to human readable values
+          -t, --tags stringArray   tags, multiple allowed
+          -T, --template string    go template for '-m template'
+          -l, --wide-output        output mode: wide
+
+    In its simplest form - without any options - , the list command just
+    prints all keys with their values to STDOUT. Values are being truncated
+    to maximum of 60 characters, that is, multiline values are not
+    completely shown in order to keep the tabular view readable.
+
+    To get more informations about each entry, use the "-o wide" or "-l"
+    option. In addition to the key and value also the size, update timestamp
+    and tags will be printed. Time and size values are converted into a
+    human readable form, you can suppress this behavior with the "-N"
+    option. You may omit the headers using the option "-n"
+
+    Sometimes you might want to filter the list of entries. Either because
+    your database grew too large or because you're searching for something.
+    In that case you have two options: You may supply one or more tags or
+    provide a filter regexp. To filter by tag, do:
+
+        anydb list -t tag1
+        anydb list -t tag1,tag2
+        anydb list -t tag1 -t tag2
+
+    To filter using a regular expression, do:
+
+       anydb list "foo.*bar"
+
+    Regular expressions follow the golang re2 syntax. For more details about
+    the syntax, refer to <https://github.com/google/re2/wiki/Syntax>. Please
+    note, that this regexp dialect is not PCRE compatible, but supports most
+    of its features.
+
+    You can - as with the get command - use other output modes. The default
+    mode is "table". The "wide" mode is, as already mentioned, a more
+    detailed table. Also supported is "json" mode and "template" mode. For
+    details about using templates see TEMPLATES.
+
+  del
+    Use the del command to delete database entries.
+
+    Usage:
+
+        Usage:
+          anydb del <key> [flags]
+    
+        Aliases:
+          del, d, rm
+    
+        Flags:
+          -h, --help   help for del
+
+    The subcommand del does not provide any further options, it just deletes
+    the entry referred to by the given key. No questions are being asked.
+
+  edit
+    The edit command makes it easier to modify larger entries.
+
+    Usage:
+
+       Usage:
+          anydb edit <key> [flags]
+    
+        Aliases:
+          edit, modify, mod, ed, vi
+    
+        Flags:
+          -h, --help   help for edit
+
+    The subcommand edit does not provide any further options. It works like
+    this:
+
+    1. Write the value info a temporary file.
+    2. Execute the editor (which one, see below!) with that file.
+    3. Now you can edit the file and save+close it when done.
+    4. Anydb picks up the file and if the content has changed, puts its
+    value into the DB.
+
+    By default anydb executes the "vi" command. You can modify this behavior
+    by setting the environment variable "EDITOR" appropriately.
+
+    Please note, that this does not work with binary content!
+
+  export
+    Since the bbold database file is not portable across platforms (it is
+    bound to the endianess of the CPU it was being created on), you might
+    want to create a backup file of your database. You can do this with the
+    export subcommand.
+
+    Usage:
+
+        Usage:
+          anydb export [-o <json filename>] [flags]
+    
+        Aliases:
+          export, dump, backup
+    
+        Flags:
+          -h, --help            help for export
+          -o, --output string   output to file
+
+    The database dump is a JSON representation of the whole database and
+    will be printed to STDOUT by default. Redirect it to a file or use the
+    "-o" option:
+
+        anydb export > dump.json
+        anydb export -o dump.json
+
+    Please note, that encrypted values will not be decrypted. This might
+    change in a future version of anydb.
+
+  import
+    The import subcommand can be used to restore a database from a JSON
+    dump.
+
+    Usage:
+
+        Usage:
+          anydb import [<json file>] [flags]
+    
+        Aliases:
+          import, restore
+    
+        Flags:
+          -r, --file string        Filename or - for STDIN
+          -h, --help               help for import
+          -t, --tags stringArray   tags, multiple allowed
+
+    By default the "import" subcommand reads the JSON contents from STDIN.
+    You might pipe the dump into it or use the option "-r":
+
+        anydb import < dump.json
+        anydb import -r dump.json
+        cat dump.json | anydb import
+
+    If there is already a database, it will be saved by appending a
+    timestamp and a new database with the contents of the dump will be
+    created.
+
+  serve
+    Anydb provides a RESTful API, which you can use to manage the database
+    from somewhere else. The API does not provide any authentication or any
+    other security measures, so better only use it on localhost.
+
+    Usage:
+
+        Usage:
+          anydb serve [-l host:port] [flags]
+    
+        Flags:
+          -h, --help            help for serve
+          -l, --listen string   host:port (default "localhost:8787")
+
+    To start the listener, just execute the serve subcommand. You can tweak
+    the ip address and tcp port using the "-l" option. The listener will not
+    fork and run in the foreground. Logs are being printed to STDOUT as long
+    as the listener runs.
+
+    For more details about the API, please see the "REST API" section.
+
+  info
+    The info subcommand shows you some information about your current
+    database.
+
+    Usage:
+
+        Usage:
+          anydb info [flags]
+    
+        Flags:
+          -h, --help       help for info
+          -N, --no-human   do not translate to human readable values
+
+    Data being shown are: filename and size, number of keys per bucket. If
+    you supply the "-d" option (debug), some bbolt internals are being
+    displayed as well.
+
+  man
+    The man subcommand shows an unformatted text variant of the manual page
+    (which are currently reading).
+
+    Usage:
+
+        Usage:
+          anydb man [flags]
+    
+        Flags:
+          -h, --help   help for man
+
+    The manual is being piped into the "more" command, which is being
+    expected to exist according to the POSIX standard on all supported unix
+    platforms. It might not work on Windows.
+
+TEMPLATES
+    The get and list commands support a template feature, which is very
+    handy to create you own kind of formatting. The template syntax being
+    used is the GO template language, refer to
+    <https://pkg.go.dev/text/template> for details.
+
+    Each template operates on one or more entries, no loop construct is
+    required, the template provided applies to every matching entry
+    separatley.
+
+    The following template variables can be used:
+
+    Key - string =item Value - string =item Bin - []byte =item Created -
+    time.Time =item Tags - []string =item Encrypted bool
+
+    Prepend a single dot (".") before each variable name.
+
+    Here are some examples how to use the feature:
+
+    Only show the keys of all entries:
+
+        anydb list -m template -T "{{ .Key }}"
+
+    Format the list in a way so that is possible to evaluate it in a shell:
+
+        eval $(anydb get foo -m template -T "key='{{ .Key }}' value='{{ .Value }}' ts='{{ .Created}}'")
+        echo "Key: $key, Value: $value"
+
+    Print the values in CSV format ONLY if they have some tag:
+
+        anydb list -m template -T "{{ if .Tags }}{{ .Key }},{{ .Value }},{{ .Created}}{{ end }}"
+
+CONFIGURATION
+    Anydb looks at the following location for a configuration file, in that
+    order:
+
+    "$HOME/.config/anydb/anydb.toml"
+    "$HOME/.anydb.toml"
+    "anydb.toml" in the current directory
+    or specify one using "-c"
+        The configuration format uses the TOML language, refer to
+        <https://toml.io/en/> for more details. The key names correspond to
+        the commandline options in most cases.
+
+        Configuration follows a certain precedence: the files are tried to
+        be read in the given order, followed by commandline options. That
+        is, the last configuration file wins, unless the user provides a
+        commandline option, then this setting will be taken.
+
+        A complete configuration file might look like this:
+
+            # defaults
+            dbfile     = "~/.config/anydb/default.db"
+            dbbucket   = "data"
+            noheaders  = false
+            nohumanize = false
+            encrypt    = false
+            listen     = "localhost:8787"
+    
+            # different setups for different buckets
+            [buckets.data]
+            encrypt = true
+    
+            [buckets.test]
+            encrypt = false
+
+        Under normal circumstances you don't need a configuration file.
+        However, if you want to use different buckets, then this might be a
+        handy option. Buckets are being configured in ini-style with the
+        term "bucket." followed by the bucket name. In the example above we
+        enable encryption for the default bucket "data" and disable it for a
+        bucket "test". To use different buckets, use the "-b" option.
+
+REST API
+    The subcommand serve starts a simple HTTP service, which responds to
+    RESTful HTTP requests. The listener responds to all requests with a JSON
+    encoded response. The response contains the status and the content - if
+    any - of the requested resource.
+
+    The following requests are supported:
+
+    GET /anydb/v1/
+        Returns a JSON encoded list of all entries.
+
+    GET /anydb/v1/key
+        Returns the JSON encoded entry, if found.
+
+    PUT /anydb/v1/
+        Create an entry. Expects a JSON encoded request object in POST data.
+
+    DELETE /anydb/v1/key
+        Delete an entry.
+
+    Some curl example calls to the API:
+
+    Post a new key: curl -X PUT localhost:8787/anydb/v1/ \ -H 'Content-Type:
+    application/json' \ -d '{"key":"foo","val":"bar"}'
+
+    Retrieve the value:
+
+        curl localhost:8787/anydb/v1/foo
+
+    List all keys:
+
+        curl localhost:8787/anydb/v1/
+
+BUGS
+    In order to report a bug, unexpected behavior, feature requests or to
+    submit a patch, please open an issue on github:
+    <https://github.com/TLINDEN/anydb/issues>.
+
+    Please repeat the failing command with debugging enabled "-d" and
+    include the output in the issue.
+
+LIMITATIONS
+    The REST API list request doesn't provide any filtering capabilities
+    yet.
 
 LICENSE
     This software is licensed under the GNU GENERAL PUBLIC LICENSE version

cmd/crud.go

@@ -138,7 +138,7 @@ func Get(conf *cfg.Config) *cobra.Command {
 	}
 
 	cmd.PersistentFlags().StringVarP(&attr.File, "output", "o", "", "output value to file (ignores -m)")
-	cmd.PersistentFlags().StringVarP(&conf.Mode, "mode", "m", "", "output format (simple|wide|json) (default 'simple')")
+	cmd.PersistentFlags().StringVarP(&conf.Mode, "mode", "m", "", "output format (simple|wide|json|template) (default 'simple')")
 	cmd.PersistentFlags().BoolVarP(&conf.NoHeaders, "no-headers", "n", false, "omit headers in tables")
 	cmd.PersistentFlags().BoolVarP(&conf.NoHumanize, "no-human", "N", false, "do not translate to human readable values")
 	cmd.PersistentFlags().StringVarP(&conf.Template, "template", "T", "", "go template for '-m template'")
@@ -188,7 +188,7 @@ func List(conf *cfg.Config) *cobra.Command {
 	)
 
 	var cmd = &cobra.Command{
-		Use:   "list  [<filter-regex>] [-t <tag>] [-m <mode>] [-n -N] [-T <tpl>]",
+		Use:   "list  [<filter-regex>] [-t <tag>] [-m <mode>] [-n -N] [-T <tpl>] [-i]",
 		Short: "List database contents",
 		Long:  `List database contents`,
 		RunE: func(cmd *cobra.Command, args []string) error {
@@ -196,7 +196,11 @@ func List(conf *cfg.Config) *cobra.Command {
 			cmd.SilenceUsage = true
 
 			if len(args) > 0 {
-				attr.Args = args
+				if conf.CaseInsensitive {
+					attr.Args = []string{"(?i)" + args[0]}
+				} else {
+					attr.Args = args
+				}
 			}
 
 			// turn comma list into slice, if needed
@@ -217,11 +221,12 @@ func List(conf *cfg.Config) *cobra.Command {
 		},
 	}
 
-	cmd.PersistentFlags().StringVarP(&conf.Mode, "mode", "m", "", "output format (table|wide|json), wide is a verbose table. (default 'table')")
+	cmd.PersistentFlags().StringVarP(&conf.Mode, "mode", "m", "", "output format (table|wide|json|template), wide is a verbose table. (default 'table')")
 	cmd.PersistentFlags().StringVarP(&conf.Template, "template", "T", "", "go template for '-m template'")
 	cmd.PersistentFlags().BoolVarP(&wide, "wide-output", "l", false, "output mode: wide")
 	cmd.PersistentFlags().BoolVarP(&conf.NoHeaders, "no-headers", "n", false, "omit headers in tables")
 	cmd.PersistentFlags().BoolVarP(&conf.NoHumanize, "no-human", "N", false, "do not translate to human readable values")
+	cmd.PersistentFlags().BoolVarP(&conf.CaseInsensitive, "case-insensitive", "i", false, "filter case insensitive")
 	cmd.PersistentFlags().StringArrayVarP(&attr.Tags, "tags", "t", nil, "tags, multiple allowed")
 
 	cmd.Aliases = append(cmd.Aliases, "/")

t/workflow.txtar

@@ -18,6 +18,9 @@
 # simple entry
 exec anydb -f test.db set foo bar
 
+# single entry uc()
+exec anydb -f test.db set MUCHAS gracias
+
 # entry with tags
 exec anydb -f test.db set color grey -t flower,plant
 
@@ -37,6 +40,10 @@ exec anydb -f test.db list -t flower
 exec anydb -f test.db list b.r
 stdout bar
 
+# list with -i filter
+exec anydb -f test.db list -i mucha
+stdout MUCHA
+
 # get single entry
 exec anydb -f test.db get color
 stdout grey