config.example.yaml

# verbose is a boolean flag that governs whether detailed log information are
# to be displayed
# It can be set at runtime using the '--debug' flag
#verbose: false
#
# debug is a boolean flag that governs whether debug information are
# to be displayed
# It can be set at runtime using the '--debug' flag
#debug: false

# readonly is a boolean flag that governs whether gostore is allowed to
# alter the content of the collection. It is typically useful to dry-run
# some actions.
# It can be set at runtime using the '--pretend' flag
#readonly:   false

# DeleteGhosts is a flag that instructs gostore.Check to delete any  
# database entries that does not correspond to an existing file in th
# store's filesystem (so called ghost record)
# It can be set at runtime using '--delete-ghosts' flag
#deleteGhosts: false
                    
# deleteOrphans is a flag that instructs gostore.Check to delete any
# file of the store's filesystem that is not recorded in the store's
# database.
# It can be set at runtime using '--delete-orphans' flag
#deleteOrphans: false
                                            
# importOrphans is a flag that instructs gostore.Check to re-import a
# file of the store's filesystem that is not recorded in the store's
# database.
# It can be set at runtime using '--import-orphans' flag
#importOrphans: false


# store contains any customization to manage the way the collection is stored
store:
    # path is the path to the collection's root folder. The database, index and
    # media files are stored inside this root folder.
    # It can be set at runtime using the '--root=xxx' flag.
    # If root is not set, it defaults to the current working directory.
    path: /tmp/gostore-test/

    # indexingAnalyzer identify the analyzer used to index the media metadata when
    # searching your collection.
    # To take benefit of a new analyzer you usually have to rebuild an existing index.
    # Available analyzers can be listed using `gostore config`.
    indexingAnalyzer: fr


# ui contains any customization to manage the way gostore interacts with the
# user
ui:
    # auto is a boolean flag that governs whether user manual interaction (like
    # metadata editing) is expected.
    # It can be set at runtime using the '--auto' flag
    #auto: false
    
    # editorcmd contains the invocation stanza to fire-up a text editor
    # allowing the user to modify metadata.
    # editorcmd accepts one argument which is the name of file to be edited.
    # editorcmd can rely on environment variable.
    # If editorcmd is not set, metadata manual modification is skipped.
    editorcmd: $EDITOR "%s"
          
    # mergercmd contains the invocation stanza to fire-up a text editor
    # allowing the user to mergel two set of metadata.
    # mergecmd accepts two argument which is the name of files to be merged.
    # mergecmd can rely on environment variable.
    # If mergercmd is not set, metadata manual modification is skipped.
    mergercmd: vim "%[2]s" -c "vert diffsplit %[1]s"

    # outputformat points to the style to print-out metadata. It should be one
    # of the defined formatters.
    # If not set or set to a non-specified formatter, it defaults to listing
    # media file name.
    # It can be set at runtime using '--style=xxx' flag.
    outputformat: full

    # formatters is the set of available styles for printing media metadata. For
    # each style, customized template can be proposed for a given media type.
    # Default template can be defined using the "media" keyword.
    #
    # Templates follow golang text/template specification.
    # Stored metadata can be retrieved using their field name. List of known
    # fields can be obtained using `gostore fields`.
    # Extensions to template idiom are available to simplify common operations:
    # - tmpl: execute a sub-template by name
    # - tmplExec: execute a template text
    # - tmplFile: execute a template stored in a file
    # - get/getAll: filter out specific fields from a set of medias and order them accordingly.
    #      Specific indications can be provided:
    #      . "*": any remaining fields
    #      . "?xxx"/"?*": only display the field if non empty
    #      . "!xxx": do not display the information
    #  - extend: create a new field that can be called afterwards to be displayed.
    #  - json/jsonForHuman: convert to json
    #
    # It is also possible to define intermediate sub-templates and re-use
    # them.
    formatters:
        list:
            media:  '{{ getAll . "Name" "Title" "Authors" | bold | byrow }}'
            book:   '{{ getAll . "Name" "Title" "?SubTitle" "?Serie" "?SeriePosition" "Authors" | bold | byrow }}'

        full:
            media: |
                {{ range $i, $r := . -}}
                {{ if $i }}{{ printf "\n\n" }}{{ end -}}
                {{ get $r "Name" "Title" "?*" "Type" "?QALevel" "?SourceHash" "CreatedAt" "UpdatedAt" | bold | bycol -}}
                {{ end -}}

            book: |
                {{ range $i, $r := . -}}
                {{ if .Serie }}{{ extend . "SerieName" (tmplExec "{{.Serie}} {{.SeriePosition}}" .)}}{{ end -}}
                {{ if $i }}{{ printf "\n\n" }}{{ end -}}
                {{ get $r "Name" "Title" "?SubTitle" "?SerieName" "!Serie" "!SeriePosition" "Authors" "Description" "?*" "Type" "?QALevel" "?SourceHash" "CreatedAt" "UpdatedAt" | bold | bycol -}}
                {{ end -}}

        json:
            media: '{{ json . }}'

        #html:
        #   media:
        #       {{ tmplFile "html.tmpl" . }}

# import lists the different modules to operate on metadata during the import step.
# Modules are run in the provided order.
# List of modules is available using `gostore config`.
import:
    # hasher computes a checksum-based signature for the record's file and make
    # sure it does not already exist in the store.
    - name: hasher
      config:
         # HashMethod specifies the hash algorithm to use.
         # Available methods are: md5, sha1, sha256. Default to md5
         #HashMethod: md5

    # mdatareader reads metadata from a media file and populates the
    # corresponding record's values.
    - name : mdatareader     
      
    # fetcher a module that retrieves metadata from online databases.
    - name: fetcher
      
    # scrubber is a module that removes any fields a media metadata.
    - name: scrubber
      config: &scrubber
          # fields is the list of metadata fields to scrub
          fields: [ Comment ]
      
    # dehtmlizer is a modules that removes any html tags from a media metadata.
    # It can optionally convert the html tagging into another idioms (like
    # markdown)
    - name: dehtmlizer
      config: &dehtmlizer
          # fields is the list of metadata fields to clean from html tags
          fields: [ Description ]

          # outputstyle is the idiom in which html tagging is converted to.
          # Available formats are plain text (default) or markdown
          outputstyle: markdown
          
    # dupfinder is a module that checks whether a record is already in the
    # collection before importing it.
    - name: dupfinder
      config:
          # DupQueries is a collection of queries templates that identify records'
          # duplicates.
          # Extensions to template idiom are available to simplify common operations:
          # - escape: escape special characters in query phrase or names.
          # 
          # Queries, once expanded, should follow https://blevesearch.com/docs/Query-String-Query/
          dupqueries:
              - 'ISBN:{{ escape .ISBN }}'
              - |
                {{if .Title}}+Title:{{ escape .Title }}~1.2{{end -}}
                {{range $i, $a := .Authors }} +Authors:{{ escape $a }}~1.2{{end -}}
                {{if .Publisher}} +Publisher:{{ escape .Publisher }}~1.2{{end -}}
          
    # checker is a module that verifies that a metadata set has a good quality
    # level.
    - name: checker
      config: &checker
          # fieldName is the name of the metadata's field to store the quality
          # level assessment.
          fieldName: QALevel

          # minLevel is the minimum allowed level. Any level below this threshold
          # will result in an error aborting operation. Default to 0 (all quality
          # level is accepted)
          minLevel: 30

    # normalizer is a module that proposes normalized field values based
    # on already similar existing records' values.
    - name: normalizer
      config:
          # Fields is the list of fields where you expect normalization to happen.
          # Normalisation is only working for fields featuring strings (or list of
          # strings) like Authors, Publisher...
          fields: [ Authors, Publisher, Subject ]

          # SimilarityLevel is the measure of similarity that is accepted
          # between two records. Default to 1.
          # .   0: the input text is analyzed first. An attempt is made to use the same
          #        analyzer that was used when the field was indexed.
          # . > 0: the input text is analysed first, the match is done with the given
          #        level of fuzziness.
          # Default to 0.
          # SimilarityLevel: 0
    
    # organizer is a module that rationalizes the name of a media based on its
    # metadata values
    - name: organizer
      config: &organizer
          # namingschemes is a collection of templates to define a media name
          # based on its metadata values.
          #
          # Different naming schemes can be adopted based on the media's type.
          # Fall-back templates or generic template can be defined using the
          # "media" keyword.
          # 
          # Templates follow golang text/template specification.
          # Stored metadata can be retrieved using their field name. List of known
          # fields can be obtained using `gostore fields`.
          # Extensions to template idiom are available to simplify common operations:
          # - tmpl: execute a sub-template by name
          # - tmplExec: execute a template text
          # - tmplFile: execute a template stored in a file
          # - ext: get file extension
          # - sanitizePath: replace special chars that don't usually work well when
          #   use in path name
          # - nospace: get rid of spaces
          #
          # It is also possible to define intermediate sub-templates and re-use
          # them.
          namingschemes:
              author: "{{if .Authors}}{{index .Authors 0}}{{end}}"
              serie:  "{{if .Serie}}{{.Serie}} {{.SeriePosition}}{{end}}"
              book: |
                    {{ $a := (tmpl "author" .) -}}
                    {{ $s := (tmpl "serie" .)  -}}
                    {{ print (or $a "unknown") (and $s (printf " - [%s]" $s)) " - " .Title (ext .Name) | sanitizePath -}}



# update lists the different modules to operate on metadata during the update
# step. Modules are run in the provided order.
# List of modules is available using `gostore config`.
update:
    - name: scrubber
      config:
          <<: *scrubber

    - name: dehtmlizer
      config:
        <<: *dehtmlizer

    - name: organizer
      config:
          <<: *organizer

    - name: checker
      config:
          <<: *checker