This section of options on the right side of the Options Panel allows the user to control which reports will be generated by jMusicMetaManager and, in some cases, what parameters will be used to generate them. Some of these reports profile a music collection and some describe probable metadata errors in it. Some reports also describe the details of how the metadata was processed.

Note that most reports list items alphabetically, and these alphabetical listings typically list lowercase characters after all uppercase characters.

This section is itself divided into seven sub-sections:

1) Reports Relating to iTunes/ID3 Parsing and Merging

The preferences in this section control which reports are generated relating to the parsing of an Apple iTunes XML file and/or the extraction of ID3 tags from MP3 files, as well as how the metadata from the two sources is combined.

These reports are in general not particularly relevant to music collection profiling or error checking. They are, however, useful for validating the correspondence between the two sources of metadata and for helping to debug processing at future stages.

The specific options in this section are as follows:

• List MP3 files found that could not be parsed: Whether or not the MP3 files found that could not be parsed report is generated. This report lists all files with an MP3 extension that were found in the directory specified in the Browse MP3 Directory text field (or its subdirectories) whose ID3 tags could not be parsed. It should be noted that jMusicMetaManager skips all such files during later processing.
• List all non-MP3 files found: Whether or not the Non-MP3 files found report is generated. This report lists all files in the directory specified in the Browse MP3 Directory text field (or its subdirectories) that do not have an MP3 extension. The case of the extension is ignored. It should be noted that these files do not have any effect on processing, as they are ignored by jMusicMetaManager other than for the purpose of this report. Only the extension of the file is considered when generating this report, not actual file content.
• List all recordings parsed (before merge): Whether or not the All recordings parsed (before merge) report is generated. This report lists the path, title and artist of all recordings for which metadata was extracted, before any processing is performed. Two separate lists are produced, one for the recordings extracted from ID3 tags and one for recordings extracted from an iTunes XML file.
• List files in iTunes XML but not at specified path: Whether or not the Files in iTunes XML but not at specified path report is generated. This report lists each recording that is parsed from the iTunes XML file for which there is no valid readable file at the specified path. No verification is performed during the generation of this particular report of whether files that are present are in fact valid MP3 files, as this information is available in other reports. Unlike the Recordings in one source but not the other report, this report may be generated even if the Use MP3s option is not selected.
• List recordings in one source but not the other: Whether or not the Recordings in one source but not the other report is generated. This report lists any recordings that were found in the iTunes XML file but for which corresponding MP3 files were not found, and vice versa.
• Report fields that do not correspond between sources: Whether or not the Fields that do not correspond between sources report is generated. This report lists any differences in significant fields in the metadata for each recording parsed from the iTunes XML file and its corresponding MP3 file. It may be appropriate to review this metadata to ensure that it is consistent. jMusicMetaManager uses the metadata extracted from the iTunes file by default when there is such a disagreement.
• List all post-iTunes and ID3 merge metadata: Whether or not the All post-iTunes and ID3 merge metadata report is generated. This report lists all available metadata for each recording immediately after the metadata extracted from the iTunes file and/or the MP3 files has been combined. If an entry in the iTunes file contains a reference to a file with a path that is also found during ID3 parsing, then the metadata from these two sources is combined into a single recording. jMusicMetaManager uses the metadata extracted from the iTunes file by default when there is a disagreement between the fields of any two recordings from different sources with the same path.

2) General Profiles of Music Collection

The reports referred to in this section provide general profiles of a music collection that has been parsed by jMusicMetaManager. These reports are useful for providing statistics describing the makeup of music collections. Further profiles are available in the Profiles of Albums in Music Collection section.

The reports in this section are generated after metadata parsed from iTunes XML files and MP3 ID3 tags have been merged, but before any find/replace, reordered word subset or edit distance operations have been performed. Fields with identical values have not yet been merged when these reports are generated, as only recordings with identical file paths have been merged at the point when these reports are generated.

The specific options in this section are as follows:

• List all recordings post-sorting: Whether or not the All recordings post-sorting report is generated. This report gives all available metadata for each recording. This report is similar to the All post-iTunes and ID3 merge metadata report, except that the recordings are sorted by title.
• Report artist breakdown: Whether or not the Artist breakdown report is generated. This report lists the unique names of all artists found, as well as the number of recordings and the percentage of total recordings corresponding to each artist. The number of unique composers, albums and genres that have at least one recording corresponding to each artist is also reported. The tallies listed do not include counts for unknown composers, genres and albums.
• Report composer breakdown: Whether or not the Composer breakdown report is generated. This report lists the unique names of all composers found, as well as the number of recordings and the percentage of total recordings corresponding to each composer. The number of unique artists, albums and genres that have at least one recording corresponding to each composer is also reported. The tallies listed do not include counts for unknown artists, genres and albums.
• Report genre breakdown: Whether or not the Genre breakdown report is generated. This report lists the unique names of all genres found, as well as the number of recordings and the percentage of total recordings belonging to each genre. The number of unique artists, composers and albums that have at least one recording belonging to each genre is also reported. The tallies listed do not include counts for unknown artists, composers and albums.
• Report comment statistics: Whether or not the Comment statistics report is generated. This report lists all unique file comments found, as well as the number of recordings and the percentage of total recordings corresponding to each comment. The number of unique artists, composers, albums and genres that have at least one recording corresponding to each comment is also reported. The tallies listed do not include counts for unknown artists, composers, genres and albums.
• List artists by genre: Whether or not the Artists listed by genre report is generated. This report alphabetically lists all unique genres. Artists are listed under a given genre if they have at least one recording belonging to the genre. Artists also have the number of recordings they have belonging to the given genre listed beside their entries, along with the percentage of the total number of recordings in the genre that they represent.
• List composers by genre: Whether or not the Composers listed by genre report is generated. This report alphabetically lists all unique genres. Composers are listed under a given genre if they have at least one recording belonging to the genre. Composers also have the number of recordings they have belonging to the given genre listed beside their entries, along with the percentage of the total number of recordings in the genre that they represent.
• Report recordings missing key metadata: Whether or not the Recordings missing key metadata report is generated. This report lists all metadata for those recordings that have empty title, artist, composer, album and/or genre fields. A separate table is produced for each of these fields.
• List artists with few recordings: Whether or not the artist table of the Artists and composers with few recordings report is generated. This report alphabetically lists artists with fewer recordings than the number specified in the corresponding Cutoff (2-90) text box. Artists with fewer than half this number of recordings are listed in bold.
• List composers with few recordings: Whether or not the composer table of the Artists and composers with few recordings report is generated. This report alphabetically lists composers with fewer recordings than the number specified in the corresponding Cutoff (2-90) text box. Composers with fewer than half this number of recordings are listed in bold.

3) Miscellaneous Reports

The reports referred to in this section provide miscellaneous information relating to a music collection that has been processed by jMusicMetaManager.

The reports in this section are generated after metadata parsed from iTunes XML files and MP3 ID3 tags have been merged, but before any find/replace, reordered word subset or edit distance operations have been performed. Fields with identical values have not yet been merged when these reports are generated, as only recordings with identical file paths have been merged at the point when these reports are generated.

The specific options in this section are as follows:

• Report exactly identical recording titles: Whether or not the Exactly identical recording titles report is generated. This report lists all recordings that have exactly the same title (before any processing). This is useful in seeing how many versions of the same piece exist in the music database. These duplicates may be identical copies of the same recording occurring in different places, or they may be different versions of the same piece. The Probable duplicates of the same recording report can be consulted to detect which of these recordings are likely multiple copies of the same recording. The Exactly identical recording titles report begins with several overall statistics, followed by a list of all titles for which there is more than one recording with the same name. Finally, the details of all such recordings in each identical title cluster are reported.
• Report fields starting with a space: Whether or not the Fields starting with a space report is generated. This report separately lists all titles, artists, composers, albums and genres that start with a space. Each such field is only listed once when multiple fields are identical. This report is useful in immediately highlighting this particularly common error.

4) Profiles of Albums in Music Collection

The reports referred to in this section provide general profiles of the albums in a music collection that has been parsed by jMusicMetaManager, as well as reports of detected metadata errors relating to track numbers, compilation albums and missing year metadata. More general profiles are available in the General Profiles of Music Collection section.

jMusicMetaManager uses the notion of compilation albums that is most consistent with that used by Apple iTunes. This means that every album that contains at least two tracks with different values in their respective artist fields is considered a compilation album, and every track in such an album should be marked as a being part of a compilation in its metadata. A greatest hits compilation of a single band, for example, is not considered a compilation under this definition. Individual users are of course free to adopt any definition of a compilation album that they wish, but this is the definition used by jMusicMetaManager in its compilation error checking. It should be noted that Apple iTunes places all tracks marked as compilations in a special folder named “Compilations” that exists at the same level as artist folders.

The error checking used by jMusicMetaManager also checks to see if each recording has a track number as well as a corresponding number of tracks in its album. Multi-disc albums are considered as a single unity by jMusicMetaManager if their disc number and discs in album fields are properly set. Empty values for these two fields are assumed to correspond to single disc albums.

The reports in this section are generated after metadata parsed from iTunes XML files and MP3 ID3 tags have been merged, but before any find/replace, reordered word subset or edit distance operations have been performed. Fields with identical values have not yet been merged when these reports are generated, as only recordings with identical file paths have been merged at the point when these reports are generated.

The specific options in this section are as follows:

• List albums by artist: Whether or not the Albums listed by artist report is generated. This report alphabetically lists each unique artist present. All albums present that have at least one recording by each given artist are listed under that artist's entry. Note is also made of how many tracks by each given artist belonging to the given album are present in the music collection, how many total tracks the album is marked as containing (cumulative for mult-disc sets, if annotated), what percentage of the total tracks are by the given artist (marked in bold if not equal to 100%) and whether each album is marked as a compilation.
• List albums by composer: Whether or not the Albums listed by composer report is generated. This report alphabetically lists each unique composer present. All albums present that have at least one recording by each given composer are listed under that composer's entry. Note is also made of how many tracks by each given composer belonging to the given album are present in the music collection, how many total tracks the album is marked as containing (cumulative for mult-disc sets, if annotated), what percentage of the total tracks are by the given composer (marked in bold if not equal to 100%) and whether each album is marked as a compilation.
• List incomplete albums: Whether or not the Incomplete albums report is generated. This report consists of an alphabetical list of all albums that are missing tracks, followed by a list of all albums with an unknown number of tracks. In the case of multi-album sets, the total number of tracks includes all tracks from all component albums, as long as they all have the same name. Albums that include less than the percentage specified in the % Tracks Present (1-100) text box of the total number of tracks that the album is marked as containing are listed in bold.
• List albums with duplicate or unknown track numbers: Whether or not the Albums with duplicate or unknown track numbers report is generated. This report consists of an alphabetical list of all albums with the same name that contain more than one recording with the same track number or that contain one or more recordings that do not have a track number specified. Note that some albums may be inappropriately listed here in cases where two different albums have exactly the same name.
• List albums with unspecified year: Whether or not the Albums with unspecified year report is generated. This report lists all albums that have one or more recordings missing year metadata.
• Report on compilation albums: Whether or not the Report on compilation albums is generated. This report alphabetically lists all unique albums that contain at least one recording marked as a compilation or with an unknown compilation status. Albums where some recordings are marked as compilations and others are not are written in bold, since all tracks in an album should have the same compilation marking. An album should be marked as a compilation if and only if not all of its component recordings have the same value for their artist fields. A separate list is therefore also included of all albums that are incorrectly not marked as compilations but contain multiple artists (note that some entirely different albums sometimes have an identical name, which will result in incorrect inclusion here). Another list of albums that are incorrectly marked as compilations but contain only one artist is also generated (some of these may in fact be compilation albums that are missing all tracks by all but one artists).

5) Reports Describing Detailed Processing Results

The reports in this section describe the details of the operations that were used to detect metadata errors. These reports are generally only of interest to those technically inclined, as the actual errors detected are output in the reports found in the Summaries of Probable Errors in Metadata section.

Processing for titles, artists, composers, genres or albums are each only performed and reported if each corresponding Probable Error Report is set to be generated.

The specific options in this section are as follows:

• Report fields differing only in case: Whether or not the Fields differing only in case report is generated. Entries listed in this report indicate field values where otherwise identical (before any other find/replace, reordered word subset or edit distance processing has been performed) title, artist, composer, genre and album fields differ in case. One list is produced for each of these fields. It will likely be appropriate to correct metadata so that all fields that are otherwise identical also have the same case. Only one sample value is listed for each cluster of results, and it reflects all selected processing performed up until the time that this report was generated. For example, field values of "Reggae" and "reggae" will result in a singe value of "reggae" reported here.
• Report detailed replacements made: Whether or not the Detailed replacements made report is generated. Entries in this report indicate all changes that were made during all find/replace operations. These operations were performed after lowercase conversion (if requested) and before edit distance operations. The reported changes are cumulative. These find/replace operations were also performed before any requested reordered word subset operations, except for space removals, which were performed after.
• Report newly identical fields after find/replace: Whether or not the Newly identical fields after find and replace report is generated. Entries listed in this report indicate all title, artist, composer, album and genre fields that were made newly identical after the find/replace operations were performed. Any such occurrences are probably due to errors in the metadata and should be corrected. The Detailed replacements made report can be consulted for more information. These operations were performed after lowercase conversion (if requested) and before edit distance operations. These find/replace operations were also performed before any requested reordered word subset operations, except for space removals, which were performed after. Only one sample value is listed for each cluster of results, and it reflects all selected processing performed up until the time that this report was generated. For example, field values of "Dr. John" and "Dr John" would result in a singe value of "dr john" reported here if the appropriate processing preferences were set.
• Report fields whose words are scrambled or subsets: Whether or not the Fields with scrambled word orderings and Fields whose words are subsets of another reports are generated. The Fields with scrambled word orderings report lists all title, artist, composer, album and genre fields that contain identical words, but in any order, and where sometimes the words in one field are a subset of the words in the other. For example, occurrences of both "Duke Ellington" and "Ellington Duke" might be reported here. In order to be listed here, two field values must have at least the percentage of identical words specified next to the Check word ordering option (the words may occur in any order). The percentage is calculated based on the field with the higher number of words, and is calculated after all selected find/replace operations have been performed (except the removal of spaces), but before any edit distance calculations have been performed. Only one sample value is listed for each cluster of results, and it reflects all selected processing performed up until the time that this report was generated. For example, field values of "Jackson, Mahalia" and "Mahalia Jackson" will result in a singe value of "mahalia jackson" reported here. The Fields whose words are subsets of another report lists all title, artist, composer, album and genre fields that contain similar words, usually where the words in one are a subset of the words in the other, and where the words may be in any order. For example, occurrences of both "The Royal Philharmonic Orchestra" and "The Royal Philharmonic" might be reported here. In order to be listed here, the two fields must have at least the percentage of the same words that is specified next to the Check word subsets preference on the left side of the Options Panel (the words may occur in any order). The percentage is calculated based on the field with the lower number of words, and is calculated after all selected find/replace operations have been performed (except the removal of spaces), but before any edit distance calculations have been performed. Note that items that might have otherwise been listed here may be included in the Fields with scrambled word orderings report instead if this report is also set to be generated. Only one sample value is listed for each cluster of results, and it reflects all selected processing performed up until the time that this report was generated. For example, field values of "Duke Ellington and His Orchestra" and "Duke Ellington" will result in a singe value of "duke ellington" reported here. Processing for titles, artists, composers, albums and genres are each only performed and reported if each corresponding Probable Errors report is set to be generated.
• Report edit distance values: Whether or not the Edit distances report is generated. This report lists the edit distances between all pairs of values in each of the title, artist, composer, album and genre fields. One table is generated for each type of field. Each cell in a table contains up to three values, depending on the preferences selected by the user. These are (in the following order): absolute edit distance, proportional edit distance and subset edit distance. Absolute edit distance is the basic Levenshtein distance. Proportional edit distance is the absolute edit distance divided by the length of the longer string, expressed as a percentage. Subset edit distance is the absolute edit distance minus the difference in lengths of the strings, all divided by the length of the shorter string and expressed as a percentage. Distances in bold on the table fall below the thresholds set by the user in the preferences. Entries consisting of a hyphen indicate one or both of the field values are unknown, so the edit distance cannot be calculated. All edit distance calculations are performed on fields after any selected find/replace and/or reordered word subset operations have been performed and any resulting merging of fields has occurred. Processing for titles, artists, composers, albums and genres are each only performed and reported if each corresponding Probable Errors report is set to be generated.
• List filtered candidate duplicate recordings: Whether or not the Filtered candidate duplicate recordings report is generated. This report lists all recordings that were found to have similar titles, but were rejected as being duplicates of other recordings because of the selected filters on the left side of the Options Panel (similar duration, identical artist, identical composer, one or more identical genres and/or identical album). The similarity of titles is assessed after all selected find/replace, reordered word subset and/or edit distance operations are performed, but no such operations are applied to any other fields for the purposes of this report. Each cluster of similar titles where one or more candidate duplicates were filtered out is listed, and each such listing includes either one or two tables. The first table, which is always present, lists recordings that were not found to be duplicates of any other recordings in the cluster. The second table lists all recordings in the cluster, if any, that were in fact found to be duplicates of one or more other recordings in the cluster.

6) Summaries of Probable Errors in Metadata

The reports referred to in this section indicate probable metadata errors that have been found. These can then be used to correct the metadata using software such as Apple iTunes. jMusicMetaManager errs on the side of reporting false errors rather than missing true errors, so some clusters may not in fact be true errors.

The reports listed here indicate two main types of problems: redundant duplicate recordings and metadata fields that should likely be the same but are not. The reports in this section report clusters of recordings that likely should be corrected. In some cases, overly large clusters are collected. These large clusters can either be ignored or the analysis parameters can be adjusted on the left side of the Options Panel so that thresholds are less permissive.

The specific options in this section are as follows:

• Report probable duplicates of the same recording: Whether or not the Probable duplicates of the same recording report is generated. This report lists recordings that are likely to be redundant duplicates of the same recording. First, all recordings whose title fields are judged to be similar after any selected find/replace, reordered word subset and edit distance processing are grouped together into clusters. Then the filters selected in the Duplicate Recording Detection Settings section of the left side of the Options Panel are applied to each such cluster. Specifically, recordings are filtered out of each cluster unless they have similar durations, have the same value for their artist fields, have the same value for their composer fields, share at least one genre and/or have different values for their album fields. Note that no processing is performed on artist, composer, genre or album fields as far as the generation of this report itself is concerned. Each cluster may be broken into further clusters during this processing, if appropriate. Each cluster with more than one surviving recording is then reported. It may be appropriate to delete some of these multiples for the sake of removing redundancy in the music collection. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters on the left side of the Options Panel.
• Report probable errors in title field: Whether or not the Probable errors in title field report is generated. This report lists probable errors in the title field of the input music database's metadata. Each cluster listed indicates a set of several differing yet similar values for the title field that are found in the music database. The values found in each cluster should likely, in reality, be the same. These possible errors should be reviewed and, if appropriate, the metadata should be corrected accordingly. Some clusters may have one or more entries that should not in fact be the same as the others. This is a result of processing that generally errs on the side of reporting false errors rather than missing true errors. Some of the possible transformations that could make them the same are also listed. Only a subset of the necessary find/replace operations may be listed. Additional metadata for each recording in each cluster is also listed. Probable errors reported here were all discovered as a result of the find/replace, reordered word subset and/or edit distance processing that was selected and performed. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters on the left side of the Options Panel.
• Report probable errors in artist field: Whether or not the Probable errors in artist field report is generated. This report lists probable errors in the artist field of the input music database's metadata. Each table listed indicates a cluster of several differing yet similar values for the artist field that are found in the music database. The values found in each cluster should likely, in reality, be the same. These possible errors should be reviewed and, if appropriate, the metadata should be corrected accordingly. Some clusters may have one or more entries that should not in fact be the same as the others. This is a result of processing that generally errs on the side of reporting false errors rather than missing true errors. Some of the possible transformations that could make them the same are also listed. Only a subset of the necessary find/replace operations may be listed. Probable errors reported here were all discovered as a result of the find/replace, reordered word subset and/or edit distance processing that was selected and performed. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters on the left side of the Options Panel.
• Report probable errors in composer field: Whether or not the Probable errors in composer field report is generated. This report lists probable errors in the composer field of the input music database's metadata. Each table listed indicates a cluster of several differing yet similar values for the composer field that are found in the music database. The values found in each cluster should likely, in reality, be the same. These possible errors should be reviewed and, if appropriate, the metadata should be corrected accordingly. Some clusters may have one or more entries that should not in fact be the same as the others. This is a result of processing that generally errs on the side of reporting false errors rather than missing true errors. Some of the possible transformations that could make them the same are also listed. Only a subset of the necessary find/replace operations may be listed. Probable errors reported here were all discovered as a result of the find/replace, reordered word subset and/or edit distance processing that was selected and performed. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters on the left side of the Options Panel.
• Report probable errors in album field: Whether or not the Probable errors in album field report is generated. This report lists probable errors in the album field of the input music database's metadata. Each table listed indicates a cluster of several differing yet similar values for the album field that are found in the music database. The values found in each cluster should likely, in reality, be the same. These possible errors should be reviewed and, if appropriate, the metadata should be corrected accordingly. Some clusters may have one or more entries that should not in fact be the same as the others. This is a result of processing that generally errs on the side of reporting false errors rather than missing true errors. Some of the possible transformations that could make them the same are also listed. Only a subset of the necessary find/replace operations may be listed. Probable errors reported here were all discovered as a result of the find/replace, reordered word subset and/or edit distance processing that was selected and performed. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters on the left side of the Options Panel.
• Report probable errors in genre field: Whether or not the Probable errors in genre field report is generated. This report lists probable errors in the genre field of the input music database's metadata. Each table listed indicates a cluster of several differing yet similar values for the genre field that are found in the music database. The values found in each cluster should likely, in reality, be the same. These possible errors should be reviewed and, if appropriate, the metadata should be corrected accordingly. Some clusters may have one or more entries that should not in fact be the same as the others. This is a result of processing that generally errs on the side of reporting false errors rather than missing true errors. Some of the reported clusters may include some genres that may in fact not be part of the cluster. This is due to a minor processing anomaly when there are multiple genres. Some of the possible transformations that could make them the same are also listed. Only a subset of the necessary find/replace operations may be listed. Probable errors reported here were all discovered as a result of the find/replace, reordered word subset and/or edit distance processing that was selected and performed. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters in the Options Panel. Note that some false errors may be reported here, as the system is biased under the default settings towards catching all true errors. Large clusters of false errors sometimes form when processing extremely large collections. Such clusters may be ignored by the user, or he or she may wish to change the extraction parameters on the left side of the Options Panel.

7) Summaries of Probable Errors in Metadata

The reports referred to in this section contain information regarding processing itself, and do not provide any information about the music that was examined.

The specific options in this section are as follows:

• Report options selected: Whether or not the Options selected report is generated. This report indicates the preferences set by the user in the Options Panel. These preferences determine which reports are generated and control the parameters of the processing that affects the contents of the reports generated.
• Report processing times: Whether or not the Processing time log report is generated. This report indicates how long each processing task took.

-top of page-